Skip to contentSkip to conversation

last xfs_repair time?

DISCUSSION Lars Damerow 5 participants 8 messages Nov 12, 2008 – Nov 13, 2008

lore
Share & tools
lore
#0 lore raw

Hi folks,

Is it possible to query XFS for the last time xfs_repair fixed a filesystem? I've been searching the man pages and haven't found anything useful.

thanks!
-lars

--
lars r. damerow :: button pusher :: pixar animation studios <[email protected]>

#1 lore raw
Quoted 6 lines · Lars Damerow

Hi folks,

Is it possible to query XFS for the last time xfs_repair fixed a filesystem? I've been searching the man pages and haven't found anything useful.

Quoted 7 lines

thanks!
-lars

--
lars r. damerow :: button pusher :: pixar animation studios <[email protected]>

Not that I know of.
Barry is the repair maintainer/expert though.

--Tim

#3 lore raw
In reply to Eric Sandeen
Quoted 9 lines

From Eric Sandeen <[email protected]>, Wed, Nov 12, 2008 at 07:11:21PM -0600:

Quoted 7 lines · Timothy Shimmin
Quoted 6 lines · Lars Damerow

Hi folks,

Is it possible to query XFS for the last time xfs_repair fixed a filesystem? I've been searching the man pages and haven't found anything useful.

Quoted 8 lines
Quoted 6 lines

Not that I know of.
Barry is the repair maintainer/expert though.

--Tim

AFAIK it doesn't, but that'd be a neat feature - adding space for that, and maybe date of mkfs (or mkfs version) etc might be nice too.

Thanks for the responses! It's handy to have that information available, especially when you're maintaining hundreds of workstations whose power cables occasionally get yanked out by their users. :/

Here's what we can get from ext3:

    % sudo dumpe2fs -h /dev/sda1 | grep 2008
    dumpe2fs 1.40.8 (13-Mar-2008)
    Inodes per group:         2008
    Filesystem created:       Thu May 29 11:11:12 2008
    Last mount time:          Tue Sep 30 18:01:18 2008
    Last write time:          Tue Sep 30 18:01:18 2008
    Last checked:             Thu May 29 11:11:12 2008

cheers,
-lars

--
lars r. damerow :: button pusher :: pixar animation studios <[email protected]>

#2 lore raw
In reply to Timothy Shimmin
Quoted 10 lines · Timothy Shimmin
Quoted 9 lines · Lars Damerow

Hi folks,

Is it possible to query XFS for the last time xfs_repair fixed a filesystem? I've been searching the man pages and haven't found anything useful.

thanks!
-lars

Quoted 10 lines
Quoted 4 lines

--
lars r. damerow :: button pusher :: pixar animation studios <[email protected]>

Not that I know of.
Barry is the repair maintainer/expert though.

--Tim

AFAIK it doesn't, but that'd be a neat feature - adding space for that, and maybe date of mkfs (or mkfs version) etc might be nice too.

-Eric

#4 lore raw
In reply to Lars Damerow
Quoted 15 lines · Lars Damerow

From Eric Sandeen <[email protected]>, Wed, Nov 12, 2008 at 07:11:21PM -0600:

Quoted 14 lines
Quoted 13 lines · Timothy Shimmin
Quoted 6 lines · Lars Damerow

Hi folks,

Is it possible to query XFS for the last time xfs_repair fixed a filesystem? I've been searching the man pages and haven't found anything useful.

Not that I know of.
Barry is the repair maintainer/expert though.

--Tim

Quoted 6 lines
Quoted 2 lines

AFAIK it doesn't, but that'd be a neat feature - adding space for that, and maybe date of mkfs (or mkfs version) etc might be nice too.

Thanks for the responses! It's handy to have that information available, especially when you're maintaining hundreds of workstations whose power cables occasionally get yanked out by their users. :/

So, for what it's worth, power loss should not necessarily require a repair; as long as you have barriers enabled and/or you're not losing log writes to a volatile write cache, power loss should never corrupt the filesystem metadata; that is what the log is for, after all...

-Eric

#7 lore raw
In reply to Eric Sandeen
Quoted 18 lines · Eric Sandeen
Quoted 17 lines · Timothy Shimmin
Quoted 6 lines · Lars Damerow

Hi folks,

Is it possible to query XFS for the last time xfs_repair fixed a filesystem? I've been searching the man pages and haven't found anything useful.

Quoted 7 lines

thanks!
-lars

--
lars r. damerow :: button pusher :: pixar animation studios <[email protected]>

Not that I know of.
Barry is the repair maintainer/expert though.

Quoted 7 lines
Quoted 3 lines

--Tim

AFAIK it doesn't, but that'd be a neat feature - adding space for that, and maybe date of mkfs (or mkfs version) etc might be nice too.

-Eric

That would sort of fall under the last repair logs still on the todo list.

#5 lore raw
In reply to Eric Sandeen
Quoted 5 lines

From Eric Sandeen <[email protected]>, Wed, Nov 12, 2008 at 07:33:25PM -0600:
So, for what it's worth, power loss should not necessarily require a repair; as long as you have barriers enabled and/or you're not losing log writes to a volatile write cache, power loss should never corrupt the filesystem metadata; that is what the log is for, after all...

Our disks are partitioned with LVM, so we don't have write barriers. We just recently disabled write caching, though, so our future should be brighter than it's been. :)

-lars

--
lars r. damerow :: button pusher :: pixar animation studios <[email protected]>

#6 lore raw

Hi!

Lets see whether MailMan lets GPG signature unmangled or whether it is something else that mangles it.

Am Donnerstag 13 November 2008 schrieb Lars Damerow:

Quoted 1 line

From Eric Sandeen <[email protected]>, Wed, Nov 12, 2008 at

07:33:25PM -0600:

Quoted 8 lines
Quoted 4 lines

So, for what it's worth, power loss should not necessarily require a repair; as long as you have barriers enabled and/or you're not losing log writes to a volatile write cache, power loss should never corrupt the filesystem metadata; that is what the log is for, after all...

Our disks are partitioned with LVM, so we don't have write barriers. We just recently disabled write caching, though, so our future should be brighter than it's been. :)

I really hope that LVM gets write barrier support for write barrier capable targets. Unless it does my notebook will use plain partitions for anything but testing filesystems. Otherwise with 2.6.16 back then I hardly felt a difference in XFS speed after turning off write caches.

If you like add your vote here:
http://bugzilla.kernel.org/show_bug.cgi?id=9554

Ciao,

Signature · 2 lines
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA  B82F 991B EAAC A599 84C7

_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs