zfs/lib/libzfs/libzfs_util.c

2146 lines
51 KiB
C
Raw Normal View History

2008-11-21 04:01:55 +08:00
/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* or http://www.opensolaris.org/os/licensing.
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
2008-11-21 04:01:55 +08:00
/*
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
* Copyright (c) 2013, Joyent, Inc. All rights reserved.
* Copyright (c) 2011, 2014 by Delphix. All rights reserved.
* Copyright 2016 Igor Kozhukhov <ikozhukhov@gmail.com>
* Copyright (c) 2017 Datto Inc.
2008-11-21 04:01:55 +08:00
*/
/*
* Internal utility routines for the ZFS library.
*/
#include <errno.h>
#include <fcntl.h>
#include <libintl.h>
#include <stdarg.h>
#include <stdio.h>
#include <stdlib.h>
#include <strings.h>
#include <unistd.h>
#include <ctype.h>
#include <math.h>
#include <sys/stat.h>
2008-11-21 04:01:55 +08:00
#include <sys/mnttab.h>
#include <sys/mntent.h>
#include <sys/types.h>
#include <sys/wait.h>
2008-11-21 04:01:55 +08:00
#include <libzfs.h>
Illumos #2882, #2883, #2900 2882 implement libzfs_core 2883 changing "canmount" property to "on" should not always remount dataset 2900 "zfs snapshot" should be able to create multiple, arbitrary snapshots at once Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Chris Siden <christopher.siden@delphix.com> Reviewed by: Garrett D'Amore <garrett@damore.org> Reviewed by: Bill Pijewski <wdp@joyent.com> Reviewed by: Dan Kruchinin <dan.kruchinin@gmail.com> Approved by: Eric Schrock <Eric.Schrock@delphix.com> References: https://www.illumos.org/issues/2882 https://www.illumos.org/issues/2883 https://www.illumos.org/issues/2900 illumos/illumos-gate@4445fffbbb1ea25fd0e9ea68b9380dd7a6709025 Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1293 Porting notes: WARNING: This patch changes the user/kernel ABI. That means that the zfs/zpool utilities built from master are NOT compatible with the 0.6.2 kernel modules. Ensure you load the matching kernel modules from master after updating the utilities. Otherwise the zfs/zpool commands will be unable to interact with your pool and you will see errors similar to the following: $ zpool list failed to read pool configuration: bad address no pools available $ zfs list no datasets available Add zvol minor device creation to the new zfs_snapshot_nvl function. Remove the logging of the "release" operation in dsl_dataset_user_release_sync(). The logging caused a null dereference because ds->ds_dir is zeroed in dsl_dataset_destroy_sync() and the logging functions try to get the ds name via the dsl_dataset_name() function. I've got no idea why this particular code would have worked in Illumos. This code has subsequently been completely reworked in Illumos commit 3b2aab1 (3464 zfs synctask code needs restructuring). Squash some "may be used uninitialized" warning/erorrs. Fix some printf format warnings for %lld and %llu. Apply a few spa_writeable() changes that were made to Illumos in illumos/illumos-gate.git@cd1c8b8 as part of the 3112, 3113, 3114 and 3115 fixes. Add a missing call to fnvlist_free(nvl) in log_internal() that was added in Illumos to fix issue 3085 but couldn't be ported to ZoL at the time (zfsonlinux/zfs@9e11c73) because it depended on future work.
2013-08-28 19:45:09 +08:00
#include <libzfs_core.h>
2008-11-21 04:01:55 +08:00
#include "libzfs_impl.h"
#include "zfs_prop.h"
#include "zfeature_common.h"
#include <zfs_fletcher.h>
2008-11-21 04:01:55 +08:00
int
libzfs_errno(libzfs_handle_t *hdl)
{
return (hdl->libzfs_error);
}
const char *
libzfs_error_init(int error)
{
switch (error) {
case ENXIO:
return (dgettext(TEXT_DOMAIN, "The ZFS modules are not "
"loaded.\nTry running '/sbin/modprobe zfs' as root "
"to load them.\n"));
case ENOENT:
return (dgettext(TEXT_DOMAIN, "/dev/zfs and /proc/self/mounts "
"are required.\nTry running 'udevadm trigger' and 'mount "
"-t proc proc /proc' as root.\n"));
case ENOEXEC:
return (dgettext(TEXT_DOMAIN, "The ZFS modules cannot be "
"auto-loaded.\nTry running '/sbin/modprobe zfs' as "
"root to manually load them.\n"));
case EACCES:
return (dgettext(TEXT_DOMAIN, "Permission denied the "
"ZFS utilities must be run as root.\n"));
default:
return (dgettext(TEXT_DOMAIN, "Failed to initialize the "
"libzfs library.\n"));
}
}
2008-11-21 04:01:55 +08:00
const char *
libzfs_error_action(libzfs_handle_t *hdl)
{
return (hdl->libzfs_action);
}
const char *
libzfs_error_description(libzfs_handle_t *hdl)
{
if (hdl->libzfs_desc[0] != '\0')
return (hdl->libzfs_desc);
switch (hdl->libzfs_error) {
case EZFS_NOMEM:
return (dgettext(TEXT_DOMAIN, "out of memory"));
case EZFS_BADPROP:
return (dgettext(TEXT_DOMAIN, "invalid property value"));
case EZFS_PROPREADONLY:
return (dgettext(TEXT_DOMAIN, "read-only property"));
2008-11-21 04:01:55 +08:00
case EZFS_PROPTYPE:
return (dgettext(TEXT_DOMAIN, "property doesn't apply to "
"datasets of this type"));
case EZFS_PROPNONINHERIT:
return (dgettext(TEXT_DOMAIN, "property cannot be inherited"));
case EZFS_PROPSPACE:
return (dgettext(TEXT_DOMAIN, "invalid quota or reservation"));
case EZFS_BADTYPE:
return (dgettext(TEXT_DOMAIN, "operation not applicable to "
"datasets of this type"));
case EZFS_BUSY:
return (dgettext(TEXT_DOMAIN, "pool or dataset is busy"));
case EZFS_EXISTS:
return (dgettext(TEXT_DOMAIN, "pool or dataset exists"));
case EZFS_NOENT:
return (dgettext(TEXT_DOMAIN, "no such pool or dataset"));
case EZFS_BADSTREAM:
return (dgettext(TEXT_DOMAIN, "invalid backup stream"));
case EZFS_DSREADONLY:
return (dgettext(TEXT_DOMAIN, "dataset is read-only"));
2008-11-21 04:01:55 +08:00
case EZFS_VOLTOOBIG:
return (dgettext(TEXT_DOMAIN, "volume size exceeds limit for "
"this system"));
case EZFS_INVALIDNAME:
return (dgettext(TEXT_DOMAIN, "invalid name"));
case EZFS_BADRESTORE:
return (dgettext(TEXT_DOMAIN, "unable to restore to "
"destination"));
case EZFS_BADBACKUP:
return (dgettext(TEXT_DOMAIN, "backup failed"));
case EZFS_BADTARGET:
return (dgettext(TEXT_DOMAIN, "invalid target vdev"));
case EZFS_NODEVICE:
return (dgettext(TEXT_DOMAIN, "no such device in pool"));
case EZFS_BADDEV:
return (dgettext(TEXT_DOMAIN, "invalid device"));
case EZFS_NOREPLICAS:
return (dgettext(TEXT_DOMAIN, "no valid replicas"));
case EZFS_RESILVERING:
return (dgettext(TEXT_DOMAIN, "currently resilvering"));
case EZFS_BADVERSION:
return (dgettext(TEXT_DOMAIN, "unsupported version or "
"feature"));
2008-11-21 04:01:55 +08:00
case EZFS_POOLUNAVAIL:
return (dgettext(TEXT_DOMAIN, "pool is unavailable"));
case EZFS_DEVOVERFLOW:
return (dgettext(TEXT_DOMAIN, "too many devices in one vdev"));
case EZFS_BADPATH:
return (dgettext(TEXT_DOMAIN, "must be an absolute path"));
case EZFS_CROSSTARGET:
return (dgettext(TEXT_DOMAIN, "operation crosses datasets or "
"pools"));
case EZFS_ZONED:
return (dgettext(TEXT_DOMAIN, "dataset in use by local zone"));
case EZFS_MOUNTFAILED:
return (dgettext(TEXT_DOMAIN, "mount failed"));
case EZFS_UMOUNTFAILED:
return (dgettext(TEXT_DOMAIN, "umount failed"));
case EZFS_UNSHARENFSFAILED:
return (dgettext(TEXT_DOMAIN, "unshare(1M) failed"));
case EZFS_SHARENFSFAILED:
return (dgettext(TEXT_DOMAIN, "share(1M) failed"));
case EZFS_UNSHARESMBFAILED:
return (dgettext(TEXT_DOMAIN, "smb remove share failed"));
case EZFS_SHARESMBFAILED:
return (dgettext(TEXT_DOMAIN, "smb add share failed"));
case EZFS_PERM:
return (dgettext(TEXT_DOMAIN, "permission denied"));
case EZFS_NOSPC:
return (dgettext(TEXT_DOMAIN, "out of space"));
case EZFS_FAULT:
return (dgettext(TEXT_DOMAIN, "bad address"));
2008-11-21 04:01:55 +08:00
case EZFS_IO:
return (dgettext(TEXT_DOMAIN, "I/O error"));
case EZFS_INTR:
return (dgettext(TEXT_DOMAIN, "signal received"));
case EZFS_ISSPARE:
return (dgettext(TEXT_DOMAIN, "device is reserved as a hot "
"spare"));
case EZFS_INVALCONFIG:
return (dgettext(TEXT_DOMAIN, "invalid vdev configuration"));
case EZFS_RECURSIVE:
return (dgettext(TEXT_DOMAIN, "recursive dataset dependency"));
case EZFS_NOHISTORY:
return (dgettext(TEXT_DOMAIN, "no history available"));
case EZFS_POOLPROPS:
return (dgettext(TEXT_DOMAIN, "failed to retrieve "
"pool properties"));
case EZFS_POOL_NOTSUP:
return (dgettext(TEXT_DOMAIN, "operation not supported "
"on this type of pool"));
case EZFS_POOL_INVALARG:
return (dgettext(TEXT_DOMAIN, "invalid argument for "
"this pool operation"));
case EZFS_NAMETOOLONG:
return (dgettext(TEXT_DOMAIN, "dataset name is too long"));
case EZFS_OPENFAILED:
return (dgettext(TEXT_DOMAIN, "open failed"));
case EZFS_NOCAP:
return (dgettext(TEXT_DOMAIN,
"disk capacity information could not be retrieved"));
case EZFS_LABELFAILED:
return (dgettext(TEXT_DOMAIN, "write of label failed"));
case EZFS_BADWHO:
return (dgettext(TEXT_DOMAIN, "invalid user/group"));
case EZFS_BADPERM:
return (dgettext(TEXT_DOMAIN, "invalid permission"));
case EZFS_BADPERMSET:
return (dgettext(TEXT_DOMAIN, "invalid permission set name"));
case EZFS_NODELEGATION:
return (dgettext(TEXT_DOMAIN, "delegated administration is "
"disabled on pool"));
case EZFS_BADCACHE:
return (dgettext(TEXT_DOMAIN, "invalid or missing cache file"));
case EZFS_ISL2CACHE:
return (dgettext(TEXT_DOMAIN, "device is in use as a cache"));
case EZFS_VDEVNOTSUP:
return (dgettext(TEXT_DOMAIN, "vdev specification is not "
"supported"));
case EZFS_NOTSUP:
return (dgettext(TEXT_DOMAIN, "operation not supported "
"on this dataset"));
case EZFS_ACTIVE_SPARE:
return (dgettext(TEXT_DOMAIN, "pool has active shared spare "
"device"));
2009-07-03 06:44:48 +08:00
case EZFS_UNPLAYED_LOGS:
return (dgettext(TEXT_DOMAIN, "log device has unplayed intent "
"logs"));
2009-08-19 02:43:27 +08:00
case EZFS_REFTAG_RELE:
return (dgettext(TEXT_DOMAIN, "no such tag on this dataset"));
case EZFS_REFTAG_HOLD:
return (dgettext(TEXT_DOMAIN, "tag already exists on this "
"dataset"));
case EZFS_TAGTOOLONG:
return (dgettext(TEXT_DOMAIN, "tag too long"));
case EZFS_PIPEFAILED:
return (dgettext(TEXT_DOMAIN, "pipe create failed"));
case EZFS_THREADCREATEFAILED:
return (dgettext(TEXT_DOMAIN, "thread create failed"));
case EZFS_POSTSPLIT_ONLINE:
return (dgettext(TEXT_DOMAIN, "disk was split from this pool "
"into a new one"));
case EZFS_SCRUB_PAUSED:
return (dgettext(TEXT_DOMAIN, "scrub is paused; "
"use 'zpool scrub' to resume"));
case EZFS_SCRUBBING:
return (dgettext(TEXT_DOMAIN, "currently scrubbing; "
"use 'zpool scrub -s' to cancel current scrub"));
case EZFS_NO_SCRUB:
return (dgettext(TEXT_DOMAIN, "there is no active scrub"));
case EZFS_DIFF:
return (dgettext(TEXT_DOMAIN, "unable to generate diffs"));
case EZFS_DIFFDATA:
return (dgettext(TEXT_DOMAIN, "invalid diff data"));
case EZFS_POOLREADONLY:
return (dgettext(TEXT_DOMAIN, "pool is read-only"));
Multi-modifier protection (MMP) Add multihost=on|off pool property to control MMP. When enabled a new thread writes uberblocks to the last slot in each label, at a set frequency, to indicate to other hosts the pool is actively imported. These uberblocks are the last synced uberblock with an updated timestamp. Property defaults to off. During tryimport, find the "best" uberblock (newest txg and timestamp) repeatedly, checking for change in the found uberblock. Include the results of the activity test in the config returned by tryimport. These results are reported to user in "zpool import". Allow the user to control the period between MMP writes, and the duration of the activity test on import, via a new module parameter zfs_multihost_interval. The period is specified in milliseconds. The activity test duration is calculated from this value, and from the mmp_delay in the "best" uberblock found initially. Add a kstat interface to export statistics about Multiple Modifier Protection (MMP) updates. Include the last synced txg number, the timestamp, the delay since the last MMP update, the VDEV GUID, the VDEV label that received the last MMP update, and the VDEV path. Abbreviated output below. $ cat /proc/spl/kstat/zfs/mypool/multihost 31 0 0x01 10 880 105092382393521 105144180101111 txg timestamp mmp_delay vdev_guid vdev_label vdev_path 20468 261337 250274925 68396651780 3 /dev/sda 20468 261339 252023374 6267402363293 1 /dev/sdc 20468 261340 252000858 6698080955233 1 /dev/sdx 20468 261341 251980635 783892869810 2 /dev/sdy 20468 261342 253385953 8923255792467 3 /dev/sdd 20468 261344 253336622 042125143176 0 /dev/sdab 20468 261345 253310522 1200778101278 2 /dev/sde 20468 261346 253286429 0950576198362 2 /dev/sdt 20468 261347 253261545 96209817917 3 /dev/sds 20468 261349 253238188 8555725937673 3 /dev/sdb Add a new tunable zfs_multihost_history to specify the number of MMP updates to store history for. By default it is set to zero meaning that no MMP statistics are stored. When using ztest to generate activity, for automated tests of the MMP function, some test functions interfere with the test. For example, the pool is exported to run zdb and then imported again. Add a new ztest function, "-M", to alter ztest behavior to prevent this. Add new tests to verify the new functionality. Tests provided by Giuseppe Di Natale. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #745 Closes #6279
2017-07-08 11:20:35 +08:00
case EZFS_ACTIVE_POOL:
return (dgettext(TEXT_DOMAIN, "pool is imported on a "
"different host"));
Native Encryption for ZFS on Linux This change incorporates three major pieces: The first change is a keystore that manages wrapping and encryption keys for encrypted datasets. These commands mostly involve manipulating the new DSL Crypto Key ZAP Objects that live in the MOS. Each encrypted dataset has its own DSL Crypto Key that is protected with a user's key. This level of indirection allows users to change their keys without re-encrypting their entire datasets. The change implements the new subcommands "zfs load-key", "zfs unload-key" and "zfs change-key" which allow the user to manage their encryption keys and settings. In addition, several new flags and properties have been added to allow dataset creation and to make mounting and unmounting more convenient. The second piece of this patch provides the ability to encrypt, decyrpt, and authenticate protected datasets. Each object set maintains a Merkel tree of Message Authentication Codes that protect the lower layers, similarly to how checksums are maintained. This part impacts the zio layer, which handles the actual encryption and generation of MACs, as well as the ARC and DMU, which need to be able to handle encrypted buffers and protected data. The last addition is the ability to do raw, encrypted sends and receives. The idea here is to send raw encrypted and compressed data and receive it exactly as is on a backup system. This means that the dataset on the receiving system is protected using the same user key that is in use on the sending side. By doing so, datasets can be efficiently backed up to an untrusted system without fear of data being compromised. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Reviewed-by: Jorgen Lundman <lundman@lundman.net> Signed-off-by: Tom Caputi <tcaputi@datto.com> Closes #494 Closes #5769
2017-08-15 01:36:48 +08:00
case EZFS_CRYPTOFAILED:
return (dgettext(TEXT_DOMAIN, "encryption failure"));
2008-11-21 04:01:55 +08:00
case EZFS_UNKNOWN:
return (dgettext(TEXT_DOMAIN, "unknown error"));
default:
assert(hdl->libzfs_error == 0);
return (dgettext(TEXT_DOMAIN, "no error"));
}
}
/*PRINTFLIKE2*/
void
zfs_error_aux(libzfs_handle_t *hdl, const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
(void) vsnprintf(hdl->libzfs_desc, sizeof (hdl->libzfs_desc),
fmt, ap);
hdl->libzfs_desc_active = 1;
va_end(ap);
}
static void
zfs_verror(libzfs_handle_t *hdl, int error, const char *fmt, va_list ap)
{
(void) vsnprintf(hdl->libzfs_action, sizeof (hdl->libzfs_action),
fmt, ap);
hdl->libzfs_error = error;
if (hdl->libzfs_desc_active)
hdl->libzfs_desc_active = 0;
else
hdl->libzfs_desc[0] = '\0';
if (hdl->libzfs_printerr) {
if (error == EZFS_UNKNOWN) {
(void) fprintf(stderr, dgettext(TEXT_DOMAIN, "internal "
"error: %s\n"), libzfs_error_description(hdl));
abort();
}
(void) fprintf(stderr, "%s: %s\n", hdl->libzfs_action,
libzfs_error_description(hdl));
if (error == EZFS_NOMEM)
exit(1);
}
}
int
zfs_error(libzfs_handle_t *hdl, int error, const char *msg)
{
return (zfs_error_fmt(hdl, error, "%s", msg));
}
/*PRINTFLIKE3*/
int
zfs_error_fmt(libzfs_handle_t *hdl, int error, const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
zfs_verror(hdl, error, fmt, ap);
va_end(ap);
return (-1);
}
static int
zfs_common_error(libzfs_handle_t *hdl, int error, const char *fmt,
va_list ap)
{
switch (error) {
case EPERM:
case EACCES:
zfs_verror(hdl, EZFS_PERM, fmt, ap);
return (-1);
case ECANCELED:
zfs_verror(hdl, EZFS_NODELEGATION, fmt, ap);
return (-1);
case EIO:
zfs_verror(hdl, EZFS_IO, fmt, ap);
return (-1);
case EFAULT:
zfs_verror(hdl, EZFS_FAULT, fmt, ap);
return (-1);
2008-11-21 04:01:55 +08:00
case EINTR:
zfs_verror(hdl, EZFS_INTR, fmt, ap);
return (-1);
}
return (0);
}
int
zfs_standard_error(libzfs_handle_t *hdl, int error, const char *msg)
{
return (zfs_standard_error_fmt(hdl, error, "%s", msg));
}
/*PRINTFLIKE3*/
int
zfs_standard_error_fmt(libzfs_handle_t *hdl, int error, const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
if (zfs_common_error(hdl, error, fmt, ap) != 0) {
va_end(ap);
return (-1);
}
switch (error) {
case ENXIO:
case ENODEV:
case EPIPE:
2008-11-21 04:01:55 +08:00
zfs_verror(hdl, EZFS_IO, fmt, ap);
break;
case ENOENT:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"dataset does not exist"));
zfs_verror(hdl, EZFS_NOENT, fmt, ap);
break;
case ENOSPC:
case EDQUOT:
zfs_verror(hdl, EZFS_NOSPC, fmt, ap);
break;
2008-11-21 04:01:55 +08:00
case EEXIST:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"dataset already exists"));
zfs_verror(hdl, EZFS_EXISTS, fmt, ap);
break;
case EBUSY:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"dataset is busy"));
zfs_verror(hdl, EZFS_BUSY, fmt, ap);
break;
case EROFS:
zfs_verror(hdl, EZFS_POOLREADONLY, fmt, ap);
2008-11-21 04:01:55 +08:00
break;
case ENAMETOOLONG:
zfs_verror(hdl, EZFS_NAMETOOLONG, fmt, ap);
break;
case ENOTSUP:
zfs_verror(hdl, EZFS_BADVERSION, fmt, ap);
break;
2009-07-03 06:44:48 +08:00
case EAGAIN:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"pool I/O is currently suspended"));
zfs_verror(hdl, EZFS_POOLUNAVAIL, fmt, ap);
break;
Multi-modifier protection (MMP) Add multihost=on|off pool property to control MMP. When enabled a new thread writes uberblocks to the last slot in each label, at a set frequency, to indicate to other hosts the pool is actively imported. These uberblocks are the last synced uberblock with an updated timestamp. Property defaults to off. During tryimport, find the "best" uberblock (newest txg and timestamp) repeatedly, checking for change in the found uberblock. Include the results of the activity test in the config returned by tryimport. These results are reported to user in "zpool import". Allow the user to control the period between MMP writes, and the duration of the activity test on import, via a new module parameter zfs_multihost_interval. The period is specified in milliseconds. The activity test duration is calculated from this value, and from the mmp_delay in the "best" uberblock found initially. Add a kstat interface to export statistics about Multiple Modifier Protection (MMP) updates. Include the last synced txg number, the timestamp, the delay since the last MMP update, the VDEV GUID, the VDEV label that received the last MMP update, and the VDEV path. Abbreviated output below. $ cat /proc/spl/kstat/zfs/mypool/multihost 31 0 0x01 10 880 105092382393521 105144180101111 txg timestamp mmp_delay vdev_guid vdev_label vdev_path 20468 261337 250274925 68396651780 3 /dev/sda 20468 261339 252023374 6267402363293 1 /dev/sdc 20468 261340 252000858 6698080955233 1 /dev/sdx 20468 261341 251980635 783892869810 2 /dev/sdy 20468 261342 253385953 8923255792467 3 /dev/sdd 20468 261344 253336622 042125143176 0 /dev/sdab 20468 261345 253310522 1200778101278 2 /dev/sde 20468 261346 253286429 0950576198362 2 /dev/sdt 20468 261347 253261545 96209817917 3 /dev/sds 20468 261349 253238188 8555725937673 3 /dev/sdb Add a new tunable zfs_multihost_history to specify the number of MMP updates to store history for. By default it is set to zero meaning that no MMP statistics are stored. When using ztest to generate activity, for automated tests of the MMP function, some test functions interfere with the test. For example, the pool is exported to run zdb and then imported again. Add a new ztest function, "-M", to alter ztest behavior to prevent this. Add new tests to verify the new functionality. Tests provided by Giuseppe Di Natale. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #745 Closes #6279
2017-07-08 11:20:35 +08:00
case EREMOTEIO:
zfs_verror(hdl, EZFS_ACTIVE_POOL, fmt, ap);
break;
2008-11-21 04:01:55 +08:00
default:
zfs_error_aux(hdl, strerror(error));
2008-11-21 04:01:55 +08:00
zfs_verror(hdl, EZFS_UNKNOWN, fmt, ap);
break;
}
va_end(ap);
return (-1);
}
int
zpool_standard_error(libzfs_handle_t *hdl, int error, const char *msg)
{
return (zpool_standard_error_fmt(hdl, error, "%s", msg));
}
/*PRINTFLIKE3*/
int
zpool_standard_error_fmt(libzfs_handle_t *hdl, int error, const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
if (zfs_common_error(hdl, error, fmt, ap) != 0) {
va_end(ap);
return (-1);
}
switch (error) {
case ENODEV:
zfs_verror(hdl, EZFS_NODEVICE, fmt, ap);
break;
case ENOENT:
zfs_error_aux(hdl,
dgettext(TEXT_DOMAIN, "no such pool or dataset"));
zfs_verror(hdl, EZFS_NOENT, fmt, ap);
break;
case EEXIST:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"pool already exists"));
zfs_verror(hdl, EZFS_EXISTS, fmt, ap);
break;
case EBUSY:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "pool is busy"));
zfs_verror(hdl, EZFS_BUSY, fmt, ap);
2008-11-21 04:01:55 +08:00
break;
case ENXIO:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"one or more devices is currently unavailable"));
zfs_verror(hdl, EZFS_BADDEV, fmt, ap);
break;
case ENAMETOOLONG:
zfs_verror(hdl, EZFS_DEVOVERFLOW, fmt, ap);
break;
case ENOTSUP:
zfs_verror(hdl, EZFS_POOL_NOTSUP, fmt, ap);
break;
case EINVAL:
zfs_verror(hdl, EZFS_POOL_INVALARG, fmt, ap);
break;
case ENOSPC:
case EDQUOT:
zfs_verror(hdl, EZFS_NOSPC, fmt, ap);
return (-1);
2009-07-03 06:44:48 +08:00
case EAGAIN:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"pool I/O is currently suspended"));
zfs_verror(hdl, EZFS_POOLUNAVAIL, fmt, ap);
break;
2008-11-21 04:01:55 +08:00
case EROFS:
zfs_verror(hdl, EZFS_POOLREADONLY, fmt, ap);
break;
case EDOM:
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"block size out of range or does not match"));
zfs_verror(hdl, EZFS_BADPROP, fmt, ap);
break;
Multi-modifier protection (MMP) Add multihost=on|off pool property to control MMP. When enabled a new thread writes uberblocks to the last slot in each label, at a set frequency, to indicate to other hosts the pool is actively imported. These uberblocks are the last synced uberblock with an updated timestamp. Property defaults to off. During tryimport, find the "best" uberblock (newest txg and timestamp) repeatedly, checking for change in the found uberblock. Include the results of the activity test in the config returned by tryimport. These results are reported to user in "zpool import". Allow the user to control the period between MMP writes, and the duration of the activity test on import, via a new module parameter zfs_multihost_interval. The period is specified in milliseconds. The activity test duration is calculated from this value, and from the mmp_delay in the "best" uberblock found initially. Add a kstat interface to export statistics about Multiple Modifier Protection (MMP) updates. Include the last synced txg number, the timestamp, the delay since the last MMP update, the VDEV GUID, the VDEV label that received the last MMP update, and the VDEV path. Abbreviated output below. $ cat /proc/spl/kstat/zfs/mypool/multihost 31 0 0x01 10 880 105092382393521 105144180101111 txg timestamp mmp_delay vdev_guid vdev_label vdev_path 20468 261337 250274925 68396651780 3 /dev/sda 20468 261339 252023374 6267402363293 1 /dev/sdc 20468 261340 252000858 6698080955233 1 /dev/sdx 20468 261341 251980635 783892869810 2 /dev/sdy 20468 261342 253385953 8923255792467 3 /dev/sdd 20468 261344 253336622 042125143176 0 /dev/sdab 20468 261345 253310522 1200778101278 2 /dev/sde 20468 261346 253286429 0950576198362 2 /dev/sdt 20468 261347 253261545 96209817917 3 /dev/sds 20468 261349 253238188 8555725937673 3 /dev/sdb Add a new tunable zfs_multihost_history to specify the number of MMP updates to store history for. By default it is set to zero meaning that no MMP statistics are stored. When using ztest to generate activity, for automated tests of the MMP function, some test functions interfere with the test. For example, the pool is exported to run zdb and then imported again. Add a new ztest function, "-M", to alter ztest behavior to prevent this. Add new tests to verify the new functionality. Tests provided by Giuseppe Di Natale. Reviewed by: Matthew Ahrens <mahrens@delphix.com> Reviewed-by: Giuseppe Di Natale <dinatale2@llnl.gov> Reviewed-by: Ned Bass <bass6@llnl.gov> Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #745 Closes #6279
2017-07-08 11:20:35 +08:00
case EREMOTEIO:
zfs_verror(hdl, EZFS_ACTIVE_POOL, fmt, ap);
break;
2008-11-21 04:01:55 +08:00
default:
zfs_error_aux(hdl, strerror(error));
zfs_verror(hdl, EZFS_UNKNOWN, fmt, ap);
}
va_end(ap);
return (-1);
}
/*
* Display an out of memory error message and abort the current program.
*/
int
no_memory(libzfs_handle_t *hdl)
{
return (zfs_error(hdl, EZFS_NOMEM, "internal error"));
}
/*
* A safe form of malloc() which will die if the allocation fails.
*/
void *
zfs_alloc(libzfs_handle_t *hdl, size_t size)
{
void *data;
if ((data = calloc(1, size)) == NULL)
(void) no_memory(hdl);
return (data);
}
/*
* A safe form of asprintf() which will die if the allocation fails.
*/
/*PRINTFLIKE2*/
char *
zfs_asprintf(libzfs_handle_t *hdl, const char *fmt, ...)
{
va_list ap;
char *ret;
int err;
va_start(ap, fmt);
err = vasprintf(&ret, fmt, ap);
va_end(ap);
if (err < 0)
(void) no_memory(hdl);
return (ret);
}
2008-11-21 04:01:55 +08:00
/*
* A safe form of realloc(), which also zeroes newly allocated space.
*/
void *
zfs_realloc(libzfs_handle_t *hdl, void *ptr, size_t oldsize, size_t newsize)
{
void *ret;
if ((ret = realloc(ptr, newsize)) == NULL) {
(void) no_memory(hdl);
return (NULL);
}
bzero((char *)ret + oldsize, (newsize - oldsize));
return (ret);
}
/*
* A safe form of strdup() which will die if the allocation fails.
*/
char *
zfs_strdup(libzfs_handle_t *hdl, const char *str)
{
char *ret;
if ((ret = strdup(str)) == NULL)
(void) no_memory(hdl);
return (ret);
}
/*
* Convert a number to an appropriately human-readable output.
*/
void
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
zfs_nicenum_format(uint64_t num, char *buf, size_t buflen,
enum zfs_nicenum_format format)
2008-11-21 04:01:55 +08:00
{
uint64_t n = num;
int index = 0;
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
const char *u;
const char *units[3][7] = {
[ZFS_NICENUM_1024] = {"", "K", "M", "G", "T", "P", "E"},
[ZFS_NICENUM_BYTES] = {"B", "K", "M", "G", "T", "P", "E"},
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
[ZFS_NICENUM_TIME] = {"ns", "us", "ms", "s", "?", "?", "?"}
};
const int units_len[] = {[ZFS_NICENUM_1024] = 6,
[ZFS_NICENUM_BYTES] = 6,
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
[ZFS_NICENUM_TIME] = 4};
const int k_unit[] = { [ZFS_NICENUM_1024] = 1024,
[ZFS_NICENUM_BYTES] = 1024,
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
[ZFS_NICENUM_TIME] = 1000};
double val;
2008-11-21 04:01:55 +08:00
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
if (format == ZFS_NICENUM_RAW) {
snprintf(buf, buflen, "%llu", (u_longlong_t)num);
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
return;
} else if (format == ZFS_NICENUM_RAWTIME && num > 0) {
snprintf(buf, buflen, "%llu", (u_longlong_t)num);
return;
} else if (format == ZFS_NICENUM_RAWTIME && num == 0) {
snprintf(buf, buflen, "%s", "-");
return;
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
}
while (n >= k_unit[format] && index < units_len[format]) {
n /= k_unit[format];
2008-11-21 04:01:55 +08:00
index++;
}
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
u = units[format][index];
2008-11-21 04:01:55 +08:00
/* Don't print zero latencies since they're invalid */
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
if ((format == ZFS_NICENUM_TIME) && (num == 0)) {
(void) snprintf(buf, buflen, "-");
} else if ((index == 0) || ((num %
(uint64_t)powl(k_unit[format], index)) == 0)) {
2008-11-21 04:01:55 +08:00
/*
* If this is an even multiple of the base, always display
* without any decimal precision.
*/
(void) snprintf(buf, buflen, "%llu%s", (u_longlong_t)n, u);
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
2008-11-21 04:01:55 +08:00
} else {
/*
* We want to choose a precision that reflects the best choice
* for fitting in 5 characters. This can get rather tricky when
* we have numbers that are very close to an order of magnitude.
* For example, when displaying 10239 (which is really 9.999K),
* we want only a single place of precision for 10.0K. We could
* develop some complex heuristics for this, but it's much
* easier just to try each combination in turn.
*/
int i;
for (i = 2; i >= 0; i--) {
val = (double)num /
(uint64_t)powl(k_unit[format], index);
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
/*
* Don't print floating point values for time. Note,
* we use floor() instead of round() here, since
* round can result in undesirable results. For
* example, if "num" is in the range of
* 999500-999999, it will print out "1000us". This
* doesn't happen if we use floor().
*/
if (format == ZFS_NICENUM_TIME) {
if (snprintf(buf, buflen, "%d%s",
(unsigned int) floor(val), u) <= 5)
break;
} else {
if (snprintf(buf, buflen, "%.*f%s", i,
val, u) <= 5)
break;
}
2008-11-21 04:01:55 +08:00
}
}
}
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
/*
* Convert a number to an appropriately human-readable output.
*/
void
zfs_nicenum(uint64_t num, char *buf, size_t buflen)
{
zfs_nicenum_format(num, buf, buflen, ZFS_NICENUM_1024);
}
/*
* Convert a time to an appropriately human-readable output.
* @num: Time in nanoseconds
*/
void
zfs_nicetime(uint64_t num, char *buf, size_t buflen)
{
zfs_nicenum_format(num, buf, buflen, ZFS_NICENUM_TIME);
}
/*
* Print out a raw number with correct column spacing
*/
void
zfs_niceraw(uint64_t num, char *buf, size_t buflen)
{
zfs_nicenum_format(num, buf, buflen, ZFS_NICENUM_RAW);
}
/*
* Convert a number of bytes to an appropriately human-readable output.
*/
void
zfs_nicebytes(uint64_t num, char *buf, size_t buflen)
{
zfs_nicenum_format(num, buf, buflen, ZFS_NICENUM_BYTES);
}
Add -lhHpw options to "zpool iostat" for avg latency, histograms, & queues Update the zfs module to collect statistics on average latencies, queue sizes, and keep an internal histogram of all IO latencies. Along with this, update "zpool iostat" with some new options to print out the stats: -l: Include average IO latencies stats: total_wait disk_wait syncq_wait asyncq_wait scrub read write read write read write read write wait ----- ----- ----- ----- ----- ----- ----- ----- ----- - 41ms - 2ms - 46ms - 4ms - - 5ms - 1ms - 1us - 4ms - - 5ms - 1ms - 1us - 4ms - - - - - - - - - - - 49ms - 2ms - 47ms - - - - - - - - - - - - - 2ms - 1ms - - - 1ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- 1ms 1ms 1ms 413us 16us 25us - 5ms - 1ms 1ms 1ms 413us 16us 25us - 5ms - 2ms 1ms 2ms 412us 26us 25us - 5ms - - 1ms - 413us - 25us - 5ms - - 1ms - 460us - 29us - 5ms - 196us 1ms 196us 370us 7us 23us - 5ms - ----- ----- ----- ----- ----- ----- ----- ----- ----- -w: Print out latency histograms: sdb total disk sync_queue async_queue latency read write read write read write read write scrub ------- ------ ------ ------ ------ ------ ------ ------ ------ ------ 1ns 0 0 0 0 0 0 0 0 0 ... 33us 0 0 0 0 0 0 0 0 0 66us 0 0 107 2486 2 788 12 12 0 131us 2 797 359 4499 10 558 184 184 6 262us 22 801 264 1563 10 286 287 287 24 524us 87 575 71 52086 15 1063 136 136 92 1ms 152 1190 5 41292 4 1693 252 252 141 2ms 245 2018 0 50007 0 2322 371 371 220 4ms 189 7455 22 162957 0 3912 6726 6726 199 8ms 108 9461 0 102320 0 5775 2526 2526 86 17ms 23 11287 0 37142 0 8043 1813 1813 19 34ms 0 14725 0 24015 0 11732 3071 3071 0 67ms 0 23597 0 7914 0 18113 5025 5025 0 134ms 0 33798 0 254 0 25755 7326 7326 0 268ms 0 51780 0 12 0 41593 10002 10002 0 537ms 0 77808 0 0 0 64255 13120 13120 0 1s 0 105281 0 0 0 83805 20841 20841 0 2s 0 88248 0 0 0 73772 14006 14006 0 4s 0 47266 0 0 0 29783 17176 17176 0 9s 0 10460 0 0 0 4130 6295 6295 0 17s 0 0 0 0 0 0 0 0 0 34s 0 0 0 0 0 0 0 0 0 69s 0 0 0 0 0 0 0 0 0 137s 0 0 0 0 0 0 0 0 0 ------------------------------------------------------------------------------- -h: Help -H: Scripted mode. Do not display headers, and separate fields by a single tab instead of arbitrary space. -q: Include current number of entries in sync & async read/write queues, and scrub queue: syncq_read syncq_write asyncq_read asyncq_write scrubq_read pend activ pend activ pend activ pend activ pend activ ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 0 0 78 29 0 0 0 0 0 0 0 0 78 29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 - - - - - - - - - - 0 0 0 0 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- 0 0 227 394 0 19 0 0 0 0 0 0 227 394 0 19 0 0 0 0 0 0 108 98 0 19 0 0 0 0 0 0 19 98 0 0 0 0 0 0 0 0 78 98 0 0 0 0 0 0 0 0 19 88 0 0 0 0 0 0 ----- ----- ----- ----- ----- ----- ----- ----- ----- ----- -p: Display numbers in parseable (exact) values. Also, update iostat syntax to allow the user to specify specific vdevs to show statistics for. The three options for choosing pools/vdevs are: Display a list of pools: zpool iostat ... [pool ...] Display a list of vdevs from a specific pool: zpool iostat ... [pool vdev ...] Display a list of vdevs from any pools: zpool iostat ... [vdev ...] Lastly, allow zpool command "interval" value to be floating point: zpool iostat -v 0.5 Signed-off-by: Tony Hutter <hutter2@llnl.gov Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #4433
2016-03-01 02:05:23 +08:00
2008-11-21 04:01:55 +08:00
void
libzfs_print_on_error(libzfs_handle_t *hdl, boolean_t printerr)
{
hdl->libzfs_printerr = printerr;
}
static int
libzfs_module_loaded(const char *module)
{
const char path_prefix[] = "/sys/module/";
char path[256];
memcpy(path, path_prefix, sizeof (path_prefix) - 1);
strcpy(path + sizeof (path_prefix) - 1, module);
return (access(path, F_OK) == 0);
}
2017-04-22 00:27:04 +08:00
/*
* Read lines from an open file descriptor and store them in an array of
* strings until EOF. lines[] will be allocated and populated with all the
* lines read. All newlines are replaced with NULL terminators for
* convenience. lines[] must be freed after use with libzfs_free_str_array().
*
* Returns the number of lines read.
*/
static int
libzfs_read_stdout_from_fd(int fd, char **lines[])
{
FILE *fp;
int lines_cnt = 0;
size_t len = 0;
char *line = NULL;
char **tmp_lines = NULL, **tmp;
char *nl = NULL;
int rc;
fp = fdopen(fd, "r");
if (fp == NULL)
return (0);
while (1) {
rc = getline(&line, &len, fp);
if (rc == -1)
break;
tmp = realloc(tmp_lines, sizeof (*tmp_lines) * (lines_cnt + 1));
if (tmp == NULL) {
/* Return the lines we were able to process */
break;
}
tmp_lines = tmp;
/* Terminate newlines */
if ((nl = strchr(line, '\n')) != NULL)
*nl = '\0';
tmp_lines[lines_cnt] = line;
lines_cnt++;
line = NULL;
}
fclose(fp);
*lines = tmp_lines;
return (lines_cnt);
}
static int
libzfs_run_process_impl(const char *path, char *argv[], char *env[], int flags,
char **lines[], int *lines_cnt)
{
pid_t pid;
int error, devnull_fd;
2017-04-22 00:27:04 +08:00
int link[2];
/*
* Setup a pipe between our child and parent process if we're
* reading stdout.
*/
if ((lines != NULL) && pipe(link) == -1)
return (-ESTRPIPE);
pid = vfork();
if (pid == 0) {
2017-04-22 00:27:04 +08:00
/* Child process */
devnull_fd = open("/dev/null", O_WRONLY);
if (devnull_fd < 0)
_exit(-1);
2017-04-22 00:27:04 +08:00
if (!(flags & STDOUT_VERBOSE) && (lines == NULL))
(void) dup2(devnull_fd, STDOUT_FILENO);
2017-04-22 00:27:04 +08:00
else if (lines != NULL) {
/* Save the output to lines[] */
dup2(link[1], STDOUT_FILENO);
close(link[0]);
close(link[1]);
}
if (!(flags & STDERR_VERBOSE))
(void) dup2(devnull_fd, STDERR_FILENO);
close(devnull_fd);
2017-04-22 00:27:04 +08:00
if (flags & NO_DEFAULT_PATH) {
if (env == NULL)
execv(path, argv);
else
execve(path, argv, env);
} else {
if (env == NULL)
execvp(path, argv);
else
execvpe(path, argv, env);
}
_exit(-1);
} else if (pid > 0) {
2017-04-22 00:27:04 +08:00
/* Parent process */
int status;
while ((error = waitpid(pid, &status, 0)) == -1 &&
errno == EINTR) { }
if (error < 0 || !WIFEXITED(status))
return (-1);
2017-04-22 00:27:04 +08:00
if (lines != NULL) {
close(link[1]);
*lines_cnt = libzfs_read_stdout_from_fd(link[0], lines);
}
return (WEXITSTATUS(status));
}
return (-1);
}
2017-04-22 00:27:04 +08:00
int
libzfs_run_process(const char *path, char *argv[], int flags)
{
return (libzfs_run_process_impl(path, argv, NULL, flags, NULL, NULL));
}
/*
* Run a command and store its stdout lines in an array of strings (lines[]).
* lines[] is allocated and populated for you, and the number of lines is set in
* lines_cnt. lines[] must be freed after use with libzfs_free_str_array().
* All newlines (\n) in lines[] are terminated for convenience.
*/
int
libzfs_run_process_get_stdout(const char *path, char *argv[], char *env[],
char **lines[], int *lines_cnt)
{
return (libzfs_run_process_impl(path, argv, env, 0, lines, lines_cnt));
}
/*
* Same as libzfs_run_process_get_stdout(), but run without $PATH set. This
* means that *path needs to be the full path to the executable.
*/
int
libzfs_run_process_get_stdout_nopath(const char *path, char *argv[],
char *env[], char **lines[], int *lines_cnt)
{
return (libzfs_run_process_impl(path, argv, env, NO_DEFAULT_PATH,
lines, lines_cnt));
}
/*
* Free an array of strings. Free both the strings contained in the array and
* the array itself.
*/
void
libzfs_free_str_array(char **strs, int count)
{
while (--count >= 0)
free(strs[count]);
free(strs);
}
/*
* Returns 1 if environment variable is set to "YES", "yes", "ON", "on", or
* a non-zero number.
*
* Returns 0 otherwise.
*/
int
libzfs_envvar_is_set(char *envvar)
{
char *env = getenv(envvar);
if (env && (strtoul(env, NULL, 0) > 0 ||
(!strncasecmp(env, "YES", 3) && strnlen(env, 4) == 3) ||
(!strncasecmp(env, "ON", 2) && strnlen(env, 3) == 2)))
return (1);
return (0);
}
Wait in libzfs_init() for the /dev/zfs device While module loading itself is synchronous the creation of the /dev/zfs device is not. This is because /dev/zfs is typically created by a udev rule after the module is registered and presented to user space through sysfs. This small window between module loading and device creation can result in spurious failures of libzfs_init(). This patch closes that race by extending libzfs_init() so it can detect that the modules are loaded and only if required wait for the /dev/zfs device to be created. This allows scripts to reliably use the following shell construct without the need for additional error handling. $ /sbin/modprobe zfs && /sbin/zpool import -a To minimize the potential time waiting in libzfs_init() a strategy similar to adaptive mutexes is employed. The function will busy-wait for up to 10ms based on the expectation that the modules were just loaded and therefore the /dev/zfs will be created imminently. If it takes longer than this it will fall back to polling for up to 10 seconds. This behavior can be customized to some degree by setting the following new environment variables. This functionality is provided for backwards compatibility with existing scripts which depend on the module auto-load behavior. By default module auto-loading is now disabled. * ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules. * ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for /dev/zfs The zfs-import-* systemd service files have been updated to call '/sbin/modprobe zfs' so they no longer rely on the legacy auto-loading behavior. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org> Closes #2556
2015-05-21 05:36:37 +08:00
/*
* Verify the required ZFS_DEV device is available and optionally attempt
* to load the ZFS modules. Under normal circumstances the modules
* should already have been loaded by some external mechanism.
*
* Environment variables:
* - ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules.
* - ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for ZFS_DEV
*/
static int
libzfs_load_module(const char *module)
{
char *argv[4] = {"/sbin/modprobe", "-q", (char *)module, (char *)0};
Wait in libzfs_init() for the /dev/zfs device While module loading itself is synchronous the creation of the /dev/zfs device is not. This is because /dev/zfs is typically created by a udev rule after the module is registered and presented to user space through sysfs. This small window between module loading and device creation can result in spurious failures of libzfs_init(). This patch closes that race by extending libzfs_init() so it can detect that the modules are loaded and only if required wait for the /dev/zfs device to be created. This allows scripts to reliably use the following shell construct without the need for additional error handling. $ /sbin/modprobe zfs && /sbin/zpool import -a To minimize the potential time waiting in libzfs_init() a strategy similar to adaptive mutexes is employed. The function will busy-wait for up to 10ms based on the expectation that the modules were just loaded and therefore the /dev/zfs will be created imminently. If it takes longer than this it will fall back to polling for up to 10 seconds. This behavior can be customized to some degree by setting the following new environment variables. This functionality is provided for backwards compatibility with existing scripts which depend on the module auto-load behavior. By default module auto-loading is now disabled. * ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules. * ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for /dev/zfs The zfs-import-* systemd service files have been updated to call '/sbin/modprobe zfs' so they no longer rely on the legacy auto-loading behavior. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org> Closes #2556
2015-05-21 05:36:37 +08:00
char *load_str, *timeout_str;
long timeout = 10; /* seconds */
long busy_timeout = 10; /* milliseconds */
int load = 0, fd;
hrtime_t start;
/* Optionally request module loading */
if (!libzfs_module_loaded(module)) {
load_str = getenv("ZFS_MODULE_LOADING");
if (load_str) {
if (!strncasecmp(load_str, "YES", strlen("YES")) ||
!strncasecmp(load_str, "ON", strlen("ON")))
load = 1;
else
load = 0;
}
Add kernel module auto-loading Historically a dynamic misc minor number was registered for the /dev/zfs device in order to prevent minor number collisions. This was fine but it prevented us from being able to use the kernel module auto-loaded which requires a known reserved value. Resolve this issue by adding a configure test to find an available misc minor number which can then be used in MODULE_ALIAS_MISCDEV at build time. By adding this alias the zfs kmod is added to the list of known static-nodes and the systemd-tmpfiles-setup-dev service will create a /dev/zfs character device at boot time. This in turn allows us to update the 90-zfs.rules file to make it aware this is a static node. The upshot of this is that whenever a process (zpool, zfs, zed) opens the /dev/zfs the kmods will be automatic loaded. This even works for unprivileged users so there is no longer a need to manually load the modules at boot time. As an additional bonus the zed now no longer needs to start after the zfs-import.service since it will trigger the module load. In the unlikely event the minor number we selected conflicts with another out of tree unregistered minor number the code falls back to dynamically allocating it. In this case the modules again must be manually loaded. Note that due to the change in the method of registering the minor number the zimport.sh test case may incorrectly fail when the static node for the installed packages is created instead of the dynamic one. This issue will only transiently impact zimport.sh for this single commit when we transition and are mixing and matching methods. Reviewed-by: Fabian Grünbichler <f.gruenbichler@proxmox.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> TEST_ZIMPORT_SKIP="yes" Closes #7287
2018-03-14 01:45:55 +08:00
if (load) {
if (libzfs_run_process("/sbin/modprobe", argv, 0))
return (ENOEXEC);
Wait in libzfs_init() for the /dev/zfs device While module loading itself is synchronous the creation of the /dev/zfs device is not. This is because /dev/zfs is typically created by a udev rule after the module is registered and presented to user space through sysfs. This small window between module loading and device creation can result in spurious failures of libzfs_init(). This patch closes that race by extending libzfs_init() so it can detect that the modules are loaded and only if required wait for the /dev/zfs device to be created. This allows scripts to reliably use the following shell construct without the need for additional error handling. $ /sbin/modprobe zfs && /sbin/zpool import -a To minimize the potential time waiting in libzfs_init() a strategy similar to adaptive mutexes is employed. The function will busy-wait for up to 10ms based on the expectation that the modules were just loaded and therefore the /dev/zfs will be created imminently. If it takes longer than this it will fall back to polling for up to 10 seconds. This behavior can be customized to some degree by setting the following new environment variables. This functionality is provided for backwards compatibility with existing scripts which depend on the module auto-load behavior. By default module auto-loading is now disabled. * ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules. * ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for /dev/zfs The zfs-import-* systemd service files have been updated to call '/sbin/modprobe zfs' so they no longer rely on the legacy auto-loading behavior. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org> Closes #2556
2015-05-21 05:36:37 +08:00
Add kernel module auto-loading Historically a dynamic misc minor number was registered for the /dev/zfs device in order to prevent minor number collisions. This was fine but it prevented us from being able to use the kernel module auto-loaded which requires a known reserved value. Resolve this issue by adding a configure test to find an available misc minor number which can then be used in MODULE_ALIAS_MISCDEV at build time. By adding this alias the zfs kmod is added to the list of known static-nodes and the systemd-tmpfiles-setup-dev service will create a /dev/zfs character device at boot time. This in turn allows us to update the 90-zfs.rules file to make it aware this is a static node. The upshot of this is that whenever a process (zpool, zfs, zed) opens the /dev/zfs the kmods will be automatic loaded. This even works for unprivileged users so there is no longer a need to manually load the modules at boot time. As an additional bonus the zed now no longer needs to start after the zfs-import.service since it will trigger the module load. In the unlikely event the minor number we selected conflicts with another out of tree unregistered minor number the code falls back to dynamically allocating it. In this case the modules again must be manually loaded. Note that due to the change in the method of registering the minor number the zimport.sh test case may incorrectly fail when the static node for the installed packages is created instead of the dynamic one. This issue will only transiently impact zimport.sh for this single commit when we transition and are mixing and matching methods. Reviewed-by: Fabian Grünbichler <f.gruenbichler@proxmox.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> TEST_ZIMPORT_SKIP="yes" Closes #7287
2018-03-14 01:45:55 +08:00
if (!libzfs_module_loaded(module))
return (ENXIO);
}
}
Wait in libzfs_init() for the /dev/zfs device While module loading itself is synchronous the creation of the /dev/zfs device is not. This is because /dev/zfs is typically created by a udev rule after the module is registered and presented to user space through sysfs. This small window between module loading and device creation can result in spurious failures of libzfs_init(). This patch closes that race by extending libzfs_init() so it can detect that the modules are loaded and only if required wait for the /dev/zfs device to be created. This allows scripts to reliably use the following shell construct without the need for additional error handling. $ /sbin/modprobe zfs && /sbin/zpool import -a To minimize the potential time waiting in libzfs_init() a strategy similar to adaptive mutexes is employed. The function will busy-wait for up to 10ms based on the expectation that the modules were just loaded and therefore the /dev/zfs will be created imminently. If it takes longer than this it will fall back to polling for up to 10 seconds. This behavior can be customized to some degree by setting the following new environment variables. This functionality is provided for backwards compatibility with existing scripts which depend on the module auto-load behavior. By default module auto-loading is now disabled. * ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules. * ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for /dev/zfs The zfs-import-* systemd service files have been updated to call '/sbin/modprobe zfs' so they no longer rely on the legacy auto-loading behavior. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org> Closes #2556
2015-05-21 05:36:37 +08:00
/*
* Device creation by udev is asynchronous and waiting may be
* required. Busy wait for 10ms and then fall back to polling every
* 10ms for the allowed timeout (default 10s, max 10m). This is
* done to optimize for the common case where the device is
* immediately available and to avoid penalizing the possible
* case where udev is slow or unable to create the device.
*/
timeout_str = getenv("ZFS_MODULE_TIMEOUT");
if (timeout_str) {
timeout = strtol(timeout_str, NULL, 0);
timeout = MAX(MIN(timeout, (10 * 60)), 0); /* 0 <= N <= 600 */
}
start = gethrtime();
do {
fd = open(ZFS_DEV, O_RDWR);
if (fd >= 0) {
(void) close(fd);
return (0);
} else if (errno != ENOENT) {
return (errno);
} else if (NSEC2MSEC(gethrtime() - start) < busy_timeout) {
sched_yield();
} else {
usleep(10 * MILLISEC);
}
} while (NSEC2MSEC(gethrtime() - start) < (timeout * MILLISEC));
Wait in libzfs_init() for the /dev/zfs device While module loading itself is synchronous the creation of the /dev/zfs device is not. This is because /dev/zfs is typically created by a udev rule after the module is registered and presented to user space through sysfs. This small window between module loading and device creation can result in spurious failures of libzfs_init(). This patch closes that race by extending libzfs_init() so it can detect that the modules are loaded and only if required wait for the /dev/zfs device to be created. This allows scripts to reliably use the following shell construct without the need for additional error handling. $ /sbin/modprobe zfs && /sbin/zpool import -a To minimize the potential time waiting in libzfs_init() a strategy similar to adaptive mutexes is employed. The function will busy-wait for up to 10ms based on the expectation that the modules were just loaded and therefore the /dev/zfs will be created imminently. If it takes longer than this it will fall back to polling for up to 10 seconds. This behavior can be customized to some degree by setting the following new environment variables. This functionality is provided for backwards compatibility with existing scripts which depend on the module auto-load behavior. By default module auto-loading is now disabled. * ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules. * ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for /dev/zfs The zfs-import-* systemd service files have been updated to call '/sbin/modprobe zfs' so they no longer rely on the legacy auto-loading behavior. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org> Closes #2556
2015-05-21 05:36:37 +08:00
return (ENOENT);
}
2008-11-21 04:01:55 +08:00
libzfs_handle_t *
libzfs_init(void)
{
libzfs_handle_t *hdl;
Wait in libzfs_init() for the /dev/zfs device While module loading itself is synchronous the creation of the /dev/zfs device is not. This is because /dev/zfs is typically created by a udev rule after the module is registered and presented to user space through sysfs. This small window between module loading and device creation can result in spurious failures of libzfs_init(). This patch closes that race by extending libzfs_init() so it can detect that the modules are loaded and only if required wait for the /dev/zfs device to be created. This allows scripts to reliably use the following shell construct without the need for additional error handling. $ /sbin/modprobe zfs && /sbin/zpool import -a To minimize the potential time waiting in libzfs_init() a strategy similar to adaptive mutexes is employed. The function will busy-wait for up to 10ms based on the expectation that the modules were just loaded and therefore the /dev/zfs will be created imminently. If it takes longer than this it will fall back to polling for up to 10 seconds. This behavior can be customized to some degree by setting the following new environment variables. This functionality is provided for backwards compatibility with existing scripts which depend on the module auto-load behavior. By default module auto-loading is now disabled. * ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules. * ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for /dev/zfs The zfs-import-* systemd service files have been updated to call '/sbin/modprobe zfs' so they no longer rely on the legacy auto-loading behavior. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org> Closes #2556
2015-05-21 05:36:37 +08:00
int error;
2008-11-21 04:01:55 +08:00
Wait in libzfs_init() for the /dev/zfs device While module loading itself is synchronous the creation of the /dev/zfs device is not. This is because /dev/zfs is typically created by a udev rule after the module is registered and presented to user space through sysfs. This small window between module loading and device creation can result in spurious failures of libzfs_init(). This patch closes that race by extending libzfs_init() so it can detect that the modules are loaded and only if required wait for the /dev/zfs device to be created. This allows scripts to reliably use the following shell construct without the need for additional error handling. $ /sbin/modprobe zfs && /sbin/zpool import -a To minimize the potential time waiting in libzfs_init() a strategy similar to adaptive mutexes is employed. The function will busy-wait for up to 10ms based on the expectation that the modules were just loaded and therefore the /dev/zfs will be created imminently. If it takes longer than this it will fall back to polling for up to 10 seconds. This behavior can be customized to some degree by setting the following new environment variables. This functionality is provided for backwards compatibility with existing scripts which depend on the module auto-load behavior. By default module auto-loading is now disabled. * ZFS_MODULE_LOADING="YES|yes|ON|on" - Attempt to load modules. * ZFS_MODULE_TIMEOUT="<seconds>" - Seconds to wait for /dev/zfs The zfs-import-* systemd service files have been updated to call '/sbin/modprobe zfs' so they no longer rely on the legacy auto-loading behavior. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org> Closes #2556
2015-05-21 05:36:37 +08:00
error = libzfs_load_module(ZFS_DRIVER);
if (error) {
errno = error;
return (NULL);
}
if ((hdl = calloc(1, sizeof (libzfs_handle_t))) == NULL) {
2008-11-21 04:01:55 +08:00
return (NULL);
}
if ((hdl->libzfs_fd = open(ZFS_DEV, O_RDWR)) < 0) {
free(hdl);
return (NULL);
}
#ifdef HAVE_SETMNTENT
if ((hdl->libzfs_mnttab = setmntent(MNTTAB, "r")) == NULL) {
#else
2008-11-21 04:01:55 +08:00
if ((hdl->libzfs_mnttab = fopen(MNTTAB, "r")) == NULL) {
#endif
2008-11-21 04:01:55 +08:00
(void) close(hdl->libzfs_fd);
free(hdl);
return (NULL);
}
hdl->libzfs_sharetab = fopen(ZFS_SHARETAB, "r");
2008-11-21 04:01:55 +08:00
Illumos #2882, #2883, #2900 2882 implement libzfs_core 2883 changing "canmount" property to "on" should not always remount dataset 2900 "zfs snapshot" should be able to create multiple, arbitrary snapshots at once Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Chris Siden <christopher.siden@delphix.com> Reviewed by: Garrett D'Amore <garrett@damore.org> Reviewed by: Bill Pijewski <wdp@joyent.com> Reviewed by: Dan Kruchinin <dan.kruchinin@gmail.com> Approved by: Eric Schrock <Eric.Schrock@delphix.com> References: https://www.illumos.org/issues/2882 https://www.illumos.org/issues/2883 https://www.illumos.org/issues/2900 illumos/illumos-gate@4445fffbbb1ea25fd0e9ea68b9380dd7a6709025 Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1293 Porting notes: WARNING: This patch changes the user/kernel ABI. That means that the zfs/zpool utilities built from master are NOT compatible with the 0.6.2 kernel modules. Ensure you load the matching kernel modules from master after updating the utilities. Otherwise the zfs/zpool commands will be unable to interact with your pool and you will see errors similar to the following: $ zpool list failed to read pool configuration: bad address no pools available $ zfs list no datasets available Add zvol minor device creation to the new zfs_snapshot_nvl function. Remove the logging of the "release" operation in dsl_dataset_user_release_sync(). The logging caused a null dereference because ds->ds_dir is zeroed in dsl_dataset_destroy_sync() and the logging functions try to get the ds name via the dsl_dataset_name() function. I've got no idea why this particular code would have worked in Illumos. This code has subsequently been completely reworked in Illumos commit 3b2aab1 (3464 zfs synctask code needs restructuring). Squash some "may be used uninitialized" warning/erorrs. Fix some printf format warnings for %lld and %llu. Apply a few spa_writeable() changes that were made to Illumos in illumos/illumos-gate.git@cd1c8b8 as part of the 3112, 3113, 3114 and 3115 fixes. Add a missing call to fnvlist_free(nvl) in log_internal() that was added in Illumos to fix issue 3085 but couldn't be ported to ZoL at the time (zfsonlinux/zfs@9e11c73) because it depended on future work.
2013-08-28 19:45:09 +08:00
if (libzfs_core_init() != 0) {
(void) close(hdl->libzfs_fd);
(void) fclose(hdl->libzfs_mnttab);
if (hdl->libzfs_sharetab)
(void) fclose(hdl->libzfs_sharetab);
Illumos #2882, #2883, #2900 2882 implement libzfs_core 2883 changing "canmount" property to "on" should not always remount dataset 2900 "zfs snapshot" should be able to create multiple, arbitrary snapshots at once Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Chris Siden <christopher.siden@delphix.com> Reviewed by: Garrett D'Amore <garrett@damore.org> Reviewed by: Bill Pijewski <wdp@joyent.com> Reviewed by: Dan Kruchinin <dan.kruchinin@gmail.com> Approved by: Eric Schrock <Eric.Schrock@delphix.com> References: https://www.illumos.org/issues/2882 https://www.illumos.org/issues/2883 https://www.illumos.org/issues/2900 illumos/illumos-gate@4445fffbbb1ea25fd0e9ea68b9380dd7a6709025 Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1293 Porting notes: WARNING: This patch changes the user/kernel ABI. That means that the zfs/zpool utilities built from master are NOT compatible with the 0.6.2 kernel modules. Ensure you load the matching kernel modules from master after updating the utilities. Otherwise the zfs/zpool commands will be unable to interact with your pool and you will see errors similar to the following: $ zpool list failed to read pool configuration: bad address no pools available $ zfs list no datasets available Add zvol minor device creation to the new zfs_snapshot_nvl function. Remove the logging of the "release" operation in dsl_dataset_user_release_sync(). The logging caused a null dereference because ds->ds_dir is zeroed in dsl_dataset_destroy_sync() and the logging functions try to get the ds name via the dsl_dataset_name() function. I've got no idea why this particular code would have worked in Illumos. This code has subsequently been completely reworked in Illumos commit 3b2aab1 (3464 zfs synctask code needs restructuring). Squash some "may be used uninitialized" warning/erorrs. Fix some printf format warnings for %lld and %llu. Apply a few spa_writeable() changes that were made to Illumos in illumos/illumos-gate.git@cd1c8b8 as part of the 3112, 3113, 3114 and 3115 fixes. Add a missing call to fnvlist_free(nvl) in log_internal() that was added in Illumos to fix issue 3085 but couldn't be ported to ZoL at the time (zfsonlinux/zfs@9e11c73) because it depended on future work.
2013-08-28 19:45:09 +08:00
free(hdl);
return (NULL);
}
2008-11-21 04:01:55 +08:00
zfs_prop_init();
zpool_prop_init();
zpool_feature_init();
2009-07-03 06:44:48 +08:00
libzfs_mnttab_init(hdl);
fletcher_4_init();
2008-11-21 04:01:55 +08:00
if (getenv("ZFS_PROP_DEBUG") != NULL) {
hdl->libzfs_prop_debug = B_TRUE;
}
2008-11-21 04:01:55 +08:00
return (hdl);
}
void
libzfs_fini(libzfs_handle_t *hdl)
{
(void) close(hdl->libzfs_fd);
if (hdl->libzfs_mnttab)
#ifdef HAVE_SETMNTENT
(void) endmntent(hdl->libzfs_mnttab);
#else
2008-11-21 04:01:55 +08:00
(void) fclose(hdl->libzfs_mnttab);
#endif
2008-11-21 04:01:55 +08:00
if (hdl->libzfs_sharetab)
(void) fclose(hdl->libzfs_sharetab);
zfs_uninit_libshare(hdl);
zpool_free_handles(hdl);
2008-11-21 04:01:55 +08:00
namespace_clear(hdl);
2009-07-03 06:44:48 +08:00
libzfs_mnttab_fini(hdl);
Illumos #2882, #2883, #2900 2882 implement libzfs_core 2883 changing "canmount" property to "on" should not always remount dataset 2900 "zfs snapshot" should be able to create multiple, arbitrary snapshots at once Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Chris Siden <christopher.siden@delphix.com> Reviewed by: Garrett D'Amore <garrett@damore.org> Reviewed by: Bill Pijewski <wdp@joyent.com> Reviewed by: Dan Kruchinin <dan.kruchinin@gmail.com> Approved by: Eric Schrock <Eric.Schrock@delphix.com> References: https://www.illumos.org/issues/2882 https://www.illumos.org/issues/2883 https://www.illumos.org/issues/2900 illumos/illumos-gate@4445fffbbb1ea25fd0e9ea68b9380dd7a6709025 Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1293 Porting notes: WARNING: This patch changes the user/kernel ABI. That means that the zfs/zpool utilities built from master are NOT compatible with the 0.6.2 kernel modules. Ensure you load the matching kernel modules from master after updating the utilities. Otherwise the zfs/zpool commands will be unable to interact with your pool and you will see errors similar to the following: $ zpool list failed to read pool configuration: bad address no pools available $ zfs list no datasets available Add zvol minor device creation to the new zfs_snapshot_nvl function. Remove the logging of the "release" operation in dsl_dataset_user_release_sync(). The logging caused a null dereference because ds->ds_dir is zeroed in dsl_dataset_destroy_sync() and the logging functions try to get the ds name via the dsl_dataset_name() function. I've got no idea why this particular code would have worked in Illumos. This code has subsequently been completely reworked in Illumos commit 3b2aab1 (3464 zfs synctask code needs restructuring). Squash some "may be used uninitialized" warning/erorrs. Fix some printf format warnings for %lld and %llu. Apply a few spa_writeable() changes that were made to Illumos in illumos/illumos-gate.git@cd1c8b8 as part of the 3112, 3113, 3114 and 3115 fixes. Add a missing call to fnvlist_free(nvl) in log_internal() that was added in Illumos to fix issue 3085 but couldn't be ported to ZoL at the time (zfsonlinux/zfs@9e11c73) because it depended on future work.
2013-08-28 19:45:09 +08:00
libzfs_core_fini();
fletcher_4_fini();
2008-11-21 04:01:55 +08:00
free(hdl);
}
libzfs_handle_t *
zpool_get_handle(zpool_handle_t *zhp)
{
return (zhp->zpool_hdl);
}
libzfs_handle_t *
zfs_get_handle(zfs_handle_t *zhp)
{
return (zhp->zfs_hdl);
}
zpool_handle_t *
zfs_get_pool_handle(const zfs_handle_t *zhp)
{
return (zhp->zpool_hdl);
}
2008-11-21 04:01:55 +08:00
/*
* Given a name, determine whether or not it's a valid path
* (starts with '/' or "./"). If so, walk the mnttab trying
* to match the device number. If not, treat the path as an
* fs/vol/snap/bkmark name.
2008-11-21 04:01:55 +08:00
*/
zfs_handle_t *
zfs_path_to_zhandle(libzfs_handle_t *hdl, char *path, zfs_type_t argtype)
{
struct stat64 statbuf;
struct extmnttab entry;
int ret;
if (path[0] != '/' && strncmp(path, "./", strlen("./")) != 0) {
/*
* It's not a valid path, assume it's a name of type 'argtype'.
*/
return (zfs_open(hdl, path, argtype));
}
if (stat64(path, &statbuf) != 0) {
(void) fprintf(stderr, "%s: %s\n", path, strerror(errno));
return (NULL);
}
/* Reopen MNTTAB to prevent reading stale data from open file */
if (freopen(MNTTAB, "r", hdl->libzfs_mnttab) == NULL)
return (NULL);
2008-11-21 04:01:55 +08:00
while ((ret = getextmntent(hdl->libzfs_mnttab, &entry, 0)) == 0) {
if (makedevice(entry.mnt_major, entry.mnt_minor) ==
statbuf.st_dev) {
break;
}
}
if (ret != 0) {
return (NULL);
}
if (strcmp(entry.mnt_fstype, MNTTYPE_ZFS) != 0) {
(void) fprintf(stderr, gettext("'%s': not a ZFS filesystem\n"),
path);
return (NULL);
}
return (zfs_open(hdl, entry.mnt_special, ZFS_TYPE_FILESYSTEM));
}
Add helper functions for manipulating device names This change adds two helper functions for working with vdev names and paths. zfs_resolve_shortname() resolves a shorthand vdev name to an absolute path of a file in /dev, /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-path, /dev/disk/by-uuid, /dev/disk/zpool. This was previously done only in the function is_shorthand_path(), but we need a general helper function to implement shorthand names for additional zpool subcommands like remove. is_shorthand_path() is accordingly updated to call the helper function. There is a minor change in the way zfs_resolve_shortname() tests if a file exists. is_shorthand_path() effectively used open() and stat64() to test for file existence, since its scope includes testing if a device is a whole disk and collecting file status information. zfs_resolve_shortname(), on the other hand, only uses access() to test for existence and leaves it to the caller to perform any additional file operations. This seemed like the most general and lightweight approach, and still preserves the semantics of is_shorthand_path(). zfs_append_partition() appends a partition suffix to a device path. This should be used to generate the name of a whole disk as it is stored in the vdev label. The user-visible names of whole disks do not contain the partition information, while the name in the vdev label does. The code was lifted from the function make_disks(), which now just calls the helper function. Again, having a helper function to do this supports general handling of shorthand names in the user interface. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-10-14 07:16:25 +08:00
/*
2012-10-18 07:58:54 +08:00
* Append partition suffix to an otherwise fully qualified device path.
* This is used to generate the name the full path as its stored in
* ZPOOL_CONFIG_PATH for whole disk devices. On success the new length
* of 'path' will be returned on error a negative value is returned.
Add helper functions for manipulating device names This change adds two helper functions for working with vdev names and paths. zfs_resolve_shortname() resolves a shorthand vdev name to an absolute path of a file in /dev, /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-path, /dev/disk/by-uuid, /dev/disk/zpool. This was previously done only in the function is_shorthand_path(), but we need a general helper function to implement shorthand names for additional zpool subcommands like remove. is_shorthand_path() is accordingly updated to call the helper function. There is a minor change in the way zfs_resolve_shortname() tests if a file exists. is_shorthand_path() effectively used open() and stat64() to test for file existence, since its scope includes testing if a device is a whole disk and collecting file status information. zfs_resolve_shortname(), on the other hand, only uses access() to test for existence and leaves it to the caller to perform any additional file operations. This seemed like the most general and lightweight approach, and still preserves the semantics of is_shorthand_path(). zfs_append_partition() appends a partition suffix to a device path. This should be used to generate the name of a whole disk as it is stored in the vdev label. The user-visible names of whole disks do not contain the partition information, while the name in the vdev label does. The code was lifted from the function make_disks(), which now just calls the helper function. Again, having a helper function to do this supports general handling of shorthand names in the user interface. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-10-14 07:16:25 +08:00
*/
int
2012-10-18 07:58:54 +08:00
zfs_append_partition(char *path, size_t max_len)
{
int len = strlen(path);
Add the ZFS Test Suite Add the ZFS Test Suite and test-runner framework from illumos. This is a continuation of the work done by Turbo Fredriksson to port the ZFS Test Suite to Linux. While this work was originally conceived as a stand alone project integrating it directly with the ZoL source tree has several advantages: * Allows the ZFS Test Suite to be packaged in zfs-test package. * Facilitates easy integration with the CI testing. * Users can locally run the ZFS Test Suite to validate ZFS. This testing should ONLY be done on a dedicated test system because the ZFS Test Suite in its current form is destructive. * Allows the ZFS Test Suite to be run directly in the ZoL source tree enabled developers to iterate quickly during development. * Developers can easily add/modify tests in the framework as features are added or functionality is changed. The tests will then always be in sync with the implementation. Full documentation for how to run the ZFS Test Suite is available in the tests/README.md file. Warning: This test suite is designed to be run on a dedicated test system. It will make modifications to the system including, but not limited to, the following. * Adding new users * Adding new groups * Modifying the following /proc files: * /proc/sys/kernel/core_pattern * /proc/sys/kernel/core_uses_pid * Creating directories under / Notes: * Not all of the test cases are expected to pass and by default these test cases are disabled. The failures are primarily due to assumption made for illumos which are invalid under Linux. * When updating these test cases it should be done in as generic a way as possible so the patch can be submitted back upstream. Most existing library functions have been updated to be Linux aware, and the following functions and variables have been added. * Functions: * is_linux - Used to wrap a Linux specific section. * block_device_wait - Waits for block devices to be added to /dev/. * Variables: Linux Illumos * ZVOL_DEVDIR "/dev/zvol" "/dev/zvol/dsk" * ZVOL_RDEVDIR "/dev/zvol" "/dev/zvol/rdsk" * DEV_DSKDIR "/dev" "/dev/dsk" * DEV_RDSKDIR "/dev" "/dev/rdsk" * NEWFS_DEFAULT_FS "ext2" "ufs" * Many of the disabled test cases fail because 'zfs/zpool destroy' returns EBUSY. This is largely causes by the asynchronous nature of device handling on Linux and is expected, the impacted test cases will need to be updated to handle this. * There are several test cases which have been disabled because they can trigger a deadlock. A primary example of this is to recursively create zpools within zpools. These tests have been disabled until the root issue can be addressed. * Illumos specific utilities such as (mkfile) should be added to the tests/zfs-tests/cmd/ directory. Custom programs required by the test scripts can also be added here. * SELinux should be either is permissive mode or disabled when running the tests. The test cases should be updated to conform to a standard policy. * Redundant test functionality has been removed (zfault.sh). * Existing test scripts (zconfig.sh) should be migrated to use the framework for consistency and ease of testing. * The DISKS environment variable currently only supports loopback devices because of how the ZFS Test Suite expects partitions to be named (p1, p2, etc). Support must be added to generate the correct partition name based on the device location and name. * The ZFS Test Suite is part of the illumos code base at: https://github.com/illumos/illumos-gate/tree/master/usr/src/test Original-patch-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Olaf Faaland <faaland1@llnl.gov> Closes #6 Closes #1534
2015-07-02 06:23:09 +08:00
if ((strncmp(path, UDISK_ROOT, strlen(UDISK_ROOT)) == 0) ||
(strncmp(path, ZVOL_ROOT, strlen(ZVOL_ROOT)) == 0)) {
2012-10-18 07:58:54 +08:00
if (len + 6 >= max_len)
return (-1);
(void) strcat(path, "-part1");
len += 6;
} else {
if (len + 2 >= max_len)
return (-1);
if (isdigit(path[len-1])) {
(void) strcat(path, "p1");
len += 2;
} else {
(void) strcat(path, "1");
len += 1;
}
Add helper functions for manipulating device names This change adds two helper functions for working with vdev names and paths. zfs_resolve_shortname() resolves a shorthand vdev name to an absolute path of a file in /dev, /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-path, /dev/disk/by-uuid, /dev/disk/zpool. This was previously done only in the function is_shorthand_path(), but we need a general helper function to implement shorthand names for additional zpool subcommands like remove. is_shorthand_path() is accordingly updated to call the helper function. There is a minor change in the way zfs_resolve_shortname() tests if a file exists. is_shorthand_path() effectively used open() and stat64() to test for file existence, since its scope includes testing if a device is a whole disk and collecting file status information. zfs_resolve_shortname(), on the other hand, only uses access() to test for existence and leaves it to the caller to perform any additional file operations. This seemed like the most general and lightweight approach, and still preserves the semantics of is_shorthand_path(). zfs_append_partition() appends a partition suffix to a device path. This should be used to generate the name of a whole disk as it is stored in the vdev label. The user-visible names of whole disks do not contain the partition information, while the name in the vdev label does. The code was lifted from the function make_disks(), which now just calls the helper function. Again, having a helper function to do this supports general handling of shorthand names in the user interface. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-10-14 07:16:25 +08:00
}
2012-10-18 07:58:54 +08:00
return (len);
Add helper functions for manipulating device names This change adds two helper functions for working with vdev names and paths. zfs_resolve_shortname() resolves a shorthand vdev name to an absolute path of a file in /dev, /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-path, /dev/disk/by-uuid, /dev/disk/zpool. This was previously done only in the function is_shorthand_path(), but we need a general helper function to implement shorthand names for additional zpool subcommands like remove. is_shorthand_path() is accordingly updated to call the helper function. There is a minor change in the way zfs_resolve_shortname() tests if a file exists. is_shorthand_path() effectively used open() and stat64() to test for file existence, since its scope includes testing if a device is a whole disk and collecting file status information. zfs_resolve_shortname(), on the other hand, only uses access() to test for existence and leaves it to the caller to perform any additional file operations. This seemed like the most general and lightweight approach, and still preserves the semantics of is_shorthand_path(). zfs_append_partition() appends a partition suffix to a device path. This should be used to generate the name of a whole disk as it is stored in the vdev label. The user-visible names of whole disks do not contain the partition information, while the name in the vdev label does. The code was lifted from the function make_disks(), which now just calls the helper function. Again, having a helper function to do this supports general handling of shorthand names in the user interface. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-10-14 07:16:25 +08:00
}
/*
2012-10-18 07:58:54 +08:00
* Given a shorthand device name check if a file by that name exists in any
* of the 'zpool_default_import_path' or ZPOOL_IMPORT_PATH directories. If
* one is found, store its fully qualified path in the 'path' buffer passed
* by the caller and return 0, otherwise return an error.
Add helper functions for manipulating device names This change adds two helper functions for working with vdev names and paths. zfs_resolve_shortname() resolves a shorthand vdev name to an absolute path of a file in /dev, /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-path, /dev/disk/by-uuid, /dev/disk/zpool. This was previously done only in the function is_shorthand_path(), but we need a general helper function to implement shorthand names for additional zpool subcommands like remove. is_shorthand_path() is accordingly updated to call the helper function. There is a minor change in the way zfs_resolve_shortname() tests if a file exists. is_shorthand_path() effectively used open() and stat64() to test for file existence, since its scope includes testing if a device is a whole disk and collecting file status information. zfs_resolve_shortname(), on the other hand, only uses access() to test for existence and leaves it to the caller to perform any additional file operations. This seemed like the most general and lightweight approach, and still preserves the semantics of is_shorthand_path(). zfs_append_partition() appends a partition suffix to a device path. This should be used to generate the name of a whole disk as it is stored in the vdev label. The user-visible names of whole disks do not contain the partition information, while the name in the vdev label does. The code was lifted from the function make_disks(), which now just calls the helper function. Again, having a helper function to do this supports general handling of shorthand names in the user interface. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-10-14 07:16:25 +08:00
*/
2012-10-18 07:58:54 +08:00
int
zfs_resolve_shortname(const char *name, char *path, size_t len)
Add helper functions for manipulating device names This change adds two helper functions for working with vdev names and paths. zfs_resolve_shortname() resolves a shorthand vdev name to an absolute path of a file in /dev, /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-path, /dev/disk/by-uuid, /dev/disk/zpool. This was previously done only in the function is_shorthand_path(), but we need a general helper function to implement shorthand names for additional zpool subcommands like remove. is_shorthand_path() is accordingly updated to call the helper function. There is a minor change in the way zfs_resolve_shortname() tests if a file exists. is_shorthand_path() effectively used open() and stat64() to test for file existence, since its scope includes testing if a device is a whole disk and collecting file status information. zfs_resolve_shortname(), on the other hand, only uses access() to test for existence and leaves it to the caller to perform any additional file operations. This seemed like the most general and lightweight approach, and still preserves the semantics of is_shorthand_path(). zfs_append_partition() appends a partition suffix to a device path. This should be used to generate the name of a whole disk as it is stored in the vdev label. The user-visible names of whole disks do not contain the partition information, while the name in the vdev label does. The code was lifted from the function make_disks(), which now just calls the helper function. Again, having a helper function to do this supports general handling of shorthand names in the user interface. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-10-14 07:16:25 +08:00
{
2012-10-18 07:58:54 +08:00
int i, error = -1;
char *dir, *env, *envdup;
env = getenv("ZPOOL_IMPORT_PATH");
errno = ENOENT;
if (env) {
envdup = strdup(env);
dir = strtok(envdup, ":");
while (dir && error) {
(void) snprintf(path, len, "%s/%s", dir, name);
error = access(path, F_OK);
dir = strtok(NULL, ":");
}
free(envdup);
} else {
for (i = 0; i < DEFAULT_IMPORT_PATH_SIZE && error < 0; i++) {
(void) snprintf(path, len, "%s/%s",
zpool_default_import_path[i], name);
error = access(path, F_OK);
}
}
return (error ? ENOENT : 0);
}
/*
* Given a shorthand device name look for a match against 'cmp_name'. This
* is done by checking all prefix expansions using either the default
* 'zpool_default_import_paths' or the ZPOOL_IMPORT_PATH environment
* variable. Proper partition suffixes will be appended if this is a
* whole disk. When a match is found 0 is returned otherwise ENOENT.
*/
static int
zfs_strcmp_shortname(char *name, char *cmp_name, int wholedisk)
{
int path_len, cmp_len, i = 0, error = ENOENT;
char *dir, *env, *envdup = NULL;
char path_name[MAXPATHLEN];
cmp_len = strlen(cmp_name);
env = getenv("ZPOOL_IMPORT_PATH");
if (env) {
envdup = strdup(env);
dir = strtok(envdup, ":");
} else {
dir = zpool_default_import_path[i];
}
while (dir) {
/* Trim trailing directory slashes from ZPOOL_IMPORT_PATH */
while (dir[strlen(dir)-1] == '/')
dir[strlen(dir)-1] = '\0';
path_len = snprintf(path_name, MAXPATHLEN, "%s/%s", dir, name);
if (wholedisk)
path_len = zfs_append_partition(path_name, MAXPATHLEN);
if ((path_len == cmp_len) && strcmp(path_name, cmp_name) == 0) {
2012-10-18 07:58:54 +08:00
error = 0;
break;
}
if (env) {
dir = strtok(NULL, ":");
} else if (++i < DEFAULT_IMPORT_PATH_SIZE) {
dir = zpool_default_import_path[i];
} else {
dir = NULL;
}
}
if (env)
free(envdup);
return (error);
}
/*
* Given either a shorthand or fully qualified path name look for a match
* against 'cmp'. The passed name will be expanded as needed for comparison
* purposes and redundant slashes stripped to ensure an accurate match.
*/
int
zfs_strcmp_pathname(char *name, char *cmp, int wholedisk)
{
int path_len, cmp_len;
char path_name[MAXPATHLEN];
char cmp_name[MAXPATHLEN];
char *dir, *dup;
2012-10-18 07:58:54 +08:00
/* Strip redundant slashes if one exists due to ZPOOL_IMPORT_PATH */
memset(cmp_name, 0, MAXPATHLEN);
dup = strdup(cmp);
dir = strtok(dup, "/");
2012-10-18 07:58:54 +08:00
while (dir) {
strlcat(cmp_name, "/", sizeof (cmp_name));
strlcat(cmp_name, dir, sizeof (cmp_name));
2012-10-18 07:58:54 +08:00
dir = strtok(NULL, "/");
}
free(dup);
2012-10-18 07:58:54 +08:00
if (name[0] != '/')
return (zfs_strcmp_shortname(name, cmp_name, wholedisk));
2012-10-18 07:58:54 +08:00
(void) strlcpy(path_name, name, MAXPATHLEN);
2012-10-18 07:58:54 +08:00
path_len = strlen(path_name);
cmp_len = strlen(cmp_name);
if (wholedisk) {
path_len = zfs_append_partition(path_name, MAXPATHLEN);
if (path_len == -1)
return (ENOMEM);
}
if ((path_len != cmp_len) || strcmp(path_name, cmp_name))
return (ENOENT);
return (0);
Add helper functions for manipulating device names This change adds two helper functions for working with vdev names and paths. zfs_resolve_shortname() resolves a shorthand vdev name to an absolute path of a file in /dev, /dev/disk/by-id, /dev/disk/by-label, /dev/disk/by-path, /dev/disk/by-uuid, /dev/disk/zpool. This was previously done only in the function is_shorthand_path(), but we need a general helper function to implement shorthand names for additional zpool subcommands like remove. is_shorthand_path() is accordingly updated to call the helper function. There is a minor change in the way zfs_resolve_shortname() tests if a file exists. is_shorthand_path() effectively used open() and stat64() to test for file existence, since its scope includes testing if a device is a whole disk and collecting file status information. zfs_resolve_shortname(), on the other hand, only uses access() to test for existence and leaves it to the caller to perform any additional file operations. This seemed like the most general and lightweight approach, and still preserves the semantics of is_shorthand_path(). zfs_append_partition() appends a partition suffix to a device path. This should be used to generate the name of a whole disk as it is stored in the vdev label. The user-visible names of whole disks do not contain the partition information, while the name in the vdev label does. The code was lifted from the function make_disks(), which now just calls the helper function. Again, having a helper function to do this supports general handling of shorthand names in the user interface. Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2010-10-14 07:16:25 +08:00
}
Fix 'zpool import' detection issues This patch addresses multiple 'zpool import' block device indentification problems which are most likely to occur on a system configured to use blkid, by_vdev paths, multipath and failover. The symptom most commonly observed is the import uses different path names to import the pool than would normally be expected. * When using blkid to identify vdevs the listed devices may be added to the cache in any order. In order to apply the preferred search order heuristic a zfs_path_order() function was added to calculate the order given full path names. * Since it's possible to have multiple block devices with different vdev guids which refer to the same ZPOOL_CONFIG_PATH the slice cache must be indexed by guid and name. By avoiding collisions the preferred ordering can be maintaining even when multiple block devices claim the same ZPOOL_CONFIG_PATH. The preferred sorting by partition was never benefitial for a Linux system and was removed as part of this change. * When adding entries to the blkid cache avl_find/avl_insert are used instead of avl_add because collisions are possible and must be handled gracefully. * For pools using multipath devices there are, at a minimum, three devices where a vdev label may be read. They are the dm-* device and each underlying /dev/sd* device. Due to the way the block cache is implemented each of these devices may have a different cached copy of the vdev label. This can result in "ghost pools" which appear to persist even after a 'zpool labelclear' has been done to the dm-* device. In order to prevent this the vdev label is read with O_DIRECT in order to bypass any caching to get the on-disk version. * When opening a block device verify that vdev guid read from the disk matches the expected vdev guid. This allows for bad labels to be filtered out. Reviewed-by: Tony Hutter <hutter2@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #5359
2016-11-08 02:28:57 +08:00
/*
* Given a full path to a device determine if that device appears in the
* import search path. If it does return the first match and store the
* index in the passed 'order' variable, otherwise return an error.
*/
int
zfs_path_order(char *name, int *order)
{
int i = 0, error = ENOENT;
char *dir, *env, *envdup;
env = getenv("ZPOOL_IMPORT_PATH");
if (env) {
envdup = strdup(env);
dir = strtok(envdup, ":");
while (dir) {
if (strncmp(name, dir, strlen(dir)) == 0) {
*order = i;
error = 0;
break;
}
dir = strtok(NULL, ":");
i++;
}
free(envdup);
} else {
for (i = 0; i < DEFAULT_IMPORT_PATH_SIZE; i++) {
if (strncmp(name, zpool_default_import_path[i],
strlen(zpool_default_import_path[i])) == 0) {
*order = i;
error = 0;
break;
}
}
}
return (error);
}
2008-11-21 04:01:55 +08:00
/*
* Initialize the zc_nvlist_dst member to prepare for receiving an nvlist from
* an ioctl().
*/
int
zcmd_alloc_dst_nvlist(libzfs_handle_t *hdl, zfs_cmd_t *zc, size_t len)
{
if (len == 0)
len = 16 * 1024;
2008-11-21 04:01:55 +08:00
zc->zc_nvlist_dst_size = len;
zc->zc_nvlist_dst =
(uint64_t)(uintptr_t)zfs_alloc(hdl, zc->zc_nvlist_dst_size);
if (zc->zc_nvlist_dst == 0)
2008-11-21 04:01:55 +08:00
return (-1);
return (0);
}
/*
* Called when an ioctl() which returns an nvlist fails with ENOMEM. This will
* expand the nvlist to the size specified in 'zc_nvlist_dst_size', which was
* filled in by the kernel to indicate the actual required size.
*/
int
zcmd_expand_dst_nvlist(libzfs_handle_t *hdl, zfs_cmd_t *zc)
{
free((void *)(uintptr_t)zc->zc_nvlist_dst);
zc->zc_nvlist_dst =
(uint64_t)(uintptr_t)zfs_alloc(hdl, zc->zc_nvlist_dst_size);
if (zc->zc_nvlist_dst == 0)
2008-11-21 04:01:55 +08:00
return (-1);
return (0);
}
/*
* Called to free the src and dst nvlists stored in the command structure.
*/
void
zcmd_free_nvlists(zfs_cmd_t *zc)
{
free((void *)(uintptr_t)zc->zc_nvlist_conf);
free((void *)(uintptr_t)zc->zc_nvlist_src);
free((void *)(uintptr_t)zc->zc_nvlist_dst);
zc->zc_nvlist_conf = 0;
zc->zc_nvlist_src = 0;
zc->zc_nvlist_dst = 0;
2008-11-21 04:01:55 +08:00
}
static int
zcmd_write_nvlist_com(libzfs_handle_t *hdl, uint64_t *outnv, uint64_t *outlen,
nvlist_t *nvl)
{
char *packed;
size_t len;
verify(nvlist_size(nvl, &len, NV_ENCODE_NATIVE) == 0);
if ((packed = zfs_alloc(hdl, len)) == NULL)
return (-1);
verify(nvlist_pack(nvl, &packed, &len, NV_ENCODE_NATIVE, 0) == 0);
*outnv = (uint64_t)(uintptr_t)packed;
*outlen = len;
return (0);
}
int
zcmd_write_conf_nvlist(libzfs_handle_t *hdl, zfs_cmd_t *zc, nvlist_t *nvl)
{
return (zcmd_write_nvlist_com(hdl, &zc->zc_nvlist_conf,
&zc->zc_nvlist_conf_size, nvl));
}
int
zcmd_write_src_nvlist(libzfs_handle_t *hdl, zfs_cmd_t *zc, nvlist_t *nvl)
{
return (zcmd_write_nvlist_com(hdl, &zc->zc_nvlist_src,
&zc->zc_nvlist_src_size, nvl));
}
/*
* Unpacks an nvlist from the ZFS ioctl command structure.
*/
int
zcmd_read_dst_nvlist(libzfs_handle_t *hdl, zfs_cmd_t *zc, nvlist_t **nvlp)
{
if (nvlist_unpack((void *)(uintptr_t)zc->zc_nvlist_dst,
zc->zc_nvlist_dst_size, nvlp, 0) != 0)
return (no_memory(hdl));
return (0);
}
int
zfs_ioctl(libzfs_handle_t *hdl, int request, zfs_cmd_t *zc)
{
Illumos #2882, #2883, #2900 2882 implement libzfs_core 2883 changing "canmount" property to "on" should not always remount dataset 2900 "zfs snapshot" should be able to create multiple, arbitrary snapshots at once Reviewed by: George Wilson <george.wilson@delphix.com> Reviewed by: Chris Siden <christopher.siden@delphix.com> Reviewed by: Garrett D'Amore <garrett@damore.org> Reviewed by: Bill Pijewski <wdp@joyent.com> Reviewed by: Dan Kruchinin <dan.kruchinin@gmail.com> Approved by: Eric Schrock <Eric.Schrock@delphix.com> References: https://www.illumos.org/issues/2882 https://www.illumos.org/issues/2883 https://www.illumos.org/issues/2900 illumos/illumos-gate@4445fffbbb1ea25fd0e9ea68b9380dd7a6709025 Ported-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes #1293 Porting notes: WARNING: This patch changes the user/kernel ABI. That means that the zfs/zpool utilities built from master are NOT compatible with the 0.6.2 kernel modules. Ensure you load the matching kernel modules from master after updating the utilities. Otherwise the zfs/zpool commands will be unable to interact with your pool and you will see errors similar to the following: $ zpool list failed to read pool configuration: bad address no pools available $ zfs list no datasets available Add zvol minor device creation to the new zfs_snapshot_nvl function. Remove the logging of the "release" operation in dsl_dataset_user_release_sync(). The logging caused a null dereference because ds->ds_dir is zeroed in dsl_dataset_destroy_sync() and the logging functions try to get the ds name via the dsl_dataset_name() function. I've got no idea why this particular code would have worked in Illumos. This code has subsequently been completely reworked in Illumos commit 3b2aab1 (3464 zfs synctask code needs restructuring). Squash some "may be used uninitialized" warning/erorrs. Fix some printf format warnings for %lld and %llu. Apply a few spa_writeable() changes that were made to Illumos in illumos/illumos-gate.git@cd1c8b8 as part of the 3112, 3113, 3114 and 3115 fixes. Add a missing call to fnvlist_free(nvl) in log_internal() that was added in Illumos to fix issue 3085 but couldn't be ported to ZoL at the time (zfsonlinux/zfs@9e11c73) because it depended on future work.
2013-08-28 19:45:09 +08:00
return (ioctl(hdl->libzfs_fd, request, zc));
2008-11-21 04:01:55 +08:00
}
/*
* ================================================================
* API shared by zfs and zpool property management
* ================================================================
*/
static void
zprop_print_headers(zprop_get_cbdata_t *cbp, zfs_type_t type)
{
zprop_list_t *pl = cbp->cb_proplist;
int i;
char *title;
size_t len;
cbp->cb_first = B_FALSE;
if (cbp->cb_scripted)
return;
/*
* Start with the length of the column headers.
*/
cbp->cb_colwidths[GET_COL_NAME] = strlen(dgettext(TEXT_DOMAIN, "NAME"));
cbp->cb_colwidths[GET_COL_PROPERTY] = strlen(dgettext(TEXT_DOMAIN,
"PROPERTY"));
cbp->cb_colwidths[GET_COL_VALUE] = strlen(dgettext(TEXT_DOMAIN,
"VALUE"));
cbp->cb_colwidths[GET_COL_RECVD] = strlen(dgettext(TEXT_DOMAIN,
"RECEIVED"));
2008-11-21 04:01:55 +08:00
cbp->cb_colwidths[GET_COL_SOURCE] = strlen(dgettext(TEXT_DOMAIN,
"SOURCE"));
2009-01-16 05:59:39 +08:00
/* first property is always NAME */
assert(cbp->cb_proplist->pl_prop ==
((type == ZFS_TYPE_POOL) ? ZPOOL_PROP_NAME : ZFS_PROP_NAME));
2008-11-21 04:01:55 +08:00
/*
* Go through and calculate the widths for each column. For the
* 'source' column, we kludge it up by taking the worst-case scenario of
* inheriting from the longest name. This is acceptable because in the
* majority of cases 'SOURCE' is the last column displayed, and we don't
* use the width anyway. Note that the 'VALUE' column can be oversized,
* if the name of the property is much longer than any values we find.
2008-11-21 04:01:55 +08:00
*/
for (pl = cbp->cb_proplist; pl != NULL; pl = pl->pl_next) {
/*
* 'PROPERTY' column
*/
if (pl->pl_prop != ZPROP_INVAL) {
const char *propname = (type == ZFS_TYPE_POOL) ?
zpool_prop_to_name(pl->pl_prop) :
zfs_prop_to_name(pl->pl_prop);
len = strlen(propname);
if (len > cbp->cb_colwidths[GET_COL_PROPERTY])
cbp->cb_colwidths[GET_COL_PROPERTY] = len;
} else {
len = strlen(pl->pl_user_prop);
if (len > cbp->cb_colwidths[GET_COL_PROPERTY])
cbp->cb_colwidths[GET_COL_PROPERTY] = len;
}
/*
2009-01-16 05:59:39 +08:00
* 'VALUE' column. The first property is always the 'name'
* property that was tacked on either by /sbin/zfs's
* zfs_do_get() or when calling zprop_expand_list(), so we
* ignore its width. If the user specified the name property
* to display, then it will be later in the list in any case.
2008-11-21 04:01:55 +08:00
*/
2009-01-16 05:59:39 +08:00
if (pl != cbp->cb_proplist &&
2008-11-21 04:01:55 +08:00
pl->pl_width > cbp->cb_colwidths[GET_COL_VALUE])
cbp->cb_colwidths[GET_COL_VALUE] = pl->pl_width;
/* 'RECEIVED' column. */
if (pl != cbp->cb_proplist &&
pl->pl_recvd_width > cbp->cb_colwidths[GET_COL_RECVD])
cbp->cb_colwidths[GET_COL_RECVD] = pl->pl_recvd_width;
2008-11-21 04:01:55 +08:00
/*
* 'NAME' and 'SOURCE' columns
*/
if (pl->pl_prop == (type == ZFS_TYPE_POOL ? ZPOOL_PROP_NAME :
ZFS_PROP_NAME) &&
pl->pl_width > cbp->cb_colwidths[GET_COL_NAME]) {
cbp->cb_colwidths[GET_COL_NAME] = pl->pl_width;
cbp->cb_colwidths[GET_COL_SOURCE] = pl->pl_width +
strlen(dgettext(TEXT_DOMAIN, "inherited from"));
}
}
/*
* Now go through and print the headers.
*/
for (i = 0; i < ZFS_GET_NCOLS; i++) {
2008-11-21 04:01:55 +08:00
switch (cbp->cb_columns[i]) {
case GET_COL_NAME:
title = dgettext(TEXT_DOMAIN, "NAME");
break;
case GET_COL_PROPERTY:
title = dgettext(TEXT_DOMAIN, "PROPERTY");
break;
case GET_COL_VALUE:
title = dgettext(TEXT_DOMAIN, "VALUE");
break;
case GET_COL_RECVD:
title = dgettext(TEXT_DOMAIN, "RECEIVED");
break;
2008-11-21 04:01:55 +08:00
case GET_COL_SOURCE:
title = dgettext(TEXT_DOMAIN, "SOURCE");
break;
default:
title = NULL;
}
if (title != NULL) {
if (i == (ZFS_GET_NCOLS - 1) ||
cbp->cb_columns[i + 1] == GET_COL_NONE)
2008-11-21 04:01:55 +08:00
(void) printf("%s", title);
else
(void) printf("%-*s ",
cbp->cb_colwidths[cbp->cb_columns[i]],
title);
}
}
(void) printf("\n");
}
/*
* Display a single line of output, according to the settings in the callback
* structure.
*/
void
zprop_print_one_property(const char *name, zprop_get_cbdata_t *cbp,
const char *propname, const char *value, zprop_source_t sourcetype,
const char *source, const char *recvd_value)
2008-11-21 04:01:55 +08:00
{
int i;
const char *str = NULL;
2008-11-21 04:01:55 +08:00
char buf[128];
/*
* Ignore those source types that the user has chosen to ignore.
*/
if ((sourcetype & cbp->cb_sources) == 0)
return;
if (cbp->cb_first)
zprop_print_headers(cbp, cbp->cb_type);
for (i = 0; i < ZFS_GET_NCOLS; i++) {
2008-11-21 04:01:55 +08:00
switch (cbp->cb_columns[i]) {
case GET_COL_NAME:
str = name;
break;
case GET_COL_PROPERTY:
str = propname;
break;
case GET_COL_VALUE:
str = value;
break;
case GET_COL_SOURCE:
switch (sourcetype) {
case ZPROP_SRC_NONE:
str = "-";
break;
case ZPROP_SRC_DEFAULT:
str = "default";
break;
case ZPROP_SRC_LOCAL:
str = "local";
break;
case ZPROP_SRC_TEMPORARY:
str = "temporary";
break;
case ZPROP_SRC_INHERITED:
(void) snprintf(buf, sizeof (buf),
"inherited from %s", source);
str = buf;
break;
case ZPROP_SRC_RECEIVED:
str = "received";
break;
default:
str = NULL;
assert(!"unhandled zprop_source_t");
2008-11-21 04:01:55 +08:00
}
break;
case GET_COL_RECVD:
str = (recvd_value == NULL ? "-" : recvd_value);
break;
2008-11-21 04:01:55 +08:00
default:
continue;
}
if (i == (ZFS_GET_NCOLS - 1) ||
cbp->cb_columns[i + 1] == GET_COL_NONE)
2008-11-21 04:01:55 +08:00
(void) printf("%s", str);
else if (cbp->cb_scripted)
(void) printf("%s\t", str);
else
(void) printf("%-*s ",
cbp->cb_colwidths[cbp->cb_columns[i]],
str);
}
(void) printf("\n");
}
/*
* Given a numeric suffix, convert the value into a number of bits that the
* resulting value must be shifted.
*/
static int
str2shift(libzfs_handle_t *hdl, const char *buf)
{
const char *ends = "BKMGTPEZ";
int i;
if (buf[0] == '\0')
return (0);
for (i = 0; i < strlen(ends); i++) {
if (toupper(buf[0]) == ends[i])
break;
}
if (i == strlen(ends)) {
if (hdl)
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"invalid numeric suffix '%s'"), buf);
2008-11-21 04:01:55 +08:00
return (-1);
}
/*
* Allow 'G' = 'GB' = 'GiB', case-insensitively.
* However, 'BB' and 'BiB' are disallowed.
2008-11-21 04:01:55 +08:00
*/
if (buf[1] == '\0' ||
(toupper(buf[0]) != 'B' &&
((toupper(buf[1]) == 'B' && buf[2] == '\0') ||
(toupper(buf[1]) == 'I' && toupper(buf[2]) == 'B' &&
buf[3] == '\0'))))
return (10 * i);
2008-11-21 04:01:55 +08:00
if (hdl)
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"invalid numeric suffix '%s'"), buf);
2008-11-21 04:01:55 +08:00
return (-1);
}
/*
* Convert a string of the form '100G' into a real number. Used when setting
* properties or creating a volume. 'buf' is used to place an extended error
* message for the caller to use.
*/
int
zfs_nicestrtonum(libzfs_handle_t *hdl, const char *value, uint64_t *num)
{
char *end;
int shift;
*num = 0;
/* Check to see if this looks like a number. */
if ((value[0] < '0' || value[0] > '9') && value[0] != '.') {
if (hdl)
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"bad numeric value '%s'"), value);
return (-1);
}
/* Rely on strtoull() to process the numeric portion. */
2008-11-21 04:01:55 +08:00
errno = 0;
2009-02-19 04:51:31 +08:00
*num = strtoull(value, &end, 10);
2008-11-21 04:01:55 +08:00
/*
* Check for ERANGE, which indicates that the value is too large to fit
* in a 64-bit value.
*/
if (errno == ERANGE) {
if (hdl)
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"numeric value is too large"));
return (-1);
}
/*
* If we have a decimal value, then do the computation with floating
* point arithmetic. Otherwise, use standard arithmetic.
*/
if (*end == '.') {
double fval = strtod(value, &end);
if ((shift = str2shift(hdl, end)) == -1)
return (-1);
fval *= pow(2, shift);
if (fval > UINT64_MAX) {
if (hdl)
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"numeric value is too large"));
return (-1);
}
*num = (uint64_t)fval;
} else {
if ((shift = str2shift(hdl, end)) == -1)
return (-1);
/* Check for overflow */
if (shift >= 64 || (*num << shift) >> shift != *num) {
if (hdl)
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"numeric value is too large"));
return (-1);
}
*num <<= shift;
}
return (0);
}
/*
* Given a propname=value nvpair to set, parse any numeric properties
* (index, boolean, etc) if they are specified as strings and add the
* resulting nvpair to the returned nvlist.
*
* At the DSL layer, all properties are either 64-bit numbers or strings.
* We want the user to be able to ignore this fact and specify properties
* as native values (numbers, for example) or as strings (to simplify
* command line utilities). This also handles converting index types
* (compression, checksum, etc) from strings to their on-disk index.
*/
int
zprop_parse_value(libzfs_handle_t *hdl, nvpair_t *elem, int prop,
zfs_type_t type, nvlist_t *ret, char **svalp, uint64_t *ivalp,
const char *errbuf)
{
data_type_t datatype = nvpair_type(elem);
zprop_type_t proptype;
const char *propname;
char *value;
boolean_t isnone = B_FALSE;
int err = 0;
2008-11-21 04:01:55 +08:00
if (type == ZFS_TYPE_POOL) {
proptype = zpool_prop_get_type(prop);
propname = zpool_prop_to_name(prop);
} else {
proptype = zfs_prop_get_type(prop);
propname = zfs_prop_to_name(prop);
}
/*
* Convert any properties to the internal DSL value types.
*/
*svalp = NULL;
*ivalp = 0;
switch (proptype) {
case PROP_TYPE_STRING:
if (datatype != DATA_TYPE_STRING) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"'%s' must be a string"), nvpair_name(elem));
goto error;
}
err = nvpair_value_string(elem, svalp);
if (err != 0) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"'%s' is invalid"), nvpair_name(elem));
goto error;
}
2008-11-21 04:01:55 +08:00
if (strlen(*svalp) >= ZFS_MAXPROPLEN) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"'%s' is too long"), nvpair_name(elem));
goto error;
}
break;
case PROP_TYPE_NUMBER:
if (datatype == DATA_TYPE_STRING) {
(void) nvpair_value_string(elem, &value);
if (strcmp(value, "none") == 0) {
isnone = B_TRUE;
} else if (zfs_nicestrtonum(hdl, value, ivalp)
!= 0) {
goto error;
}
} else if (datatype == DATA_TYPE_UINT64) {
(void) nvpair_value_uint64(elem, ivalp);
} else {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"'%s' must be a number"), nvpair_name(elem));
goto error;
}
/*
* Quota special: force 'none' and don't allow 0.
*/
if ((type & ZFS_TYPE_DATASET) && *ivalp == 0 && !isnone &&
(prop == ZFS_PROP_QUOTA || prop == ZFS_PROP_REFQUOTA)) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"use 'none' to disable quota/refquota"));
goto error;
}
/*
* Special handling for "*_limit=none". In this case it's not
* 0 but UINT64_MAX.
*/
if ((type & ZFS_TYPE_DATASET) && isnone &&
(prop == ZFS_PROP_FILESYSTEM_LIMIT ||
prop == ZFS_PROP_SNAPSHOT_LIMIT)) {
*ivalp = UINT64_MAX;
}
2008-11-21 04:01:55 +08:00
break;
case PROP_TYPE_INDEX:
if (datatype != DATA_TYPE_STRING) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"'%s' must be a string"), nvpair_name(elem));
goto error;
}
(void) nvpair_value_string(elem, &value);
if (zprop_string_to_index(prop, value, ivalp, type) != 0) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"'%s' must be one of '%s'"), propname,
zprop_values(prop, type));
goto error;
}
break;
default:
abort();
}
/*
* Add the result to our return set of properties.
*/
if (*svalp != NULL) {
if (nvlist_add_string(ret, propname, *svalp) != 0) {
(void) no_memory(hdl);
return (-1);
}
} else {
if (nvlist_add_uint64(ret, propname, *ivalp) != 0) {
(void) no_memory(hdl);
return (-1);
}
}
return (0);
error:
(void) zfs_error(hdl, EZFS_BADPROP, errbuf);
return (-1);
}
static int
addlist(libzfs_handle_t *hdl, char *propname, zprop_list_t **listp,
zfs_type_t type)
{
int prop;
zprop_list_t *entry;
prop = zprop_name_to_prop(propname, type);
if (prop != ZPROP_INVAL && !zprop_valid_for_type(prop, type, B_FALSE))
prop = ZPROP_INVAL;
/*
* When no property table entry can be found, return failure if
* this is a pool property or if this isn't a user-defined
* dataset property,
*/
if (prop == ZPROP_INVAL && ((type == ZFS_TYPE_POOL &&
!zpool_prop_feature(propname) &&
!zpool_prop_unsupported(propname)) ||
(type == ZFS_TYPE_DATASET && !zfs_prop_user(propname) &&
!zfs_prop_userquota(propname) && !zfs_prop_written(propname)))) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"invalid property '%s'"), propname);
return (zfs_error(hdl, EZFS_BADPROP,
dgettext(TEXT_DOMAIN, "bad property list")));
}
if ((entry = zfs_alloc(hdl, sizeof (zprop_list_t))) == NULL)
return (-1);
entry->pl_prop = prop;
if (prop == ZPROP_INVAL) {
if ((entry->pl_user_prop = zfs_strdup(hdl, propname)) ==
NULL) {
free(entry);
return (-1);
}
entry->pl_width = strlen(propname);
} else {
entry->pl_width = zprop_width(prop, &entry->pl_fixed,
type);
}
*listp = entry;
return (0);
}
2008-11-21 04:01:55 +08:00
/*
* Given a comma-separated list of properties, construct a property list
* containing both user-defined and native properties. This function will
* return a NULL list if 'all' is specified, which can later be expanded
* by zprop_expand_list().
*/
int
zprop_get_list(libzfs_handle_t *hdl, char *props, zprop_list_t **listp,
zfs_type_t type)
{
*listp = NULL;
/*
* If 'all' is specified, return a NULL list.
*/
if (strcmp(props, "all") == 0)
return (0);
/*
* If no props were specified, return an error.
*/
if (props[0] == '\0') {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"no properties specified"));
return (zfs_error(hdl, EZFS_BADPROP, dgettext(TEXT_DOMAIN,
"bad property list")));
}
/*
* It would be nice to use getsubopt() here, but the inclusion of column
* aliases makes this more effort than it's worth.
*/
while (*props != '\0') {
size_t len;
char *p;
char c;
if ((p = strchr(props, ',')) == NULL) {
len = strlen(props);
p = props + len;
2008-11-21 04:01:55 +08:00
} else {
len = p - props;
2008-11-21 04:01:55 +08:00
}
/*
* Check for empty options.
*/
if (len == 0) {
zfs_error_aux(hdl, dgettext(TEXT_DOMAIN,
"empty property name"));
return (zfs_error(hdl, EZFS_BADPROP,
dgettext(TEXT_DOMAIN, "bad property list")));
}
/*
* Check all regular property names.
*/
c = props[len];
props[len] = '\0';
if (strcmp(props, "space") == 0) {
static char *spaceprops[] = {
"name", "avail", "used", "usedbysnapshots",
"usedbydataset", "usedbyrefreservation",
"usedbychildren", NULL
};
int i;
for (i = 0; spaceprops[i]; i++) {
if (addlist(hdl, spaceprops[i], listp, type))
return (-1);
listp = &(*listp)->pl_next;
2008-11-21 04:01:55 +08:00
}
} else {
if (addlist(hdl, props, listp, type))
return (-1);
listp = &(*listp)->pl_next;
2008-11-21 04:01:55 +08:00
}
props = p;
2008-11-21 04:01:55 +08:00
if (c == ',')
props++;
2008-11-21 04:01:55 +08:00
}
return (0);
}
void
zprop_free_list(zprop_list_t *pl)
{
zprop_list_t *next;
while (pl != NULL) {
next = pl->pl_next;
free(pl->pl_user_prop);
free(pl);
pl = next;
}
}
typedef struct expand_data {
zprop_list_t **last;
libzfs_handle_t *hdl;
zfs_type_t type;
} expand_data_t;
int
zprop_expand_list_cb(int prop, void *cb)
{
zprop_list_t *entry;
expand_data_t *edp = cb;
if ((entry = zfs_alloc(edp->hdl, sizeof (zprop_list_t))) == NULL)
return (ZPROP_INVAL);
entry->pl_prop = prop;
entry->pl_width = zprop_width(prop, &entry->pl_fixed, edp->type);
entry->pl_all = B_TRUE;
*(edp->last) = entry;
edp->last = &entry->pl_next;
return (ZPROP_CONT);
}
int
zprop_expand_list(libzfs_handle_t *hdl, zprop_list_t **plp, zfs_type_t type)
{
zprop_list_t *entry;
zprop_list_t **last;
expand_data_t exp;
if (*plp == NULL) {
/*
* If this is the very first time we've been called for an 'all'
* specification, expand the list to include all native
* properties.
*/
last = plp;
exp.last = last;
exp.hdl = hdl;
exp.type = type;
if (zprop_iter_common(zprop_expand_list_cb, &exp, B_FALSE,
B_FALSE, type) == ZPROP_INVAL)
return (-1);
/*
* Add 'name' to the beginning of the list, which is handled
* specially.
*/
if ((entry = zfs_alloc(hdl, sizeof (zprop_list_t))) == NULL)
return (-1);
entry->pl_prop = (type == ZFS_TYPE_POOL) ? ZPOOL_PROP_NAME :
ZFS_PROP_NAME;
entry->pl_width = zprop_width(entry->pl_prop,
&entry->pl_fixed, type);
entry->pl_all = B_TRUE;
entry->pl_next = *plp;
*plp = entry;
}
return (0);
}
int
zprop_iter(zprop_func func, void *cb, boolean_t show_all, boolean_t ordered,
zfs_type_t type)
{
return (zprop_iter_common(func, cb, show_all, ordered, type));
}