Print

Print


Hi Goncalo,


we have bug 54144 open for this. The recommendation there from Maarten is 
"Until this is fixed, the admin may need to have the WM restarted 
regularly e.g. with a daily cron job". 


Best regards,
Andreas

On Mon, 31 Aug 2009, Gonçalo Borges wrote:

> Hi All...
> 
> With the last WMS update
> 
>      [root@wms01 etc]# rpm -qa | grep glite-WMS
>      glite-WMS-3.1.20-0
> 
> the glite-wms-workloag is using (again!) 1.3G of memory, and the machine 
> is heavily swapping.
> 
>       3320 glite     20   0  175   3391:37 63.5 2705m 1.3g 4608 S 
> glite-wms-workl
> 
> This was a known problem seen in the past which was solved by using 
> google_perf_tools. The problem now is that 
> /opt/glite/etc/init.d/glite-wms-wm  does not has the 
> "use_google_perf_tools=1" option anymore.
> 
> Does someone knows the workaround to this issue?
> 
> Cheers
> Goncalo
> 
> 
> 
> On 04/10/2009 11:15 AM, dongiovanni wrote:
> > Hi all.
> > high RAM usage is typical but can be enormously reduced using google 
> > malloc
> > first install from http://goog-perftools.sourceforge.net/ 
> > google-perftools-devel-1.0-1.i386.rpm
> > google-perftools-1.0-1.i386.rpm
> >
> > then in
> > /opt/glite/etc/init.d/glite-wms-wm
> > uncomment this  line
> >        #use_google_perf_tools=1
> >
> > restart wm daemon
> > then RAM usage by WM  should keep near 500MB
> > cheers
> > Danilo
> >
> >
> > Stuart Purdie wrote:
> >> We run two WMS here:
> >>
> >>   PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
> >>  4340 glite     25   0 1540m 1.4g 6300 S 98.2 17.6 368:57.33 
> >> glite-wms-workl
> >>
> >> and
> >>
> >>   PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
> >> 25263 glite     19   0 1679m 1.5g 6336 S 46.2 19.3 331:45.99 
> >> glite-wms-workl
> >>
> >> So I'd have to say that 1.6Gb of RAM is typical of what we see.
> >>
> >>
> >> On 9 Apr 2009, at 16:14, Gonçalo Borges wrote:
> >>
> >>> Dear All...
> >>>
> >>> The glite-wms-workload  is using 1.6GB of memory in my WMS. This is 
> >>> the 3rd time I see this behaviour in a 2 weeks period.
> >>>
> >>> top - 16:11:28 up 6 days,  2:43,  1 user,  load average: 0.09, 0.24, 
> >>> 0.56
> >>> Tasks: 129 total,   2 running, 127 sleeping,   0 stopped,   0 zombie
> >>> Cpu(s):  0.2% us,  0.0% sy,  0.0% ni, 99.8% id,  0.0% wa,  0.0% hi,  
> >>> 0.0% si
> >>> Mem:   2070188k total,  1969332k used,   100856k free,    42100k 
> >>> buffers
> >>> Swap:  2096472k total,  1048160k used,  1048312k free,   138520k cached
> >>>
> >>>  PID USER      PR  NI %CPU    TIME+  %MEM  VIRT  RES  SHR S 
> >>> COMMAND                              3593 glite     20   0    0   
> >>> 2177:42 79.4 1699m 1.6g 4716 S glite-wms-workl
> >>> I guess this is not normal... Can someone else check there values? 
> >>> Do I need to do some extra tunning?
> >>>
> >>> Cheers
> >>> Goncalo
> >>
> >
> 

-- 
Andreas Unterkircher
IT Department
Grid Deployment Group
CERN
CH-1211 Geneva 23