Swap Memory Grows Over Time

We have an HA cluster of two servers. We are seeing the swap memory growing and never gets lower. From the time we turned it on about a week ago, it has worked its way up to 33% swap memory used. This is a new configuration and as far as we can tell, everything else is working fine. We are just concerned that this may be exposing an underlying problem.

FreePBX 13.0.195.4
High Availability Services 13.0.11

Any suggestions?

Observed something similar, just not with swap. It looks like the memory using was growing and never going down (32GB DDR4). @tm1000 mentioned a possible memory leak, but the strange thing was:

  1. No call quality issues
  2. No Swap being utilized (stayed at 1-2%, always)

When I would run htop on the command line, it showed more of what I expected <10% activity being utilized. I concluded that it must have something to do with how the dashboard is reporting vs what is actually being used.

I was going to open a issue ticket on this, but checked my other boxes and they didn’t seem to have this issue. I chalked it up to something happening to the install process and left it alone, since I could not replicate it.

Linux will quite happily use all your available swap space without any penalty, the problem comes when it has to ‘page out’ to the hard disk. I suggest you install sysstat if you think your hard disk is ‘thrashing’ it will give hints as to when the problem occurs, and the impact it has on other areas of the system

1 Like

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.