louwrentius [Fri, 29 Aug 2014 22:14:20 +0000 (00:14 +0200)]
Change delimiter for ZED email scripts
When the ZED_EMAIL_INTERVAL_SECS="3600" option is set in zed.rc
configuration file then notification emails should be rate limited.
Rate limiting is accomplished by maintaining a colon delimited state
file which includes the device name. Unfortunately there are valid
device names which include a colon and therefore prevent the rate
limiting for working properly. For this reason the delimiter has
been changed to a semi-colon.
Signed-off-by: louwrentius <louwrentius@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Chris Dunlap <cdunlap@llnl.gov>
Closes #2645
Change the startup mode of ZED to non-forking. While systemd can
track processes that detach from the terminal just fine, running
processes in non-forking mode is the preferred mode of operation.
Also remove user/group definitions as root/root is the default.
Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2252
Chris Dunlap [Thu, 28 Aug 2014 21:39:48 +0000 (14:39 -0700)]
Cleanup zed logging
This is a set of minor cleanup changes related to zed logging:
- Remove the program identity prefix from messages written to stderr
since systemd already prepends this output with the program name.
- Replace the copy of the program identity string with a ptr reference.
- Replace "pid" with "PID" for consistency in comments & strings.
- Rename the zed_log.c struct _ctx component "level" to "priority".
- Add the LOG_PID option for messages written to syslog.
Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2252
Chris Dunlap [Wed, 27 Aug 2014 20:18:01 +0000 (13:18 -0700)]
Fix race condition with zed pidfile creation
When the zed is started as a forking daemon (by default),
a race-condition exists where the parent process can terminate before
the pidfile has been created by the grandchild process. When invoked
as a Type=forking systemd service, this can result in the following:
systemd[1]: Starting ZFS Event Daemon (zed)...
systemd[1]: PID file /var/run/zed.pid not readable (yet?) after start.
This commit adds a daemonize pipe to allow the grandchild process to
signal the parent process that initialization is complete (and the
pidfile has been created). The parent process will wait for this
notification before exiting.
Signed-off-by: Chris Dunlap <cdunlap@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2252
Providing a pkg-config file makes is easy for 3rd party applications
to link against the libzfs libraries. It also allows the libzfs
developers to modify the list of required libraries and cflags
without breaking existing applications.
The following example illustrates how pkg-config can be used:
cc `pkg-config --cflags --libs libzfs` -o myapp myapp.c
Brian Behlendorf [Wed, 27 Aug 2014 21:12:18 +0000 (14:12 -0700)]
Retire HAVE_IOCTL_* configure checks
The HAVE_IOCTL_* configure checks were originally added for
compatibility with an ancient version of glibc. This support
and additional complexity is no longer needed and is therefore
being removed.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Turbo Fredriksson <turbo@bayour.com>
Closes #585
Matthew Ahrens [Tue, 15 Jul 2014 18:58:41 +0000 (10:58 -0800)]
Illumos 4970-4974 - extreme rewind enhancements
4970 need controls on i/o issued by zpool import -XF
4971 zpool import -T should accept hex values
4972 zpool import -T implies extreme rewind, and thus a scrub
4973 spa_load_retry retries the same txg
4974 spa_load_verify() reads all data twice
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Approved by: Robert Mustacchi <rm@joyent.com>
Notes:
This set of patches adds a set of tunable parameters for the
"extreme rewind" mode of pool import which allows control over
the traversal performed during such an import.
Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2598
Matthew Ahrens [Wed, 20 Aug 2014 17:09:40 +0000 (10:09 -0700)]
Illumos 5034 - ARC's buf_hash_table is too small
5034 ARC's buf_hash_table is too small
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com>
Reviewed by: Richard Elling <richard.elling@gmail.com>
Approved by: Gordon Ross <gwr@nexenta.com>
Andrew Hamilton [Wed, 13 Aug 2014 14:58:08 +0000 (10:58 -0400)]
2493 change efi_rescan() to wait longer
Change efi_rescan() to loop 10 times instead of 5 on EBUSY and
to sleep at the end of each loop. This helps with some instances
where the kernel does not reload the partition table fast enough
for ZFS to detect.
Signed-off-by: Andrew Hamilton <ahamilto@tjhsst.edu> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2493
Matthew Ahrens [Tue, 15 Jul 2014 07:43:18 +0000 (03:43 -0400)]
Illumos 4631 - zvol_get_stats triggering too many reads
4631 zvol_get_stats triggering too many reads
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Sebastien Roy <sebastien.roy@delphix.com>
Reviewed by: Matt Ahrens <mahrens@delphix.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Richard Yao [Thu, 14 Aug 2014 18:59:12 +0000 (14:59 -0400)]
Drive database update
The Intel DC S3500 and Intel DC S3700 are optimized to handle 4KB
sectors well despite of their 8KB page sizes, so we move them to a new
category for enterprise drives where they will receive ashift=12. They
are joined by the Intel 730 series, which uses the same disk controller,
as well as a San Disk enterprise drive. The drive IDs for these two were
obtained by myself with the drive_id utility. The drive ID for the 240GB
Intel 730 model was extrapolated from the drive ID for the 480GB model.
Lastly, we also add some Western Digital mobile drives. ryuo in
\#zfsonlinux on freenode obtained "ATA WDC WD2500BEVT-0" from
running drive_id on his own hardware. The additional drives in that
family were extrapolated from that identifer.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2601
Tim Chase [Thu, 14 Aug 2014 12:01:20 +0000 (07:01 -0500)]
Don't upgrade a metaslab when the pool is not writable
Illumos 4982 added code to metaslab_fragmentation() to proactively update
space maps when the spacemap_histogram feature is enabled. This should
only happen when the pool is writeable.
George Wilson [Sat, 19 Jul 2014 20:19:24 +0000 (12:19 -0800)]
Illumos 4976-4984 - metaslab improvements
4976 zfs should only avoid writing to a failing non-redundant top-level vdev
4978 ztest fails in get_metaslab_refcount()
4979 extend free space histogram to device and pool
4980 metaslabs should have a fragmentation metric
4981 remove fragmented ops vector from block allocator
4982 space_map object should proactively upgrade when feature is enabled
4983 need to collect metaslab information via mdb
4984 device selection should use fragmentation metric
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Adam Leventhal <adam.leventhal@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
Notes:
The "zdb -M" option has been re-tasked to display the new metaslab
fragmentation metric and the new "zdb -I" option is used to control
the maximum number of in-flight I/Os.
The new fragmentation metric is derived from the space map histogram
which has been rolled up to the vdev and pool level and is presented
to the user via "zpool list".
Add a number of module parameters related to the new metaslab weighting
logic.
Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2595
Add a new 'overlay' property (default 'off') that controls whether the
filesystem should be mounted even if the mountpoint is busy or if it
should fail with a 'mountpoint not empty'.
Doing overlay mounts is the default mount behavior on Linux, but not
in ZFS. It have been decided that following the ZFS behavior should
be the default, but this overlay allows for site administrator to
override this decision on a per-dataset basis.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes: #2503
Richard Yao [Fri, 20 Jun 2014 17:23:15 +0000 (13:23 -0400)]
Include sys/taskq.h in linux/vfs_compat.h
We should have included sys/taskq.h directly because we use the taskq
code here, but we instead had files that included sys/taskq.h also
include sys/kmem.h, which happened to include sys/taskq.h. sys/kmem.h no
longer does this, so we must define the include as we should
have done in the first place.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2411
Evan Susarret [Thu, 20 Mar 2014 20:21:34 +0000 (16:21 -0400)]
Correct autodetection of bootfs property
Remove lines that contain only a hyphen (match '^-$' instead of '-').
I had a root fs with a hyphen in the name (fedora/ROOT/Fedora20-Dev),
it was not detected because sed eliminated that line of output from
'zpool list -Ho bootfs'.
Alec Salazar [Tue, 5 Aug 2014 13:10:54 +0000 (09:10 -0400)]
Replace __va_list with va_list
Most of the code base already uses va_list, which is specified by
iso-c. gcc/glibc provides 'typedef __gnuc_va_list va_list'. and
when not using gcc/glibc we can't expect to find __gnuc_va_list.
Signed-off-by: Alec Salazar <alec.j.salazar@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2588
This reverts commit 7973e46 which brings the basic flow of the
code back in line with the other ZFS implementations. This
was possible due to the following related changes.
e89260a Directory xattr znodes hold a reference on their parent 6f9548c Fix deadlock in zfs_zget() 0a50679 Add zfs_iput_async() interface 4dd1893 Avoid 128K kmem allocations in mzap_upgrade()
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org>
Closes #457
Closes #2058
Closes #2128
Closes #2240
Handle all iputs in zfs_purgedir() and zfs_inode_destroy()
asynchronously to prevent deadlocks. When the iputs are allowed
to run synchronously in the destroy call path deadlocks between
xattr directory inodes and their parent file inodes are possible.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org>
Closes #457
As originally implemented the mzap_upgrade() function will
perform up to SPA_MAXBLOCKSIZE allocations using kmem_alloc().
These large allocations can potentially block indefinitely
if contiguous memory is not available. Since this allocation
is done under the zap->zap_rwlock it can appear as if there is
a deadlock in zap_lockdir(). This is shown below.
The optimal fix for this would be to rework mzap_upgrade()
such that no large allocations are required. This could be
done but it would result in us diverging further from the other
implementations. Therefore I've opted against doing this
unless it becomes absolutely necessary.
Instead mzap_upgrade() has been updated to use zio_buf_alloc()
which can reliably provide buffers of up to SPA_MAXBLOCKSIZE.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Richard Yao <ryao@gentoo.org>
Close #2580
As part of commit e8b96c6 the search zio used by the
vdev_queue_io_to_issue() function was moved to the heap
to minimize stack usage. Functionally this is fine, but
to maximize performance it's best to minimize the number
of dynamic allocations.
To avoid this allocation temporary space for the search
zio has been reserved in the vdev_queue structure. All
access must be serialized through the vq_lock.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov>
Closes #2572
The dsl_dataset_rollback_check() function is executed in the
txg_sync context. To prevent a potential deadlock due to direct
memory reclaim it must use KM_PUSHPAGE. This was introduced by
the recent 'zfs bookmark' features, commit da53684.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Eric Dillmann <eric@jave.fr>
Closes #2569
Matthew Ahrens [Wed, 25 Jun 2014 18:37:59 +0000 (10:37 -0800)]
Illumos 4914 - zfs on-disk bookmark structure should be named *_phys_t
4914 zfs on-disk bookmark structure should be named *_phys_t
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Richard Lowe <richlowe@richlowe.net>
Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com>
Approved by: Robert Mustacchi <rm@joyent.com>
There were a number of zfsonlinux-specific uses of zbookmark_t which
needed to be updated. This should reduce the likelihood of further
problems like issue #2094 from occurring.
Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2558
Matthew Ahrens [Thu, 5 Jun 2014 21:34:21 +0000 (13:34 -0800)]
Illumos 4881 - zfs send performance regression with embedded data
4881 zfs send performance degradation when embedded block pointers
are encountered
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Saso Kiselkov [Thu, 22 May 2014 09:11:57 +0000 (10:11 +0100)]
Illumos 4897 - Space accounting mismatch in L2ARC/zpool
4897 Space accounting mismatch in L2ARC/zpool
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Boris Protopopov <bprotopopov@hotmail.com>
Approved by: Dan McDonald <danmcd@omniti.com>
From the illumos issue tracker:
L2ARC vdev space usage statistics are calculated as the delta
between the maximum and minimum vdev offset ever written to
by the L2ARC fill thread, but do not inform the user of how
much space in between these two offsets is actually taken up by
cached buffers. This fix changes that so that vdev space usage
stats on L2ARC devices accurately track the volume of buffers
stored on them, allowing users to see the exact L2ARC usage in
"zpool iostat -v".
Matthew Ahrens [Thu, 5 Jun 2014 21:20:08 +0000 (13:20 -0800)]
Illumos 4390 - I/O errors can corrupt space map when deleting fs/vol
4390 i/o errors when deleting filesystem/zvol can lead to space map corruption
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Previous stack-reduction efforts in traverse_visitb() caused a fair
number of un-mergable pieces of code. This patch should reduce its
stack footprint a bit more.
The new local bptree_entry_phys_t in bptree_add() is dynamically-allocated
using kmem_zalloc() for the purpose of stack reduction.
The new global zfs_free_leak_on_eio has been defined as an integer
rather than a boolean_t as was the case with the related zfs_recover
global. Also, zfs_free_leak_on_eio's definition has been inserted into
zfs_debug.c for consistency with the existing definition of zfs_recover.
Illumos placed it in spa_misc.c.
Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2545
Matthew Ahrens [Thu, 5 Jun 2014 21:19:08 +0000 (13:19 -0800)]
Illumos 4757, 4913
4757 ZFS embedded-data block pointers ("zero block compression")
4913 zfs release should not be subject to space checks
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Max Grossman <max.grossman@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Dan McDonald <danmcd@omniti.com>
For compatibility with the fastpath code the zio_done() function
needed to be updated. Because embedded-data block pointers do
not require DVAs to be allocated the associated vdevs will not
be marked and therefore should not be unmarked.
Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2544
Matthew Ahrens [Fri, 23 May 2014 16:21:07 +0000 (08:21 -0800)]
Illumos 3835 zfs need not store 2 copies of all metadata
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Richard Lowe <richlowe@richlowe.net>
Description from Matt Ahrens's bug report at Delphix:
Add a new zfs property, "redundant_metadata" which can have values
"all" or "most". The default will be "all", which is the current
behavior. Setting to "most" will cause us to only store 1 copy of
level-1 indirect blocks of user data files.
Additional notes:
The new man page section for this property states
"The exact behavior of which metadata blocks
are stored redundantly may change in future releases."
and:
"When set to most, ZFS stores an extra copy of most types of
metadata. This can improve performance of random writes,
because less metadata must be written."
The current implementation is as described above in Matt's blog.
It is controlled by a new global integer
"zfs_redundant_metadata_most_ditto_level", currently initialized
to 2. When "redundant_metadata" is set to "most", only indirect
blocks of the specified level and higher will have additional ditto
blocks created.
Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2542
Tim Chase [Sun, 15 Jun 2014 13:47:40 +0000 (08:47 -0500)]
zed needs libzfs_core
As of a recent group of Illumos/Delphix updates, zed needs libzfs_core
in order to resolve lzc_get_bookmarks() and likely other functions
going forward.
Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2534
George Wilson [Fri, 18 Apr 2014 16:35:03 +0000 (08:35 -0800)]
Illumos 4754, 4755
4754 io issued to near-full luns even after setting noalloc threshold
4755 mg_alloc_failures is no longer needed
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Matthew Ahrens [Wed, 16 Apr 2014 03:40:22 +0000 (19:40 -0800)]
Illumos #4374
4374 dn_free_ranges should use range_tree_t
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Max Grossman <max.grossman@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com
Reviewed by: Garrett D'Amore <garrett@damore.org>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Max Grossman [Mon, 9 Dec 2013 18:37:51 +0000 (10:37 -0800)]
Illumos 4370, 4371
4370 avoid transmitting holes during zfs send
4371 DMU code clean up
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Josef 'Jeff' Sipek <jeffpc@josefsipek.net>
Approved by: Garrett D'Amore <garrett@damore.org>a
Matthew Ahrens [Tue, 8 Oct 2013 17:13:05 +0000 (09:13 -0800)]
Illumos 4171, 4172
4171 clean up spa_feature_*() interfaces
4172 implement extensible_dataset feature for use by other zpool features
Reviewed by: Max Grossman <max.grossman@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Jerry Jelinek <jerry.jelinek@joyent.com>
Approved by: Garrett D'Amore <garrett@damore.org>a
Brian Behlendorf [Fri, 25 Jul 2014 19:22:45 +0000 (12:22 -0700)]
Remove patches directory
Support for ZFS has now been merged in to both blkid and grub.
Therefore, there is no longer a need to carry these stale
patches in the ZFS source tree.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
In no way complete - most have been trial and error and some
deducing what they could mean. It needs more information from
someone that knows the code better. But this is a start and
it lays the basic structure for adding this additional detail.
Brian Behlendorf [Tue, 22 Jul 2014 21:43:22 +0000 (14:43 -0700)]
zpool-create.sh: allow features to be disabled
The zimport.sh script makes use of the zpool-create.sh script
to construct test pools for importing with older versions of
ZoL. It is desirable to have a way to disable all the features
so new pools can be imported with older code.
The simplest and most flexible way to achieve this was to merge
the VERBOSE_FLAG and FORCE_FLAG in to a single ZPOOL_FLAGS
variable. The contents of this variable will be used in the
'zpool create' allowing us to easily pass arbitrary flags.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Prakash Surya <surya1@llnl.gov>
Closes #2524
Jan Engelhardt [Fri, 18 Jul 2014 18:00:27 +0000 (20:00 +0200)]
Do not attempt access beyond the declared end of the dn_blkptr array
This loop in dmu_objset_write_ready():
for (i = 0; i < dnp->dn_nblkptr; i++)
bp->blk_fill += dnp->dn_blkptr[i].blk_fill;
invokes _undefined behavior_ for the (common) case of dn_nblkptr=3,
therefore, the compiler is free to do whatever it wants (such as
optimizing it away, or otherwise messing up your expections).
The fix is to be honest about the array size.
Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2511
Closes #2010
Brian Behlendorf [Mon, 21 Jul 2014 17:19:25 +0000 (10:19 -0700)]
zfs_trunc() should use dmu_tx_assign(tx, TXG_WAIT)
As part of the write throttle & i/o schedule performance work the
zfs_trunc() function should have been updated to use TXG_WAIT.
Using TXG_WAIT ensures that the tx will be part of the next txg.
If TXG_NOWAIT is used and retried for ERESTART errors then the
tx can suffer from starvation.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov>
Closes #2488
George Wilson [Thu, 12 Jun 2014 23:29:11 +0000 (15:29 -0800)]
Illumos #4756 Fix metaslab_group_preload deadlock
4756 metaslab_group_preload() could deadlock
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Christopher Siden <christopher.siden@delphix.com>
Reviewed by: Dan McDonald <danmcd@omniti.com>
Reviewed by: Saso Kiselkov <saso.kiselkov@nexenta.com>
Approved by: Garrett D'Amore <garrett@damore.org>
The metaslab_group_preload() function grabs the mg_lock and then later
tries to grab the metaslab lock. This lock ordering may lead to a
deadlock since other consumers of the mg_lock will grab the metaslab
lock first.
George Wilson [Tue, 1 Oct 2013 21:25:53 +0000 (13:25 -0800)]
Illumos #4101, #4102, #4103, #4105, #4106
4101 metaslab_debug should allow for fine-grained control
4102 space_maps should store more information about themselves
4103 space map object blocksize should be increased
4105 removing a mirrored log device results in a leaked object
4106 asynchronously load metaslab
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Sebastien Roy <seb@delphix.com>
Approved by: Garrett D'Amore <garrett@damore.org>
Prior to this patch, space_maps were preferred solely based on the
amount of free space left in each. Unfortunately, this heuristic didn't
contain any information about the make-up of that free space, which
meant we could keep preferring and loading a highly fragmented space map
that wouldn't actually have enough contiguous space to satisfy the
allocation; then unloading that space_map and repeating the process.
This change modifies the space_map's to store additional information
about the contiguous space in the space_map, so that we can use this
information to make a better decision about which space_map to load.
This requires reallocating all space_map objects to increase their
bonus buffer size sizes enough to fit the new metadata.
The above feature can be enabled via a new feature flag introduced by
this change: com.delphix:spacemap_histogram
In addition to the above, this patch allows the space_map block size to
be increase. Currently the block size is set to be 4K in size, which has
certain implications including the following:
* 4K sector devices will not see any compression benefit
* large space_maps require more metadata on-disk
* large space_maps require more time to load (typically random reads)
Now the space_map block size can adjust as needed up to the maximum size
set via the space_map_max_blksz variable.
A bug was fixed which resulted in potentially leaking an object when
removing a mirrored log device. The previous logic for vdev_remove() did
not deal with removing top-level vdevs that are interior vdevs (i.e.
mirror) correctly. The problem would occur when removing a mirrored log
device, and result in the DTL space map object being leaked; because
top-level vdevs don't have DTL space map objects associated with them.
This changes moves the called to metaslab_group_alloc_update() to the
metaslab_sync_reassess() function. The original placement of the call
within metaslab_sync_done() appears to have been a simple mistake,
introduced by ac72fac3eaa569902cad88053167f7d74e7fe7e4.
This aligns us more closely to the upstream illumos code base.
Signed-off-by: Prakash Surya <surya1@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Brian Behlendorf [Tue, 15 Jul 2014 20:29:57 +0000 (13:29 -0700)]
Fix zil_commit() NULL dereference
Update the current code to ensure inodes are never dirtied if they are
part of a read-only file system or snapshot. If they do somehow get
dirtied an attempt will make made to write them to disk. In the case
of snapshots, which don't have a ZIL, this will result in a NULL
dereference in zil_commit().
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2405
Richard Yao [Sun, 8 Jun 2014 18:10:14 +0000 (14:10 -0400)]
zdb: Introduce -V for verbatim import
When given a pool name via -e, zdb would attempt an import. If it
failed, then it would attempt a verbatim import. This behavior is
not always desirable so a -V switch is added to zdb to control the
behavior. When specified, a verbatim import is done. Otherwise,
the behavior is as it was previously, except no verbatim import
is done on failure.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2372
George Wilson [Fri, 4 Oct 2013 22:13:23 +0000 (14:13 -0800)]
Illumos 4168, 4169, 4170: ztest, zdb and zhack fixes
4168 ztest assertion failure in dbuf_undirty
4169 verbatim import causes zdb to segfault
4170 zhack leaves pool in ACTIVE state
Reviewed by: Adam Leventhal <ahl@delphix.com>
Reviewed by: Eric Schrock <eric.schrock@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Approved by: Dan McDonald <danmcd@nexenta.com>
Of particular interest when troubleshooting corrupted pools, the
commonly-used "zdb -e" operation may perform verbatim imports and
furthermore, it will soon have direct support for verbatim imports via
a new "-V" option. The 4169 fix eliminates a common segfault case in
which spa_history_log_version() tries to access an un-opened dsl_pool_t.
Ported by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2451
Closes #2283
Closes #2467
Tim Chase [Thu, 10 Jul 2014 03:36:03 +0000 (22:36 -0500)]
Convert zfs_mg_noalloc_threshold to a module parameter and document
The parameter was added as illumos issue 4081 which was committed to
zfsonlinux in ac72fac3eaa569902cad88053167f7d74e7fe7e4. This patch
documents the parameter and allows for it to be set as a module parameter.
Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2483
Matthew Ahrens [Sun, 24 Mar 2013 21:24:51 +0000 (13:24 -0800)]
Illumos #3641 compressed block histograms with zdb
This patch is a zdb extension of the '-b' option, producing a histogram
of the physical compressed block sizes per DMU object type on disk. The
'-bbbb' option to zdb will uncover this new feature; here's an example
usage on a new pool and snippet of the output it generates:
The blocks are also broken down by their indirection level. Expanding on
the above example:
# zfs set recordsize=1k tank
# dd bs=1k if=/dev/urandom of=/tank/2x1kfile count=2
# zdb -bbbb tank
...
1 16K 1K 2K 2K 16.00 1.02 L1 ZFS plain file
psize (in 512-byte sectors): number of blocks
2: 1 *
5 70.0K 70.0K 70.0K 14.0K 1.00 35.71 L0 ZFS plain file
psize (in 512-byte sectors): number of blocks
2: 3 ***
3: 0
4: 0
5: 0
6: 1 *
7: 0
...
127: 0
128: 1 *
6 86.0K 71.0K 72.0K 12.0K 1.21 36.73 ZFS plain file
psize (in 512-byte sectors): number of blocks
2: 4 ****
3: 0
4: 0
5: 0
6: 1 *
7: 0
...
127: 0
128: 1 *
...
There's now a single 1K L1 block which is the indirect block needed for
the '2x1kfile' file just created, as well as two more 1K L0 blocks from
the same file.
This can be used to get a distribution of the block sizes used within
the pool, on a per object type basis.
Andrew Barnes [Tue, 10 Jun 2014 06:29:12 +0000 (16:29 +1000)]
Preserve asize when last mirror child promoted to top-level vdev
If the smaller of 2 different sized child vdev's of a mirrored vdev is
detached, and the pool has the autoexpand property set to off, as the
remaining larger vdev is promoted to a top level vdev it fails to retain
the asize of the original top level mirror vdev and therefore partially
autoexpands.
This partially autoexpanded state leaves the new vdev too large to
re-mirror by adding the smaller vdev back in, and the pool fails to
utilize the space until next imported.
If the autoexpand property is set to on, the child vdev grows
in size after it has been promoted to a top level vdev as expected.
This commit causes the remaining child mirror to retain the asize of its
old parent mirror vdev if the autoexpand property is set to off,
this allows the smaller vdev to be re-added if required the vdev
can then be told to expand if required by the usual using zpool online -e.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Andrew Barnes <barnes333@gmail.com> Signed-off-by: George Wilson <george.wilson@delphix.com>
Closes #1208
The referenced fix to get_numeric_property() caused numeric property
lookups to consider the type of the parent (head) dataset when checking
validity but there are some cases in the caller expects to see the
property's default value even when the lookup is invalid.
One case in which this is true is change_one() which is part of the
renaming infrastructure. It may look up "zoned" on a snapshot of a volume
which is not valid but it expects to see the default value of false.
There may be other, yet unidentified cases in which zfs_prop_get_int()
is used on technically invalid properties but which expect the property's
default value to be returned.
Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Turbo Fredriksson <turbo@bayour.com>
Closes #2320
Dan McDonald [Tue, 24 Jun 2014 02:25:02 +0000 (22:25 -0400)]
Illumos #4936 fix potential overflow in lz4
4936 lz4 could theoretically overflow a pointer with a certain input
Reviewed by: Saso Kiselkov <skiselkov.ml@gmail.com>
Reviewed by: Keith Wesolowski <keith.wesolowski@joyent.com>
Approved by: Gordon Ross <gordon.ross@nexenta.com>
Ported by: Tim Chase <tim@chase2k.com>
This fixes the widely-reported "20-year-old vulnerability" in
LZO/LZ4 implementations which inherited said bug from the reference
implementation.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2429
Tim Chase [Sat, 28 Jun 2014 14:08:44 +0000 (09:08 -0500)]
Comment the lack of real_LZ4_uncompress()
Added several comments regarding the removal of real_LZ4_uncompress()
which exists in the reference implementation but has been removed here
since it's not used.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Brian Behlendorf [Thu, 26 Jun 2014 23:36:13 +0000 (16:36 -0700)]
Improve differing sector size error
When adding or replacing a vdev with a different sector size the
error message should be more useful. In addition to describing
the problem provide a hint that the '-o ashift' option can be
used to override the optimal default value.
Since using a non-optimal value may incur a significant performance
penalty we should issue this error. But there a numerous reasons
why a administrator may wish to do this anyway.
Signed-off-by: Niklas Edmundsson <ZNikke@github> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2421
Add information about the -o option to zpool replace
Users need to be aware that when replacing devices in an existing
pool they may need to override automatically detected ashift value.
This will all depend on the exact hardware they are using.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2024
Brian Behlendorf [Thu, 12 Jun 2014 20:11:26 +0000 (13:11 -0700)]
Fix zfs.spec.in defaults
Commit 2ee4e7da accidentally introduced two issues which only occur
when rebuilding the ZFS source rpm outside the ZFS build system.
1) The _dracutdir, _udevdir, and _udevruledir macros must be checked
using the 'undefined' keyword. This was just overlooked in the
patch review and does not cause a failure when using 'make pkg'
because the values are provided by the make target.
2) The default _udevruledir path included a typo.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2310
Accept udev and dracut paths specified by ./configure
There are two common locations where udev and dracut components are
commonly installed. When building packages using the 'make rpm|deb'
targets check those common locations and pass them to rpmbuild. For
non-standard configurations these values can be provided by the
the following configure options:
When rebuilding using the source packages the per-distribution
default values specified in the spec file will be used. This is
the preferred way to build packages for a distribution but the
ability to override the defaults is provided as a convenience.
1) While we have slightly more headroom in user space excessively
deep stacks have been observed while running ztest, see #2293.
2) Removing this conditional makes the pipeline behave consistently
regardless of if it's executing in kernel space or user space.
This way we're more likely to uncover subtle issues with ztest.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2384
These options have existed for a long time but have historically
been undocumented because they are not guaranteed to be safe. They
should only be used as a last resort when attempting to recover a
damaged pool.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1130
* Remove the references to share(1M), unshare(1M) and dfstab(4)
since they are not applicable to Linux.
* Add the exact exportfs command line used when setting sharenfs=on.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue: #1641
Document the fact that ashift is vdev specific, not a pool global.
Users need to be aware that when adding devices to an existing pool
they may need to override automatically detected ashift value.
This will all depend on the exact hardware they are using.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Closes: #2024
Only automatically mount a clone when 'canmount == on'.
According to the man page, "When the noauto option is set, a dataset
can only be mounted and unmounted explicitly. The dataset is not
mounted automatically when the dataset is created or imported ...."
When cloning a dataset the canmount property was not being honored.
This patch adds the required check to achieve the behavior described
in the man page.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2241
Accept kernel source dir(s) specified by ./configure
This adds ability to set the location of the kernel via defines
when building from the spec files. This is useful when building
against a kernel installed in a non-standard location.
Signed-off-by: Turbo Fredriksson <turbo@bayour.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1874
From day one the various ZFS libraries should have been placed in their
own sub-packages. Primarily this allows for multiple major versions of
the libraries to be concurrently installed. It also facilitates a
smaller build environment by minimizing the required dependencies.
The specific changes required to split the libraries from the utilities
are as follows:
* libzpool2, libnvpair1, libuutil1, and libzfs2 packages were added
and contain the versioned shared libraries. The Fedora packaging
guidelines discourage providing static libraries so they are not
included in the packages.
* The zfs-devel package was renamed libzfs2-devel and the new package
obsoletes the old zfs-devel package. This package includes all
the required headers for the libzpool2, libnvpair1, libuutil1, and
libzfs2 libraries and their respective unversioned shared libraries.
This package should eventually be split in to individual lib*-devel
packages but it will still take some work to cleanly separate them.
Therefore the libzfs2-devel package provides the expected lib*-devel
packages so the all proper dependencies can still be created.
* The /usr/share/doc/ files were moved in to the libzfs2-devel package.
* Updated config/deb.am to be aware of the packaging changes. This
ensures that 'deb-utils' make target converts all the resulting
packages generated by the 'rpm-utils' target.
Richard Yao [Wed, 23 Apr 2014 04:04:41 +0000 (00:04 -0400)]
Remove superfluous statement
Clang's static analyzer reported that the value assigned to pcksum is
never used. That is because we initialize both zc and pcksum to {{ 0 }}
and then do `pcksum = zc;`. That is fairly pointless. However, it has
the effect of generating a false positive in Clang's static analyzer.
Since noise from false positives can obscure real issues, we fix it
anyway.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2330
Richard Yao [Wed, 23 Apr 2014 03:35:38 +0000 (23:35 -0400)]
Fix memory leak in zpool_clear_label()
Clang's static analyzer reported a memory leak in zpool_clear_label().
Upon review, it turns out to be right. This should be a very short lived
leak because no daemons use this functionality, but that does not
preclude the possibility of third party daemons that do use it. Lets fix
it to be a good Samaritan.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #2330
Chris Wedgwood [Tue, 27 May 2014 00:09:45 +0000 (17:09 -0700)]
Allow building without ACLs
Some kernel definitions were buried inside the #if... #endif logic for
ACLs. When ACLs are not available these definitions get lost causing
the build to fail.
Signed-off-by: Chris Wedgwood <cw@f00f.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2349
Brian Behlendorf [Thu, 29 May 2014 20:36:24 +0000 (13:36 -0700)]
Fix DKMS package upgrade and packager
Running 'yum upgrade zfs-dkms' package could appear to work properly
and still leave you with no zfs modules installed. This will occur
when only the zfs release, and not the version, are incremented.
This may be the case for a fast moving zfs-testing repository.
During the upgrade process DKMS will realize that zfs-x.y.z is already
installed and remove it. DKMS then correctly builds the new modules
for zfs-x.y.z. However, as a final step when the old zfs-x.y.z-r is
removed the %preun script runs and removes the newly build modules.
To handle this case the %preun script has been updated to only run
when the installed version exactly matches the full spec file version.
This change also updated ChangeLog section based on the DKMS
reference spec file.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Brian Behlendorf [Thu, 29 May 2014 22:03:43 +0000 (15:03 -0700)]
Restrict release number to META version
When creating packages in a git repository the release number
can be automatically set by 'git describe'. This normally works
well but if your repository has newer tags which match the form
NAME-VERSION* the release may be incorrectly calculated. To
prevent this the match patten has been restricted to the contents
of the META file, NAME-VERSION.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Brian Behlendorf [Thu, 15 May 2014 01:17:39 +0000 (18:17 -0700)]
Use default slab types
We should not override the default memory type of the kmem cache. This
was done previously to force certain objects which were slightly over
object size limit cut off in to KMC_KMEM caches for better performance.
The zfsonlinux/spl#356 patch slightly increases the default cut off
from 511 bytes 1024 bytes for x86_64. This means there is long longer
a need to override the default for the caches. And since the default
values are now being used the new spl_kmem_cache_slab_limit and
spl_kmem_cache_kmem_limit tunables will apply to all kmem caches.
The following is a list of caches that will be impacted:
Marcel Huber [Wed, 21 May 2014 09:17:23 +0000 (11:17 +0200)]
Omit compiler warning by sticking to RAII
Resolve gcc 4.9.0 20140507 warnings about uninitialized 'ptr' when
using -Wmaybe-uninitialized. The first two cases appears appear
to be legitimate but not the second two. In general this is a
good practice so they are all initialized.
Signed-off-by: Marcel Huber <marcelhuberfoo@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2345
Tim Chase [Sun, 11 May 2014 01:13:12 +0000 (20:13 -0500)]
Calculate header size correctly in sa_find_sizes()
In the case where a variable-sized SA overlaps the spill block pointer and
a new variable-sized SA is being added, the header size was improperly
calculated to include the to-be-moved SA. This problem could be
reproduced when xattr=sa enabled as follows:
ln -s $(perl -e 'print "x" x 120') blah
setfattr -n security.selinux -v blahblah -h blah
The symlink is large enough to interfere with the spill block pointer and
has a typical SA registration as follows (shown in modified "zdb -dddd"
<SA attr layout obj> format):
[ ... ZPL_DACL_COUNT ZPL_DACL_ACES ZPL_SYMLINK ]
Adding the SA xattr will attempt to extend the registration to:
Tim Chase [Thu, 8 May 2014 14:51:01 +0000 (09:51 -0500)]
Allow for lock-free reading zfsdev_state_list.
Restructure the zfsdev_state_list to allow for lock-free reading by
converting to a simple singly-linked list from which items are never
deleted and over which only forward iterations are performed. It depends
on, among other things, the atomicity of accessing the zs_minor integer
and zs_next pointer.
This fixes a lock inversion in which the zfsdev_state_lock is used by
both the sync task (txg_sync) and indirectly by any user program which
uses /dev/zfs; the zfsdev_release method uses the same lock and then
blocks on the sync task.
The most typical failure scenerio occurs when the sync task is cleaning
up a user hold while various concurrent "zfs" commands are in progress.
Neither Illumos nor Solaris are affected by this issue because they use
DDI interface which provides lock-free reading of device state via the
ddi_get_soft_state() function.
Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2301
Richard Yao [Sat, 17 May 2014 13:35:10 +0000 (09:35 -0400)]
Handle ZPOOL_STATUS_HOSTID_MISMATCH in zpool status
Verbatim imports can cause hostid mismatches, but things otherwise work. `zpool
status` does not handle this and will fail when assertions are enabled:
Lets instead add a case to display an informative message such as this:
```
pool: rpool
state: ONLINE
status: Mismatch between pool hostid and system hostid on imported pool.
This pool was previously imported into a system with a different hostid,
and then was verbatim imported into this system.
action: Export this pool on all systems on which it is imported.
Then import it to correct the mismatch.
see: http://zfsonlinux.org/msg/ZFS-8000-EY
scan: scrub repaired 0 in 0h8m with 0 errors on Thu Apr 17 19:43:57 2014
config:
NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
sda ONLINE 0 0 0
errors: No known data errors
```
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2342
Chunwei Chen [Tue, 13 May 2014 02:36:35 +0000 (10:36 +0800)]
Use a dedicated taskq for vdev_file
Originally, vdev_file used system_taskq. This would cause a deadlock,
especially on system with few CPUs. The reason is that the prefetcher
threads, which are on system_taskq, will sometimes be blocked waiting
for I/O to finish. If the prefetcher threads consume all the tasks in
system_taskq, the I/O cannot be served and thus results in a deadlock.
We fix this by creating a dedicated vdev_file_taskq for vdev_file I/O.
Signed-off-by: Chunwei Chen <tuxoko@gmail.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2270
Handle vdev_lookup_top() failure in dva_get_dsize_sync()
The dva_get_dsize_sync() function incorrectly assumes that the call
to vdev_lookup_top() cannot fail. However, the NULL dereference at
clearly shows that under certain circumstances it is possible. Note
that offset 0x570 (1376) maps as expected to vd->vdev_deflate_ratio.
BUG: unable to handle kernel NULL pointer dereference at 00000570
Given that this can happen this patch add the required error handling.
In the case where vdev_lookup_top() fails assume that no deflation
will occur for the DVA and use the asize.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Alexey Zhuravlev <alexey.zhuravlev@intel.com>
Closes #1707
Closes #1987
Closes #1891
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Tim Chase [Mon, 21 Apr 2014 18:22:08 +0000 (13:22 -0500)]
Check the dataset type more rigorously when fetching properties.
When fetching property values of snapshots, a check against the head
dataset type must be performed. Previously, this additional check was
performed only when fetching "version", "normalize", "utf8only" or "case".
This caused the ZPL properties "acltype", "exec", "devices", "nbmand",
"setuid" and "xattr" to be erroneously displayed with meaningless values
for snapshots of volumes. It also did not allow for the display of
"volsize" of a snapshot of a volume.
This patch adds the headcheck flag paramater to zfs_prop_valid_for_type()
and zprop_valid_for_type() to indicate the check is being done
against a head dataset's type in order that properties valid only for
snapshots are handled correctly. This allows the the head check in
get_numeric_property() to be performed when fetching a property for
a snapshot.
Signed-off-by: Tim Chase <tim@chase2k.com> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2265
1) This is a small piece of the metaslab improvement patch from
Illumos. It was worth bringing over before the rest, since it's
low risk and it can be useful on fragmented pools (e.g. Lustre
MDTs). metaslab_debug_unload would give the performance benefit
of the old metaslab_debug option without causing unwanted delay
during pool import.
Ported-by: Ned Bass <bass6@llnl.gov> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #2227
When the system attributes (SAs) for an object exceed what can
can be stored in the bonus area of a dnode a spill block is
allocated. These spill blocks are currently considered data
blocks. However, they should be accounted for as meta data
because they are effectively an extension of the dnode.
While this may seem like a minor accounting issue it has broader
implications. The key thing to be aware of is that each spill
block will hold a reference on its parent dnode. The dnode in
turn holds a reference on its dbuf in the dnode object. This
means that a single 512 byte data buffer for a spill block can
pin over 16k of meta data. This is analogous to the small file
situation described in 2b13331 where a relatively small number
of data buffer can cause the ARC to exceed the meta limit.
However, unlike the small file case a spill block can legitimately
be considered meta data. By changing the spill block to meta data
they will now be dropped from the cache when the meta limit is
reached. This then allows the dnodes and dbufs which the spill
block was pinning to be released.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Prakash Surya <surya1@llnl.gov>
Closes #2294
The default SELinux policy for RHEL and Fedora has been updated
to include ZFS in the list of filesystems which support xattrs.
Therefore, there's no longer a need to detect this in the init
scripts.
Richard Yao [Wed, 30 Apr 2014 16:51:28 +0000 (12:51 -0400)]
ztest: Switch to LWP rwlock interface
ztest is intended to subject the ZFS code in userland to stress that it
should be able to withstand. Any failures that occur when running it are
failures that likely would occur inside the kernel. However, being in
userland, it is much easier to debug them. In practice, this prevents
a large number of problems from reaching production code.
A design decision was made by the original authors of ztest to make a
distinction between userland locking primitives and kernel locking
primitives. The ztest code itself calls userland locking primitives
while the kernel code being run in userland will call emulated kernel
locking primitives that wrap the userland locking primitives.
When ztest was first ported to Linux, a decision was made to use the
emulated kernel interfaces everywhere. In effect, the userland
rw_rdlock()/rw_wrlock() became the kernel rw_enter() and and the userland
rw_unlock() became the kernel rw_exit(). This caused a regression
because of an assertion in rw_enter() to catch recursive locking. That
is permitted in userland, but not in the kernel. Consequently, the ztest
code itself does recursive read locking. The use of the emulated kernel
interfaces consequently caused the following failure:
That occurs because ztest_dmu_objset_create_destroy() will take a read
lock and call ztest_dmu_object_alloc_free(). That will call ztest_io(),
which will take a readlock only when asked to do ZTEST_IO_REWRITE. This
triggered the assertion.
The pthreads rwlock interface was based on the LWP rwlock interface
implemented in Illumos libc. Luckily enough, the subset used by ztest is
almost identical, so we can solve this problem by switching to the LWP
thread rwlock interface in ztest. This eliminates a point of divergence
with Illumos and should make code sharing slightly easier.
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #1970
Richard Yao [Wed, 30 Apr 2014 17:12:27 +0000 (13:12 -0400)]
libspl: Implement LWP rwlock interface
This implements a subset of the LWP rwlock interface by wrapping the
equivalent POSIX thread interface. It is a superset of the features
needed by ztest.
The missing bits are {,_}rw_read_held() and {,_}rw_write_held().
Signed-off-by: Richard Yao <ryao@gentoo.org> Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Issue #1970
Brian Behlendorf [Mon, 28 Apr 2014 20:56:47 +0000 (13:56 -0700)]
dmu_tx_assign() should not return ENOMEM
As described in the comment above dmu_tx_assign() this function must
only fail if the pool is out of space. If for some other reason the
TX cannot be assigned (such as memory pressure) ERESTART must be
returned. Alternately, EAGAIN could be returned to inject a delay
but that isn't required because the caller will block on the condition
variable waiting for the next TXG.
/*
* Assign tx to a transaction group. txg_how can be one of:
*
* (1) TXG_WAIT. If the current open txg is full, waits until there's
* a new one. This should be used when you're not holding locks.
* It will only fail if we're truly out of space (or over quota).
* ...
*/
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov> Signed-off-by: Ned Bass <bass6@llnl.gov>
Closes #2287