[lustre-discuss] MGTMDT device getting full

[Torsten Harenberg]

Thanks Chris & Andreas,

Am 16.10.15 um 19:50 schrieb Christopher J. Morrone:
> If that doesn't show anything, you probably need to mount your MDT's
> backend filesystem as a local filesytem (readonly) and look for where
> the space is going.

did as advised:

[root at lustre1 MGTMDT]# ls -l
total 581151336
-rw-r--r--  1 root root          144 Aug 13  2013 capa_keys
-rwx------  1 root root          480 Aug 13  2013 CATALOGS
-rw-r--r--  1 root root            0 Jul 21 11:42 changelog_catalog
-rw-r--r--  1 root root            0 Jul 21 11:42 changelog_users
drwxrwxrwx  2 root root         4096 Jul  6 13:51 CONFIGS
-rw-r--r--  1 root root         8192 Aug 13  2013 fld
-rw-r--r--  1 root root            0 Jul 21 11:42 hsm_actions
-rw-r--r--  1 root root        47104 Aug 13  2013 last_rcvd
-rw-r--r--  1 root root           64 Jul 21 11:42 lfsck_bookmark
-rw-r--r--  1 root root         8192 Jul 21 11:42 lfsck_namespace
drwx------  2 root root        16384 Aug 13  2013 lost+found
-rw-r--r--  1 root root          120 Aug 13  2013 lov_objid
-rw-r--r--  1 root root          120 Jul 21 11:41 lov_objseq
-rw-r--r--  1 root root        16384 Jul 21 08:04 lquota_v2.group
-rw-r--r--  1 root root        50176 Jul 21 08:04 lquota_v2.user
drwxr-xr-x  2 root root         4096 Jul 21 11:41 NIDTBL_VERSIONS
drwxr-xr-x  6 root root         4096 Jul 21 11:41 O
drwxrwxrwx  2 root root       311296 Jul 21 05:16 OBJECTS
-rw-r--r--  1 root root 594516746240 Aug 13  2013 oi.16
-rw-r--r--  1 root root          400 Jul 21 11:41 OI_scrub
drwxr-xr-x  2 root root       520192 Aug 13  2013 PENDING
drwxr-xr-x  4 root root         4096 Jul 21 11:41 quota_master
drwxr-xr-x  2 root root         4096 Jul 21 11:42 quota_slave
drwxr-xr-x  2 root root         4096 Aug 13  2013 REM_OBJ_DIR
drwxr-xr-x  2 root root         4096 Jul 21 11:41 REMOTE_PARENT_DIR
drwxr-xr-x 83 root root        12288 Oct 15 09:13 ROOT
-rw-r--r--  1 root root           24 Aug 13  2013 seq_ctl
-rw-r--r--  1 root root           24 Aug 13  2013 seq_srv
[root at lustre1 MGTMDT]# df .
Filesystem         1K-blocks      Used Available Use% Mounted on
/dev/mapper/MGTMDT 805164976 654082044  97395844  88% /srv/MGTMDT
[root at lustre1 MGTMDT]# ls -lh oi.16
-rw-r--r-- 1 root root 554G Aug 13  2013 oi.16
[root at lustre1 MGTMDT]# df -h .
Filesystem          Size  Used Avail Use% Mounted on
/dev/mapper/MGTMDT  768G  624G   93G  88% /srv/MGTMDT


There is the one large file oi.16 and it looks like

https://jira.hpdd.intel.com/browse/LU-1512

If I read that correctly, only Lustre-2.1 is still affected (which is
the version we were running until summer).

I couldn't get the clue if a fsck would help now or not. Andreas
Dillinger wrote about 2 years ago:

"""I'm not sure why this bug was closed. The patch for b2_1 was still
not landed, and the work to rebuild the OI files in LFSCK Phase 4 is not
completed."""

But I couldn't get if that statement is only for Lustre-2.1 or general.

Cheers

 Torsten



-- 
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
<>                                                              <>
<> Dr. Torsten Harenberg     Torsten.Harenberg at cern.ch          <>
<> Bergische Universitaet                                       <>
<> FB C - Physik             Tel.: +49 (0)202 439-3521          <>
<> Gaussstr. 20              Fax : +49 (0)202 439-2811          <>
<> 42097 Wuppertal           @CERN: Bat. 1-1-049                <>
<>                                                              <>
<><><><><><><>< Of course it runs NetBSD http://www.netbsd.org ><>