I'm stumped by this since it's the first time I've noticed this behavior under VMware. FreeBSD 8.4-p23, ESXi 5.0.0 (474610).
I started getting nagios alerts that swap usage on one of my
This is a more minor example that I've captured. Earlier that day I saw swap at 75%, still with over 3GB of RAM free. I noticed that some of the
The question is, why?
What happens if I don't take care of it is that swap reaches 100% and then the kernel starts killing things off:
At the time this happens, there are multiple GB of RAM free.
Is this the VMware tools doing something funny with RAM? Docs on FreeBSD + VMware are really hard to come by and usually outdated.
I started getting nagios alerts that swap usage on one of my
spamd/ clamd boxes was exceeding the threshold. Not too alarming, both spamd and clamd are memory pigs. But what I've been seeing is things like this:
Code:
Mem: 3623M Active, 526M Inact, 534M Wired, 194M Cache, 558M Buf, 3016M Free
Swap: 1024M Total, 174M Used, 850M Free, 16% Inuse
...
avail memory = 8240508928 (7858 MB)This is a more minor example that I've captured. Earlier that day I saw swap at 75%, still with over 3GB of RAM free. I noticed that some of the
spamd processes showed 0K resident, so I assume that's what's using my swap. The question is, why?
What happens if I don't take care of it is that swap reaches 100% and then the kernel starts killing things off:
Code:
swap_pager_getswapspace(16): failed
swap_pager_getswapspace(16): failed
pid 1267 (clamd), uid 106, was killed: out of swap spaceAt the time this happens, there are multiple GB of RAM free.
Is this the VMware tools doing something funny with RAM? Docs on FreeBSD + VMware are really hard to come by and usually outdated.