[DRBD-user] OCFS2 showing "No space left on device" on a device with free space

Brent Clark brentgclarklist at gmail.com
Tue Mar 8 14:23:48 CET 2016

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


Good day Guys

I apologies for posting here, but it looks like the OCFS mailing list it
dead. I am in desperate need of help and understanding, as per the subject,
and I was hoping is someone could assist (or maybe this thread could be of
future to someone on this list)

We just had an issue on one of our clients servers, whereby the server was
claiming there was no more space, but df showed there there plenty of space
(even checked the inodes).

We were really stummped, but after Googling around, the following command
was needed to be run tunefs.ocfs2 --fs-features=discontig-bg /dev/drbd0
(Link : https://www.novell.com/support/kb/doc.php?id=7008696)

Here is the output before tunefs.ocfs2 was run
user at server:/storage/www/site.com/current/sites/default/files/tmp$
<http://daddysdeals.co.za/current/sites/default/files/tmp$> df -h
Filesystem               Size  Used Avail Use% Mounted on
rootfs                    10G  302M  9.2G   4% /
udev                      10M     0   10M   0% /dev

(shorted to readability sake)

/dev/mapper/vg1-storage   99G  2.5G   91G   3% /storage
/dev/drbd0               100G   43G   58G  43% /storage/shared <= Look here.

The question I would like to ask. Is this a feature I can safely just roll
out, across all servers (I.e. run tunefs.ocfs2 --fs-features=discontig-bg
/dev/drbd0 on all servers )? If so, what are the complications and / or
risks.
Or should I just consider this a feature, and be glad this happened and let
things be.

The other question I would like to ask is,  generally we nagios the crap
out of servers.
This one slipped through the cracks, would anyone know for what flag or
variable I need to look out for.
I tried the suggested debugfs.ocfs2 -n -R "stats" /dev/drb0.

But I can't see what I need to check for (I compared debugfs.ocfs2 across
various servers.)
If I can catch this before some threshold or limit is about to be reached,
that would be great, for it looks like the standard df and dh -i is not
sufficient enough.

Thank in advance.
Regards

Brent Clark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20160308/0422f101/attachment.htm>


More information about the drbd-user mailing list