Jump to content

F5 Disk Space is Full


Cowboy Denny

Recommended Posts

If the /shared disk space is full, an alert is generated on CLI. Disk partition /shared has only 0% free

Delete the old image files (.iso) or pcap files (.pcap) saved in location - /shared/tmp/images , this frees up some space in /shared directory. 

To identify the disk space usage, run the command from the bash prompt 

df -h

To check files which occupy more space in /shared directory, execute the command                                                                                                                            

find /shared/ -xdev -type f -exec du {} \; | sort -rn | head -20

Examine the above output and determine files which occupy more space and delete the unused files using admin credentials.

rm <filename>
Link to comment
Share on other sites

If the /var disk space is full, an alert is generated on CLI. Disk partition /shared has only 0% free

 

To identify the disk space usage, run the command from the bash prompt 

df -h

First attempt at reducing the amount of /var space being used is...

Performing the following procedure, the BIG-IP REST API is temporarily inaccessible, and higher disk IO may be seen.

Run the following commands, in sequence:
# bigstart stop restjavad
# rm -rf /var/config/rest/storage*.zip
# rm -rf /var/config/rest/*.tmp
# bigstart start restjavad

You can also try running an F5 bash command to cleanup tmp directories

bash /usr/local/bin/clean_tmsh_tmp_dirs

If the above doesn't work then you have to try something else.

To check files which occupy more space in /var directory, execute the command                                                                              

find /var/ -xdev -type f -exec du {} \; | sort -rn | head -20

Examine the above output and determine files which occupy more space and delete the unused files using admin credentials.

rm <filename>

If APM is provisioned (sometimes even when its not provisioned) EPSEC packages could be filling up space on drive and in the UCS file.

Log in to tmsh by entering the following command:

tmsh

Identify if you have any EPSEC packages installed

list /apm epsec epsec-package recursive

Delete the epsec-package using the following command syntax:

delete /apm epsec epsec-package all

NOTE: the one that is in use doesn't go away, just the non-used packages

Note: It may take a few minutes for this command to reflect that the package deletion completed.

You can validate the package was deleted by entering the following command:

list /apm epsec epsec-package recursive

 

 

Link to comment
Share on other sites

You can also remove old/unused partitions

# tmsh show sys software status

-------------------------------------------------------------------------
Sys::Software Status
Volume  Slot  Product  Version   Build  Active    Status  Allowed Version
-------------------------------------------------------------------------
HD1.1      1   BIG-IP   15.1.7   0.0.6     yes  complete              yes
HD1.2      1   BIG-IP   15.1.5  0.0.10      no  complete              yes

Since HD1.2 isn't being used

# tmsh delete /sys software volume HD1.2

To check inode usage, run the df -i command.

# df -i
Filesystem                                        Inodes  IUsed   IFree IUse% Mounted on
/dev/mapper/vg--db--vda-set.1.root                112640   8081  104559    8% /
devtmpfs                                         2934640    459 2934181    1% /dev
tmpfs                                            2936621    246 2936375    1% /dev/shm
tmpfs                                            2936621    905 2935716    1% /run
tmpfs                                            2936621      2 2936619    1% /sys/fs/cgroup
/dev/mapper/vg--db--vda-set.1._usr                350880  89564  261316   26% /usr
/dev/mapper/vg--db--vda-set.1._var                196608 105372   91236   54% /var
none                                             2936621     80 2936541    1% /var/tmstat
prompt                                           2936621      8 2936613    1% /var/prompt
/dev/mapper/vg--db--vda-dat.share.1              2621440   1722 2619718    1% /shared
none                                             2936621     47 2936574    1% /shared/rrd.1.2
/dev/mapper/vg--db--vda-set.1._config             208000   5254  202746    3% /config
/dev/mapper/vg--db--vda-dat.log.1                 917504    464  917040    1% /var/log
none                                             2936621     12 2936609    1% /run/pamcache
none                                             2936621      1 2936620    1% /var/loipc
/dev/loop0                                             0      0       0     - /var/apm/mount/apmclients-7221.2022.412.1126-5816.0.iso
/dev/mapper/vg--db--vda-app.ASWADB.set.1.mysqldb  786432   3062  783370    1% /var/lib/mysql
/dev/mapper/vg--db--vda-app.avr.dat.avrdata       249984     10  249974    1% /shared/avr

If the inode usage is near or at 100 percent, move any unnecessary maintenance-related files from the BIG-IP system to a network share storage and schedule a time to reboot the BIG-IP system.

To locate directories that has many files that would contribute to inode usage, you can use the following find command to locate the 20 largest directories that have most files in the /var partition:

find /var -xdev -printf '%h\n' | sort | uniq -c | sort -k 1 -nr | head -20

You should delete old tcpdumpqkview, and core files from the system or move them to a network share.

 

 

Check for deleted files still using disk space.

# lsof -ws | grep -i 'size\|deleted'

You can ignore anything in memory since it isn't taking up any disk space which would include

  • /var/tmstat
  • /proc
  • tmpfs
  • hugetlbfs

 

Link to comment
Share on other sites



×
×
  • Create New...