[Users] vmstat and floating point exception

Mark Johanson mjohanson at a2hosting.com
Tue Jul 10 19:50:25 EDT 2012


Linux solusvm.a2hosting.com 2.6.32-042stab053.5 #1 SMP Tue Mar 27 
vzctl version 3.3
vswap is enabled

Since upgrading to a 2.6.32 kernel one of our users noticed that they
are occasionally getting a Floating Point Exception when running vmstat:

[root at server ~]# vmstat 1 5
procs -----------memory---------- ---swap-- -----io---- --system--
-----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy
id wa st
 0  0      0 2048244      0  32816    0    0     0     2    0 7627  0  0
99  0  0
 0  0      0 2048236      0  32816    0    0     0     0    0  110  0  0
100  0  0
 0  0      0 2048236      0  32816    0    0     0     0    0   78  0  0
100  0  0
 0  0      0 2048236      0  32816    0    0     0     0    0   91  0  0
100  0  0
Floating point exception

It doesn't happen every time as you can see here:

[root at server ~]# vmstat 1 10
procs -----------memory---------- ---swap-- -----io---- --system--
-----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy
id wa st
 0  0      0 2068584      0  12472    0    0     0     0    0 1745  0  0
100  0  0
 0  0      0 2068576      0  12472    0    0     0     0    0   82  0  0
100  0  0
 0  0      0 2068576      0  12472    0    0     0     0    0   77  0  0
100  0  0
 0  0      0 2068576      0  12472    0    0     0     0    0  109  0  0
100  0  0
 0  0      0 2068576      0  12472    0    0     0     0    0   92  0  0
100  0  0
 0  0      0 2068576      0  12472    0    0     0     0    0   80  0  0
100  0  0
 0  0      0 2068576      0  12472    0    0     0     0    0   87  0  0
100  0  0
 0  0      0 2068576      0  12472    0    0     0     0    0   66  0  0
100  0  0
 0  0      0 2068588      0  12472    0    0     0     0    0  130  0  0
100  0  0
 0  0      0 2068588      0  12472    0    0     0     0    0   72  0  0
100  0  0

I have been able to replicate this issue in other OS templates (theirs
is Ubuntu 11.10, but I have replicated in Ubuntu, Fedora, Centos,
etc...) running the same kernel and vzctl version. 

I can not replicate this issue on the node.

Wondering if anyone else has run across this issue and/or knows what
might be causing it.



More information about the Users mailing list