[Gluster-devel] Slow fs

[Emmanuel Dreyfus]

Vijay Bellur <vbellur at redhat.com> wrote:

> With lookups taking bulk of the time, was there any self heal in progress?

It seems so. glusterfshd.log loops on this on one server:
[2013-11-15 04:48:51.995536] I
[afr-self-heald.c:1180:afr_dir_exclusive_crawl] 0-gfs341-replicate-0:
Another crawl is in progress for gfs341-client-0

The two other ones have many things like this:
[2013-11-15 05:13:23.819290] W
[client-rpc-fops.c:1538:client3_3_inodelk_cbk] 0-gfs341-client-1: remote
operation failed: No such file or directory
[2013-11-15 05:13:23.857381] W
[client-rpc-fops.c:1640:client3_3_entrylk_cbk] 0-gfs341-client-1: remote
operation failed: No such file or directory
[2013-11-15 05:13:23.857664] E
[afr-self-heal-entry.c:2296:afr_sh_post_nonblocking_entry_cbk]
0-gfs341-replicate-0: Non Blocking entrylks failed for
<gfid:46274798-b1b2-4db7-9fd1-45c158e2d201>.
[2013-11-15 05:13:23.957235] W
[client-rpc-fops.c:1538:client3_3_inodelk_cbk] 0-gfs341-client-1: remote
operation failed: No such file or directory

> Did the situation recover after a while or has it been the same?

No, it is still crawling after several days.

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu at netbsd.org