CMS Full - what now?

[cwjr]

At one of our centers, we have an CMS R3V6 system. Avaya told us last year that they would no longer support it. We took a gamble, hoping to make it through this budget year before upgrading. Well, we rolled the dice and they came up "snake eyes"...

We are now getting an error message that says "Warning Warning Warning. The file system is 97% full. You must clean up the file system immediately...." etc, etc.

Does anyone know how to "clean up the file system"?

 

[4merAvaya]

First thing is to do a reboot. Log in as 'root' and type '/etc/shutdown -y -g0 -i6'. This wil bring the CMS down gracefully and reboot. There are files that grow over time, such as /etc/tmp, that get cleaned out. The second thing to do is check the Data Storage Allocation in the CMS menu. You may have to reduce the storage intervals.

Kevin

 

[cwjr]

Thanks for the quick response, Kevin. We'll give that a try. By the way, I said R3V6, but it is actually an R3V8.

Also, as far as changing the allocation, what is the best thing to "reallocate" to get the space we need?

 

[ezncool]

If your space is trully full and not just the allocation, you may want into looking if you still need to measure your trunks. If not, that in my opinion would be the first one ill take out.



ezncool
I have no technical solution to your management problem

 

[cwjr]

Reboot didn't help.

When I do a df command at the Unix prompt, it looks like there is several gig available to the cms volume.

Question: When I go into the "Data Storage Allocation" page, if I change, for intance "Call Records" to a lower number, does that "free up" used space, or does that reduce the available space? I guess I'm just trying to understand how/what changes need to be made and in "what direction" to free up the space that CMS is claiming is nearly full.

I hope that makes sense, forgive me if it doesn't, and I'll try to rephrase...

 

[f0netech]

You might also check to see if 'trace' data is turned on for any agents. This is data that should be used sparingly as it can also bog down the performance of the machine but easily forgotten.

 

[portcity]

If it's the / (root) partition which is 80% full it could be mail file or a log file The favorites are /var/mail/root or /var/adm/messages

The /var/mail/root file can just be deleted however if the mails relate to a an issue you should deal with this otherwise it will grow again. The /var/adm/messages file should be looked at first and again deal with any issues then it can be purged with echo>/var/adm/messages don't just delete it as you will end up with an unreferenced file. If it's not these it will be like looking for a needle in a haystack however du -k |pg will list all files and there usage, what you are looking for are large files or a large number of files which seem to be out of place.

Any Unix server should have a scheduled reboot. Depending upon number of agents and simultaneous supervisor logins 1-3 months. I would just remove the entire mail directory. Who uses CMS mail.

A second suggestion is to remove core dumps. These are extremly large files. login as root and at the root directory type: find . -name core -print

then remove the core files