svn commit: r268855 - in head: cddl/contrib/opensolaris/cmd/ztest sys/cddl/contrib/opensolaris/uts/common/fs/zfs sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys
Nikolai Lifanov
lifanov at mail.lifanov.com
Mon Jul 21 03:49:58 UTC 2014
On 2014-07-20 22:56, Steven Hartland wrote:
> ----- Original Message ----- From: "Nikolai Lifanov"
> <lifanov at mail.lifanov.com>
> To: "Xin LI" <delphij at freebsd.org>
> Cc: <src-committers at freebsd.org>; <svn-src-head at freebsd.org>
> Sent: Monday, July 21, 2014 2:22 AM
> Subject: Re: svn commit: r268855 - in head:
> cddl/contrib/opensolaris/cmd/ztest
> sys/cddl/contrib/opensolaris/uts/common/fs/zfs
> sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys
>
>
>> On 2014-07-18 16:41, Xin LI wrote:
>>> Author: delphij
>>> Date: Fri Jul 18 20:41:40 2014
>>> New Revision: 268855
>>> URL: http://svnweb.freebsd.org/changeset/base/268855
>>>
>>> Log:
>>> MFV r268848:
>>>
>>> Instead of asserting all zio's be properly aligned, only assert
>>> on the logical ones.
>>>
>>> Cap uberblocks at 8k, otherwise with ashift=17, there would be
>>> only one uberblock.
>>>
>>> This fixes a problem that zdb would trip assert on pools with
>>> ashift >= 0xe (8k).
>>>
>>> While there, also change the code so it only attempt to condense
>>> space map unless the uncondensed size consumes greater than
>>> zfs_metaslab_condense_block_threshold blocks.
>>>
>>> Illumos issue:
>>> 4958 zdb trips assert on pools with ashift >= 0xe
>>>
>>> MFC after: 2 weeks
>>>
>>
>> This specific commit makes boot hang just before mounting the root
>> dataset for me when vfs.zfs.vdev.cache.size tunable is set. Unsetting
>> this tunable or reverting this commit (currently running r268933 minus
>> r268855) fixes the boot for me.
>>
>> Please let me know if I can provide any more information.
>
> Interesting, is there a reason you enabled vdev caching as its been
> disabled by default for quite some time with the comment:
>
> * TODO: Note that with the current ZFS code, it turns out that the
> * vdev cache is not helpful, and in some cases actually harmful. It
> * is better if we disable this. Once some time has passed, we should
> * actually remove this to simplify the code. For now we just disable
> * it by setting the zfs_vdev_cache_size to zero. Note that Solaris 11
> * has made these same changes.
>
> Regards
> Steve
>
> _______________________________________________
> svn-src-head at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/svn-src-head
> To unsubscribe, send any mail to "svn-src-head-unsubscribe at freebsd.org"
I noticed that it was disabled on default and I'm not seeing any ill
effect now that I have it disabled. I don't find vdev cache beneficial,
but I had it in my loader.conf for some time, and I had to remove it to
get the system to boot. If it's going to be broken going forward,
perhaps it's worthwhile to just remove the tunable?
- Nikolai Lifanov
More information about the svn-src-head
mailing list