UNSOLVED

petra-MPt6M

updated

12 years ago

0

6638

November 20th, 2014 00:00

/nbsnas getting full

I have Celerra NS704 running 5.5.21-4.

I cannot use several commands, I guess because of /nbsnas partition getting full


[root@xxxx var]# df

Filesystem 1k-blocks      Used Available Use% Mounted on

/dev/hda3 2063536   1648348    310364  85% /

/dev/hda1 31079      2749     26726 10% /boot

none 257540         0 257540   0% /dev/shm

/dev/nde1 1818352   1818352         0 100% /nbsnas

/dev/hda5 2063504   1625120    333564  83% /nas

/dev/nda1 136368     60780     75588  45% /nas/dos

/dev/ndf1 1818352    833324    892656  49% /nas/var

[nasadmin@XXX nasadmin]$ server_export server_2
server_2 :
Error 2237: server_2 : Execution failed: No space left on device: Routine failure. [FILE.file_close]

nas_cmd: PANIC: caught signal #11 (Segmentation violation) -- Giving up...

After running /nas/sbin/js_cleandb, issue seemed to be resolved as I was able to run the command. But after few minutes partition was full again. I also checked /nbsnas/var/dump but it is empty.

What else can I delete?

  • nandas

    6 Operator

    •

    1473 Posts

    3244

    1

    Posted November 23rd, 2014 22:00

    Hi Petra,

    It seems you have not got any response to this post, thus trying to help. Is the issue still persists? You may access the NAS system folder through the directory path on the slot, if that is known to you. If not, take help from Support.

    You may consider deleting old LOG files or Core dump files if any, which are typically big in size.

    Again, I expect the issue is resolved by now.

    Thanks,
    Sandip

  • SuhelaDighe

    1 Rookie

    •

    70 Posts

    3244

    1

    Posted November 24th, 2014 02:00

    I will move this question into the Celerra support forum, so you get a faster response.

  • nandas

    6 Operator

    •

    1473 Posts

    3244

    0

    Posted November 24th, 2014 21:00

    Thanks Suhela. Will wait to hear from Petra.

  • 3245

    0

    Posted December 1st, 2014 07:00

    Thanks, I was able to get help directly from EMC, but basically there were several debug core dumps created regularly which were filling up the filesystem. We had to delete them for serveral times in order to get rid of them.

  • 3245

    0

    Posted January 29th, 2015 14:00

    Hi,

    I also have this issue and need to delete something to free up the space in /dev/hda3.

    But I can't find anything to delete.

    Can you please advise where are the dump files that you deleted.

    Thanks

    Geoff Barnes

    ------------------------------------------------------------------------------------------------------------------------------------

    Error Message:   dskMon[5044]: FS /dev/hda3 mounted on / filling up (at 92%, max = 90%).

    [nasadmin@tch_cs0 ~]$ df

    Filesystem           1K-blocks      Used Available Use% Mounted on

    /dev/hda3              2063536   1800404    158308  92% /

    ------------------------------------------------------------------------------------------------------------------------------------

  • umichklewis

    4 Apprentice

    •

    1190 Posts

    3245

    0

    Posted January 30th, 2015 06:00

    You'll need to try and find the source of your buildup.  su to root and try running this command:

    [root@test-vnxcs0 / ]# cd /

    [root@test-vnxcs0 / ]#  du -h --max-depth=1|sort -n

    This will dump the root of the directories and give a listing of file similar to this:

    [root@celerra1 /]#  du -h --max-depth=1|sort -n

    du: `./nasmcd/rootfs/slot_2': No such file or directory

    du: `./nasmcd/quota/slot_2': No such file or directory

    0       ./proc

    0       ./sys

    1.2G    ./nas

    1.6G    ./nbsnas

    4.0K    ./i386

    4.0K    ./misc

    4.0K    ./selinux

    4.0K    ./srv

    4.4G    .

    5.3M    ./boot

    6.5M    ./bin

    8.0K    ./mnt

    16K     ./lost+found

    16K     ./media

    22M     ./sbin

    35M     ./etc

    37M     ./lib

    38M     ./opt

    59M     ./nasmcd

    60K     ./root

    96K     ./dev

    104K    ./tmp

    136M    ./var

    168K    ./home

    633M    ./celerra

    692M    ./usr

    /nbsnas is not mounted on /dev/hda3, neither is /nas.  However, /celerra is mounted on / - you'll probably want to start in the /celerra directory and see if you have old dump files, audit files or such.

  • Rainer_EMC

    6 Operator

    •

    8645 Posts

    3245

    0

    Posted January 30th, 2015 08:00

    Could be core dumps, old OS updates, old logs, …

    There is a knowlegbase article about cleaning up

    If not engage then support – this is normal business for them

  • jithin

    91 Posts

    3246

    0

    Posted February 18th, 2015 05:00

    Please refer the EMC knowledge base article https://support.emc.com/kb/73359

    Jithin