Re: [Qemu-devel] [pve-devel] QEMU LIve Migration - swap_free: Bad swap file entry

[Stefan Priebe - Profihost AG]

Am 07.02.2014 09:15, schrieb Alexandre DERUMIER:
> 
> do you use xbzrle for live migration ?

no - i'm really stucked right now with this. Biggest problem i can't
reproduce with test machines ;-(

Stefan


> 
> ----- Mail original ----- 
> 
> De: "Stefan Priebe" <address@hidden> 
> À: "Dr. David Alan Gilbert" <address@hidden> 
> Cc: "Alexandre DERUMIER" <address@hidden>, "qemu-devel" <address@hidden> 
> Envoyé: Jeudi 6 Février 2014 21:00:27 
> Objet: Re: [Qemu-devel] [pve-devel] QEMU LIve Migration - swap_free: Bad swap 
> file entry 
> 
> Hi, 
> Am 06.02.2014 20:51, schrieb Dr. David Alan Gilbert: 
>> * Stefan Priebe - Profihost AG (address@hidden) wrote: 
>>> some more things which happen during migration: 
>>>
>>> php5.2[20258]: segfault at a0 ip 0000000000740656 sp 00007fff53b694a0 
>>> error 4 in php-cgi[400000+6d7000] 
>>>
>>> php5.2[20249]: segfault at c ip 00007f1fb8ecb2b8 sp 00007fff642d9c20 
>>> error 4 in ZendOptimizer.so[7f1fb8e71000+147000] 
>>>
>>> cron[3154]: segfault at 7f0008a70ed4 ip 00007fc890b9d440 sp 
>>> 00007fff08a6f9b0 error 4 in libc-2.13.so[7fc890b67000+182000] 
>>
>> OK, so lets just assume some part of memory (or CPU state, or memory 
>> loaded off disk...) 
>>
>> You said before that it was happening on a 32GB image - is it *only* 
>> happening on a 32GB or bigger VM, or is it just more likely? 
> 
> Not image, memory. I've only seen this with vms having more than 16GB or 
> 32GB memory. But maybe this also indicates that just the migration takes 
> longer. 
> 
>> I think you also said you were using 1.7; have you tried an older 
>> version - i.e. is this a regression in 1.7 or don't we know? 
> Don't know. Sadly i cannot reproduce this with test VMs only with 
> production ones. 
> 
> Stefan 
> 
>> Dave 
>> -- 
>> Dr. David Alan Gilbert / address@hidden / Manchester, UK 
>>