CUCM and UCCX : Disk partition Swap is nearly full. Usage is 92%

[Hermanus Janse van Vuuren]

Hi All,

We are on CUCM 12.5 and CCX 12.5, from time too time we get the swap disk partition going over 90% and gets stuck there.

Currently we resolve the issue by rebooting the affected server.

I have been reading up since the customer started asking questions on why the restarts from time too time and all I can see is either to open a TAC case to on/off the swap via root account or by rebooting the server,

Is there no other fix for this issue yet? It is something coming from CUCM 6.0 when CUCM moved from windows to linux and we still have no other way of clearing the swap partition?

Can one restart tomcat services or something rather than rebooting the server? In RTMT there is not really and process spiking or using the CPU when this happens.

So basically we are stuck with either change control and a reboot or a TAC case which is wasting TAC's time and our time? 

[Nithin Eluvathingal]

Have a look on below thread

https://community.cisco.com/t5/ip-telephony-and-phones/cucm-question-is-it-normal-that-swap-partition-usage-is-78/td-p/1061535

[Hermanus Janse van Vuuren]

I have read that before posting, my question is because we have monitoring and an alert gets generated every time this goes over 90% and is logged in the ticketing system, the client has a view of this. Thus the question, I know it is not a real big problem, but do we have a different way to release the swap "memory" without rebooting or using TAC to get the value below 90% if it gets stuck there for a long period, I know it comes down after a while but we have had cases where it stays at 98% and won't come down until we reboot the server.

No CLI command we can run without root privileges to clear the swap partition?

[Phig Newton]

Hermanus,

I have the same situation.  Did you find any good solutions, or alternates to rebooting?  thanks!

[abdul ilyas · Accepted solution]

I have this problem too. UCCX version 12.5(1) SU1 ES03.

Rebooted server just 19 days ago to clear swap, now it is back up at 96%. It goes down a little bit ~92% then jump back to 96%+.

Last reboot it was showing 100%. Anyone able to find a solution ?

[stuart.pannell]

I also have this problem on one of my IM&P servers, is there a way to resolve this without rebooting the server? This post says resolved and 30 people found that rebooting the server was helpful? 

I would like to know a fix and not a workaround.

[Phig Newton]

Just guessing, I bet there is a service that can be restarted.  Or maybe changing the threshold would work. 

I wonder where I can find that threshold?

[abdul ilyas]

As far as I researched there is no service to restart but Cisco TAC can run some Linux internal commands to clear swap cache. 

I heard increasing system RAM will also help if you're frequently running into high swap usage. 

Hope it helps.

[Phig Newton]

Oh yeah, RAM.  That sounds like it would work.