On Wednesday 21 March 2007 10:55, Graeme Stewart wrote:
> Hi Steve
>
> No, the box isn't overloaded:
>
> 10:46:11 up 145 days, 12:55, 1 user, load average: 0.17, 0.20, 0.19
> 625 processes: 620 sleeping, 1 running, 4 zombie, 0 stopped
> CPU states: cpu user nice system irq softirq iowait idle
> total 0.0% 0.0% 0.5% 0.0% 0.0% 0.9%
> 98.3%
> cpu00 0.0% 0.0% 0.7% 0.0% 0.0% 0.0%
> 99.2%
> cpu01 0.1% 0.0% 0.3% 0.0% 0.0% 1.9%
> 97.4%
> Mem: 8195868k av, 3505276k used, 4690592k free, 0k shrd,
> 376700k buff
> 1191472k active, 1615960k inactive
> Swap: 4192880k av, 0k used, 4192880k free
> 2008148k cached
>
> It's quite a beefy machine.
>
> Most of the load comes from the top level BDII, which runs on the
> same box.
>
> Where's the test client?
export RGMA_HOME=/opt/glite
/opt/glite/bin/rgma-client-check
You will also have to setup your cert proxy and set the environmet variable
X509_USER_PROXY to point to it.
Alastair
>
> Cheers
>
> Graeme
>
> PS. It's actually been somewhat better this week (because the cluster
> is quiet?), but see the problems logged here last weekend
>
> http://www.gridpp.ac.uk/wiki/Glasgow_Middleware_Operations_Logbook
>
> On 21 Mar 2007, at 10:38, Fisher, SM (Steve) wrote:
> > Is your mon box overloaded? Does the r-gma client check script run
> > quickly?
> >
> > Steve
> >
> >> -----Original Message-----
> >> From: Testbed Support for GridPP member institutes
> >> [mailto:[log in to unmask]] On Behalf Of Graeme Stewart
> >> Sent: 21 March 2007 10:16
> >> To: [log in to unmask]
> >> Subject: Re: SAME tests run in job wrapper
> >>
> >> Yes, I've noticed this at Glasgow - and in particular R-GMA is adding
> >> 5-15 minutes of wallclock time to every job, which is a
> >> terrible waste of resources (particularly for our GRAM GT2
> >> user groups, who run some pretty short jobs).
> >>
> >> Something I really need to look into...
> >>
> >> g
> >>
> >> On 21 Mar 2007, at 09:26, Stephen Childs wrote:
> >>> Apologies if this has already been discussed, but is it the
> >>
> >> case that
> >>
> >>> some SAME tests now get run within the job wrapper? I
> >>
> >> remember seeing
> >>
> >>> something about this at some stage and looking on a node where I'm
> >>> running a simple /bin/hostname using globus-job-run I see
> >>
> >> this kind of
> >>
> >>> stuff:
> >>>
> >>> 6142 7118 0.3 2.6 9372 6848 ? S 09:19 0:01
> >>> python /opt/lcg/same//client/bin/same-exec --nodetest WN
> >>> gridmon.cs.tcd.ie -- stage=start.stage1 sh_results=/home/gitest042/
> >>> globus-tmp.gridmon.6767.0/.same_result.xq7111 sh_results_all=/home/
> >>> gitest042/globus-tmp.gridmon.6767.0/.same_details.Fv7113
> >>> starttime=1174468780 stage2_dir=/home/gitest042/globus-tmp.gridmon.
> >>> 6767.0/.same_stage2.oF7109
> >>>
> >>> Of course the problem is that the SAME stuff is getting
> >>
> >> stuck at the
> >>
> >>> moment. Is there any way to disable it?
> >>>
> >>> Stephen
> >>> --
> >>> Dr. Stephen Childs,
> >>> Research Fellow, EGEE Project, phone:
> >>> +353-1-8961797
> >>> Computer Architecture Group, email: Stephen.Childs @
> >>> cs.tcd.ie
> >>> Trinity College Dublin, Ireland web: http://www.cs.tcd.ie/
> >>> Stephen.Childs
> >>
> >> --
> >> Dr Graeme Stewart - http://wiki.gridpp.ac.uk/wiki/User:Graeme_stewart
> >> GridPP DM Wiki - http://wiki.gridpp.ac.uk/wiki/Data_Management
> >> ScotGrid - http://www.scotgrid.ac.uk/ http://scotgrid.blogspot.com/
>
> --
> Dr Graeme Stewart - http://wiki.gridpp.ac.uk/wiki/User:Graeme_stewart
> GridPP DM Wiki - http://wiki.gridpp.ac.uk/wiki/Data_Management
> ScotGrid - http://www.scotgrid.ac.uk/ http://scotgrid.blogspot.com/
|