Re: [Xen-devel] fix for bug 651?

>changeset 10546
>(<http://xenbits.xensource.com/xen-unstable.hg?cs=3D8242c0c24db7>)
>looked promising as a fix for bug #651
><http://bugzilla.xensource.com/bugzilla/show_bug.cgi?id=3D651>, but I
>can still trigger the OOM killer.   I get a kernel panic out of memory
>error when I do a mem-set of dom0 to 512 on a 32GB system and mem-set of
>dom0 to 256 on 8GB system. 

Cset 10546 just makes it less likely - you'll pretty much always be 
able to trigger oom on linux if you reduce the memory by enough. I 
can't imagine any magic to fix this unfortunately :-)

Best practice is probably simply to empirically determine an approximate 
minimum phys mem and swap setup for the workload/machine you have, and 
then stick above that. 

cheers,

S.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel

<Prev in Thread]

Current Thread

[Next in Thread>

Previous by Date:

Re: [Xen-devel] [RFC] [PATCH] HVM SMBIOS support 1/6, Keir Fraser

Next by Date:

[Xen-devel] difference between alloc_vm_area() and balloon_alloc_empty_page_range(), Lily Huang

Previous by Thread:

[Xen-devel] fix for bug 651?, Krysan, Susan

Next by Thread:

Re: [Xen-devel] Re: python frontend for xenstat, Diwaker Gupta

Indexes:

[Date] [Thread] [Top] [All Lists]