CS1K Signaling Server partition full /dev/hda10 (Not the normal /var issue). Directory not working.

[John Tel]

We have a customer with CS1K on 7.65 that has 2 cores located in separate buildings where both Signaling servers are on CPPM cards. They have lost directory button use on their IP phones again. This has happened in the past and a freeDiskSpace followed by reboot for both Signaling servers has allowed the clearing of the /dev/hda5/var and /dev/hda4/admin folders along with some JBoss clearing that is available with the command. This drive space reduction lessens these partitions and allows for directory to function again as well as EM. This has been completed again but have found that /dev/hda10 is at 100% full capacity on SS1. SS0 is at 65%. With this on SS1 being full leads me to think that this is the issue now and not anything to do with dev/hda5/var as can be found when doing a search directory issues. I have went into the /dev/hda5/var folders in the past and deleted logs manually but have no idea how to clean up the /dev/hda10 partition as it has no name like the other partitions do. I am still learning Linux and have very limited access to the system which doesn't help matters. Any help or direction is greatly appreciated!

 

[Firebird Scrambler]

I can only find a reference to find the folder below.

Commands to show Sig / LINUX Server disc and memory size.

0;admin@moc-smg-flwrss:~[admin@moc-smg-flwrss ~]$ diskSizeShow (or just df )
Filesystem Size Used Avail Use% Mounted on
/dev/hda11 2.0G 1.3G 570M 70% /
/dev/hda8 3.8G 1.2G 2.5G 32% /opt
/dev/hda9 3.8G 74M 3.6G 2% /home
/dev/hda2 5.6G 1.1G 4.2G 21% /admin
/dev/hda6 5.7G 140M 5.3G 3% /tmp
/dev/hda5 7.4G 2.6G 4.4G 38% /var
/dev/hda10 1.9G 35M 1.8G 2% /var/coredump
/dev/hda7 3.8G 146M 3.5G 4% /var/log
/dev/hda1 93M 51M 37M 58% /boot
tmpfs 1008M 16K 1008M 1% /dev/shm

[802176699@moc-ldr ~]$ memSizeShow
Total server memory: 2048 MB

 

[John Tel]

Using df will give similar output as what you have shown. Your hda11 (/dev/hda11 2.0G 1.3G 570M 70% /) looks like my hda10 where it doesn't show a folder name but mine is 100% full.

I have the below information that i was able to record manually using df. I can reach out to the site to get the filenames or further information if that helps any.

/dev/hda10 100% (***NO Filename)
/dev/hda9 36% (has filename but i did not record while onsite)
/dev/hda8 34% (has filename but i did not record while onsite)
/dev/hda2 18% (has filename but i did not record while onsite)
/dev/hda6 15% (has filename but i did not record while onsite)
/dev/hda5 48% (has filename but i did not record while onsite)
/dev/hda11 2% (has filename but i did not record while onsite)
/dev/hda7 22% (has filename but i did not record while onsite)
/dev/hda1 42% (has filename but i did not record while onsite)
tmpfs 0% (has filename but i did not record while onsite)

 

[gbtac3]

We have a customer with CS1K on 7.65 that has 2 cores located in separate buildings where both Signaling servers are on CPPM cards. They have lost directory button use on their IP phones again. This has happened in the past and a freeDiskSpace followed by reboot for both Signaling servers has allowed the clearing of the /dev/hda5/var and /dev/hda4/admin folders along with some JBoss clearing that is available with the command. This drive space reduction lessens these partitions and allows for directory to function again as well as EM. This has been completed again but have found that /dev/hda10 is at 100% full capacity on SS1. SS0 is at 65%. With this on SS1 being full leads me to think that this is the issue now and not anything to do with dev/hda5/var as can be found when doing a search directory issues. I have went into the /dev/hda5/var folders in the past and deleted logs manually but have no idea how to clean up the /dev/hda10 partition as it has no name like the other partitions do. I am still learning Linux and have very limited access to the system which doesn't help matters. Any help or direction is greatly appreciated!

What I have done to clean this up manually, is to go into the individual folders, (var/log, home and admin), and delete unnecessary files. These may include old log files and sometimes install and log files that were copied to home and admin for some other reason, (easier to transfer using WinSCP when it is in these other folders).

 

[John Tel]

Thanks, gbtac3. I have cleaned the /var, /home and /admin using freeDiskSpace as well as going into these directories to further clean. I've followed Avaya's PSN# PSN003828u to clean these also. I haven't given WinSCP a thought but I've used it on IPO with success.

I have found that the directory that is 100% full is actually the root directory. I am trying to find out why the root directory would be full and what caused it to get full. I'm hoping someone has some insight on how to clean this without risking breaking the phone system as i found the kernel and boot process resides here.