On Mon, 2013-03-04 at 10:00 +0100, Daniele Melosi wrote: > Hi, > > after kernel upgrade i've the following situation: > > after 2 days of uptime: > 08:41:41 up 2 days, 14:56, 0 users, load average: 13.07, 13.83, 13.65 > size-192(DMA) 0 0 192 20 1 : tunables 120 60 > 8 : slabdata 0 0 0 > size-192 35599 83060 192 20 1 : tunables 120 60 > 8 : slabdata 4153 4153 273 > > after 6 days of uptime: > 08:41:41 up 6 days, 1:56, 0 users, load average: 16.60, 15.22, 14.48 > size-192(DMA) 0 0 192 20 1 : tunables 120 60 > 8 : slabdata 0 0 0 > size-192 1780288 1799240 192 20 1 : tunables 120 60 > 8 : slabdata 89962 89962 256 > > As you can see the size-192 is increasing and after the 7th days the > server start swapping. > > The server is a shared hosting linux with apache + cgi-wrapper and has a > lots of entries in /proc/mounts: > > # wc -l /proc/mounts > 329219 /proc/mounts [...] Er, wow, that's a lot. Does this *also* increase over time, or are those mostly mounted at boot? Ben. -- Ben Hutchings Always try to do things in chronological order; it's less confusing that way.
Attachment:
signature.asc
Description: This is a digitally signed message part