Varnish child killed

Jean-Francois Laurens jean-francois.laurens at rts.ch
Thu Apr 21 11:27:58 CEST 2011


Hi Michal,

Thanks for your advise ! It is kind of drastical though !

We have an other varnish instance in version 2.0.6 which is using memory
(hum even if we specified a ­sfile ;-) and actually you¹re right it¹s pretty
stable and fast.
Thus I would be interested in having some feedback from folks running cache
on disk file !
The thing is that the websites behind do get a lot of trafic and have a lot
of content as we¹re media broadcasters.

The ugly thing is that the servers running varnish instances can not get
more than 8G of memory which is clearly not enough.
 

Cheers,

Jef

Le 21/04/11 11:12, « Michal Taborsky » a écrit :

>    Hello Jean-Francois,
>  
>  we have seen similar behavior. You did not specify what platform you use so I
> assume Linux. After some studying and experimentation my recommendation is:
>  a) always make sure varnish uses only memory, never disc, if you expect good
> performance. So specify the cache size smaller than the available memory you
> have (some memory should be reserved for other processes and some varnish
> control structures, we use 14G cache size on 16G box)
>  b) use malloc storage type
>  
>  Hope this helps,
>  Michal
>  
>  
>  Dne 21.4.2011 10:51, Jean-Francois Laurens napsal(a):
>>  Varnish child killed Hi there,
>>  
>>  We¹re run varnish 2.1.5 for some week now and we still do not understand
>> some behavior regarding the shared memory activity.
>>  We specified a ­sfile,/var/lib/varnish/varnish_storage.bin,50G in the
>> configuration but it¹s impossible to go higher than 25G used by varnish.
>> Please see the following cacti graph:
>>  
>>  
>>  
>>  In addition I can see varnish doesn¹t seem to be able to handle more than 1
>> million objects:
>>  
>>  
>>  When the child process get killed, the load of the system was very high:
>>  Apr 20 21:46:44 server-01-39 varnishd[21087]: Child (5372) not responding to
>> CLI, killing it.
>>  ....
>>  Apr 20 21:49:57 server-01-39 nrpe[18101]: Command completed with return code
>> 2 and output: CRITICAL - load average: 159.00, 159.32,
>> 77.02|load1=159.000;15.000;30.000;0; load5=159.320;10.000;25.000;0;
>> load15=77.020;5.000;20.000;0;
>>  ....
>>  Apr 20 21:48:43 server-01-39 varnishd[21087]: Child (5372) not responding to
>> CLI, killing it.
>>  
>>  All this makes me believe we have an issue with some kernel parameters that
>> do not allow varnish to handle as many objects as we configured it.
>>  
>>  Would anybody have an advice for this problem ?
>>  
>>  Jef
>>  
>>  Jean-Francois Laurens
>>  Ingénieur Système Unix
>>  Resources et Développement
>>  Secteur Backend
>>  RTS - Radio Télévision Suisse
>>  Quai Ernest-Ansermet 20                        
>>  Case postale 234                                    
>>  CH - 1211 Genève 8
>>  T +41 (0)58 236 81 63
>>  
>>  
>  
>  

Jean-Francois Laurens
Ingénieur Système Unix
Resources et Développement
Secteur Backend
RTS - Radio Télévision Suisse
Quai Ernest-Ansermet 20
Case postale 234   
CH - 1211 Genève 8
T +41 (0)58 236 81 63


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.varnish-cache.org/lists/pipermail/varnish-misc/attachments/20110421/ff319aca/attachment-0003.html>


More information about the varnish-misc mailing list