Re: [developer] spa_namespace_lock vs sync thread

Andriy, I think this is basically a rule, although I don't think it's stated anywhere. We do rely heavily on this locking strategy since there are many places that will hold the namespace lock to prevent spa config changes but yet wait for a txg to sync. Is it honored everywhere? Well, I hope so

Re: [developer] panic: solaris assert: ba.ba_phys->bt_bytes == 0 (0x400 == 0x0)

Andriy, Can you give me some details about how you're able to reproduce this panic. I would like to help debug this. I'm also looking into the range_tree() panic, so any details you can provide would be very helpful. If you can publish the crash dumps, I can also download them and take a look.

Re: [OpenZFS Developer] [openzfs] Possible access beyond end of string in zpool comment (#47)

LGTM --- Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/47#issuecomment-165781640___ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer

Re: [OpenZFS Developer] [openzfs] ASSERT supported zio_types for file and disk vdevs (#43)

LGTM. --- Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/43#issuecomment-161622259___ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer

Re: [OpenZFS Developer] [openzfs] Fix mutex leak in dmu_objset_find_dp (#44)

LGTM --- Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/44#issuecomment-161621968___ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer

Re: [OpenZFS Developer] [openzfs] 6458 kmem reap thread gets blocked in reclaim callback (#38)

> @@ -1508,7 +1509,15 @@ exi_cache_trim(struct exportinfo *exi) >* used for NFSAUTH_CACHE_TRIM seconds. >*/ > for (c = avl_first(tree); c != NULL; c = AVL_NEXT(tree, c)) { > - rw_enter(>authc_lock, RW_WRITER); > +

Re: [OpenZFS Developer] [openzfs] 6392 zdb: Introduce -V for verbatim import (#22)

LGTM --- Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/22#issuecomment-152831278___ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer

Re: [OpenZFS Developer] [openzfs] 6414 vdev_config_sync could be simpler (#31)

Looks good to me --- Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/31#issuecomment-152831553___ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer

Re: [OpenZFS Developer] [openzfs] 6391 Override default SPA config location via environment (#21)

LGTM --- Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/21#issuecomment-152831323___ developer mailing list developer@open-zfs.org http://lists.open-zfs.org/mailman/listinfo/developer

Re: [OpenZFS Developer] Review Request 250: DLPX-37488 exporting a pool while an async destroy is running can leave entries in the deferred tree

sues/6292 > > > Repository: illumos-gate > > > Description > --- > > 6292 exporting a pool while an async destroy is running can leave entries in > the deferred tree > Reviewed by: Paul Dagnelie <p...@delphix.com> > Reviewed by: Matthew Ahrens

Re: [OpenZFS Developer] Review Request 258: 6319 assertion failed in zio_ddt_write: bp->blk_birth == txg

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/258/#review826 --- Ship it! Ship It! - George Wilson On Oct. 10, 2015, 11:15 p.m

Re: [OpenZFS Developer] cache device sharing

I don't think L2ARC can be shared only spare devices can be shared. - George On Fri, Sep 11, 2015 at 7:23 AM, Andriy Gapon wrote: > > I am curious what was the original reason to not allow sharing of cache > devices > (L2ARC) between pools? > After all, the main ARC

Re: [OpenZFS Developer] cache device sharing

. It would not be a trivial implementation but seems doable. - George On Fri, Sep 11, 2015 at 7:40 AM, Andriy Gapon <andriy.ga...@clusterhq.com> wrote: > On 11/09/2015 16:36, George Wilson wrote: > > I don't think L2ARC can be shared only spare devices can be shared. > > But why

Re: [OpenZFS Developer] rt-rt_space is not zero. range_tree.c:153

Jorgen, Since you're unloading the pool, there should not be any new allocations or frees happening. This is actually prevented by the call to txg_sync_stop() from spa_unload(). Here it should perform the final txg_wait_synced() to clear out all the ms_freetrees and stop the sync thread. You

Re: [OpenZFS Developer] Review Request 229: account for ashift when choosing buffers to be written to l2arc device

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/229/#review764 --- Ship it! Ship It! - George Wilson On June 25, 2015, 12:34 p.m

Re: [OpenZFS Developer] Review Request 229: account for ashift when choosing buffers to be written to l2arc device

like to see review from Saso and George as well. George, Saso, ping. - Andriy On June 25th, 2015, 3:34 p.m. EEST, Andriy Gapon wrote: Review request for OpenZFS Developer Mailing List, Justin Gibbs, George Wilson, Matthew Ahrens, and Saso Kiselkov. By Andriy Gapon. *Updated June 25

Re: [OpenZFS Developer] rt-rt_space is not zero. range_tree.c:153

Jorgen, Can you provide some details about the stack trace when you hit this failure. All of the ms_freetrees should be empty by the time you can range_tree_destroy(). So to debug this we need to understand the calling stack to determine why that isn't happening. Thanks, George On Sat, Aug 22,

Re: [OpenZFS Developer] Review Request 219: 5909 ensure that shared snap names don't become too long after promotion

. - George Wilson On May 6, 2015, 12:21 p.m., Andriy Gapon wrote: --- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/219/ --- (Updated May 6

Re: [OpenZFS Developer] Review Request 217: Fix stack overflow and panic on FreeBSD.

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/217/#review728 --- Ship it! Ship It! - George Wilson On May 8, 2015, 3:29 a.m., Gleb

Re: [OpenZFS Developer] Review Request 226: 1778 Assertion failed: rn-rn_nozpool == B_FALSE, file ../common/libzfs_import.c, line 1077, function zpool_open_func

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/226/#review735 --- Ship it! Ship It! - George Wilson On May 27, 2015, 1:51 p.m

Re: [OpenZFS Developer] Review Request 172: 5694 traverse_prefetcher does not prefetch enough

christopher.si...@delphix.com Original author: George Wilson Diffs - usr/src/uts/common/fs/zfs/dmu_traverse.c 14eef2e7516c9b5e717ef318321ee137dcd19c9f Diff: https://reviews.csiden.org/r/172/diff/ Testing --- ztest; zfs test suite http://jenkins/job/zfs-precommit

Re: [OpenZFS Developer] Review Request 167: 5669 altroot not set in zpool create when specified with -o

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/167/#review543 --- Ship it! Ship It! - George Wilson On Feb. 27, 2015, 7:45 p.m., Xin

Re: [OpenZFS Developer] Review Request 163: 5630 stale bonus buffer in recycled dnode_t leads to data corruption

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/163/#review530 --- Ship it! Ship It! - George Wilson On Feb. 21, 2015, 5:34 a.m

Re: [OpenZFS Developer] Review Request 155: 5531 NULL pointer dereference in dsl_prop_get_ds()

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/155/#review477 --- Ship it! Ship It! - George Wilson On Feb. 4, 2015, 4:59 a.m

Re: [OpenZFS Developer] Review Request 152: 5527 ::spa_space fails with mdb: couldn't find member dd_phys of type struct zfs`dsl_dir'

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/152/#review461 --- Ship it! Ship It! - George Wilson On Jan. 27, 2015, 5:55 a.m

Re: [OpenZFS Developer] Review Request 147: 5438 zfs_blkptr_verify should continue after zfs_panic_recover

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/147/#review407 --- Ship it! Ship It! - George Wilson On Dec. 16, 2014, 6:39 a.m., Xin

Re: [OpenZFS Developer] Review Request 128: 5313 Allow I/Os to be aggregated across ZIO priority classes

--- This is an automatically generated e-mail. To reply, visit: https://reviews.csiden.org/r/128/#review408 --- Ship it! Ship It! - George Wilson On Dec. 2, 2014, 12:22 p.m

Re: [OpenZFS Developer] Review Request 131: [5056] ZFS deadlock on db_mtx and dn_holds/Various improvements the dmu buf user API.

Comment block missing start and end lines. usr/src/uts/sparc/stmf_sbd/Makefile https://reviews.csiden.org/r/131/#comment348 Comment block missing start and end lines. - George Wilson On Dec. 15, 2014, 4:48 a.m., Justin Gibbs wrote

Re: [OpenZFS Developer] [zfs] Recover from corrupted space map (illumos #4390)

On 7/7/14, 3:33 AM, Jan Schmidt via illumos-zfs wrote: On Wed, June 25, 2014 at 16:15 (+0200), Keith Wesolowski Via Illumos-zfs wrote: On Wed, Jun 25, 2014 at 01:47:54PM +0200, Jan Schmidt via illumos-zfs wrote: That patch looks somewhat promising, though I have not tried it yet. How did

Re: [OpenZFS Developer] zfs zio reordering in zio_vdev_io_start causing panics

On 5/7/14, 10:55 AM, Steven Hartland wrote: - Original Message - From: George Wilson george.wil...@delphix.com To: Steven Hartland kill...@multiplay.co.uk; developer@open-zfs.org Sent: Wednesday, May 07, 2014 3:07 PM Subject: Re: [OpenZFS Developer] zfs zio reordering

Re: [OpenZFS Developer] zfs zio reordering in zio_vdev_io_start causing panics

On 5/7/14, 11:46 AM, Steven Hartland wrote: - Original Message - From: George Wilson george.wil...@delphix.com There are a couple of cases where it can work but I'm going to make it such that requires you to always return back ZIO_PIPELINE_STOP. Otherwise it's makes it too easy

Re: [OpenZFS Developer] zfs zio reordering in zio_vdev_io_start causing panics

On 5/7/14, 12:50 PM, Steven Hartland wrote: - Original Message - From: George Wilson george.wil...@delphix.com On 5/7/14, 11:46 AM, Steven Hartland wrote: - Original Message - From: George Wilson george.wil...@delphix.com There are a couple of cases where it can work

Re: [OpenZFS Developer] zfs zio reordering in zio_vdev_io_start causing panics

Steven, I think there is probably another way we can solve this problem but I first want to get a better understanding of the corruption. We have not integrated the TRIM support upstream and I suspect that's the source of most of the problems. Can you confirm that with TRIM disabled that most

Re: [OpenZFS Developer] permanent error in a pool

Hmm, 'zpool scrub' should be able to clear that error. Can you tell me which options you used to 'zinject' and I'll try reproducing it here. Thanks, George On 2/26/14 4:22 AM, Andriy Gapon wrote: I did some experimenting and used zinject to introduce some non-correctable data errors while

Re: [OpenZFS Developer] l2arc_write_buffers scanning efficiency

LGTM. - George On 1/14/14 12:57 PM, Saso Kiselkov wrote: On 1/14/14, 11:51 AM, Andriy Gapon wrote: Recently I noticed that on one of our systems l2arc_feed_thread was consuming non insignificant amount of CPU time. I looked at some stats (FreeBSD has local changes that add a few kstats for

Re: [OpenZFS Developer] Metaslab allocation question

Ilya, I think there is a bigger problem here that your vdev_mirror() change would cover up. The problems is that we should never have a bp with a dva that is out of range. So we need to understand how that is occurring. As for you question about metaslab_allocate_dva(), I'm assuming you're

Re: [OpenZFS Developer] zio parent and zio children relation.

On 12/17/13 5:21 AM, Gaurav Mahajan wrote: Hi all, I am trying to understand relations of root ZIO and children ZIO. This is what I have understood.. Please correct me if I'm wrong. Usually whenever we want to do a series of IO operations like in sync thread. We create a root ZIO with

Re: [OpenZFS Developer] priority for TRIM zio

Andriy, The fix looks good but I have a couple of questions: 1. Are you sure you want the TRIM priority to be lower than SCRUB? 2. You have zfs_vdev_trim_max_active set to 1 with an /* XXX */ next to it. Does this mean you haven't settled on the final value for active max? Thanks, George On