There was an issue sometime back that we ran into which involved log clear
during failover not always flushing the llogs completely and thus leaving
orphaned entries in defunct logs. This manifested itself by reporting no
available catalog slots error. However in this case, I would highly
recommend
I am no sure if you hit the same bug as in our case: the llog was not
cleared several times and filed the whole mdt space, but the upgrade from
2.8.x to 2.9 resolved the log clear problem.
Am Fr., 25. Jan. 2019, 07:16 hat Colin Faber geschrieben:
> Have you tried manually purging the changelo
Have you tried manually purging the changelog files and catalog then
restarting by re-registering? Also, are you sure that _all_ consumers are
requesting to clear the records?
On Mon, Jan 7, 2019 at 11:40 AM nan...@luis.uni-hannover.de <
nan...@luis.uni-hannover.de> wrote:
>
> Any advice here wha
out to me?
:Andy
From: Gibbins, Faye [mailto:faye.gibb...@cirrus.com]
Sent: Friday, June 02, 2017 2:44
To: Andy Moe mailto:m...@cray.com>>;
lustre-discuss@lists.lustre.org<mailto:lustre-discuss@lists.lustre.org>
Subject: RE: [lustre-discuss] Changelog users failing to clear records in 2.8,
me?
:Andy
From: Gibbins, Faye [mailto:faye.gibb...@cirrus.com]
Sent: Friday, June 02, 2017 2:44
To: Andy Moe ; lustre-discuss@lists.lustre.org
Subject: RE: [lustre-discuss] Changelog users failing to clear records in 2.8,
can anyone help?
Hi Andy,
Yes indeed!
There was parallel reading and cleari
...@cray.com]
Sent: 01 June 2017 19:34
To: lustre-discuss@lists.lustre.org
Cc: IT Software Systems - All
Subject: RE: [lustre-discuss] Changelog users failing to clear records in 2.8,
can anyone help?
Faye,
Up to the point when you’ve experienced problems clearing records, had at any
point Changelogs
- All ;
lustre-discuss@lists.lustre.org
Subject: Re: [lustre-discuss] Changelog users failing to clear records in 2.8,
can anyone help?
There seems to have been a few instances of this reported here on the list in
the last few months, I don't recall the earlier versions of lustre, but we
There seems to have been a few instances of this reported here on the list
in the last few months, I don't recall the earlier versions of lustre, but
we have also seen this in the wild for customer systems, so very likely a
bug which results in corruption of llog files.
-cf
On Thu, Jun 1, 2017 a
On Jun 1, 2017, at 10:55, Faccini, Bruno wrote:
>
> Hello,
> According to the error msgs, looks like there is a corrupted plain-LLOG file
> for the ChangeLogs of MDT0. And unfortunately, neither e2fsck nor lfsck can
> help to recover in this case.
Bruno,
is this bug fixed in newer Lustre relea
Hello,
According to the error msgs, looks like there is a corrupted plain-LLOG file
for the ChangeLogs of MDT0. And unfortunately, neither e2fsck nor lfsck can
help to recover in this case.
I think that to clear this situation you need to stop/umount this MDT and
re-mount it as ldiskfs to move b
Hi,
We have 4 file systems on our lustre cluster. All have changelog users
registered for robinhood to use.
We have discovered that a changelog user for one of the file systems is not
catching up to its index. Manual runs of Robinhood fail to read any more
records even though according to mdd/
11 matches
Mail list logo
