[CentOS] BUG: soft lockup - CPU#0 stuck for 36s! [swapper/0:0]

2016-08-18 13:32 GMT-04:00 JJB <jack at internetguy.net>:

>
> I'm also seeing those errors in several servers, running under 5.5.
>> Currently investigating if this
>> <https://kb.vmware.com/selfservice/microsites/search.do?
>> language=en_US&cmd=displayKC&externalId=1009996>
>> has anything to do (the resource overcommit bit).
>>
>
> Does this happen (only) while taking or consolidating snapshots? The VM is
> suspended during these operations and the OS isn't too crazy about it,
> especially if you have slow storage.
>

Nope, no snapshots. Just plain running. In fact, many times the guests are
under light usage (internal instrumentation, no external VMware stats).
We're investigating because we do have reasons to believe that our provider
is probably overcommitting or overselling (not out of malice, AFAIK).

HTH,
Carlos.