[Vagnekman-users] Configuration change of the login node
lars malinowsky
lama at pdc.kth.se
Thu Nov 18 12:50:32 CET 2010
Hello,
as several transfers (rsync/scp) in and out of ekman
seem to be hit the 10 minute cpu-time limit, the ekman
compute-node k33n41.pdc.kth.se can be used for these
transfers - you should be able to log on to it and use
it in the same manner as you log on to ekman.pdc.kth.se.
Note that this compute node does not have a 10Gbit
network to the outside world, only a 1Gbit network.
Consider this a stop-gap solution, for
transfers only, until advised to do differently.
It's unknown whether the ffv-utility works out
of the box or not on k33n41.pdc.kth.se.
k33n41.pdc.kth.se is now also known as
ekman-rsync.pdc.kth.se for the time being.
regards,
lars/pdc-staff.
[..]
> Dear Users,
>
> today we had a problem with memory overuse on the login
> node. Therefore we established here limits which showed
> helpful already on other systems like Ferlin.
> All user processes are limited now to
> * 10 minutes of CPU time
> * 8 GByte RAM, th.i. 50% of the machine RAM.
> So it should be not so easy to bring the login
> node in an unstable state.
> The consequence is that some resource hungry things you
> may have done before on the login node will not work longer.
> For example can this be a compilation of a large source file
> with the Inel compiler and IPO (inter procedure optimization)
> turned on.
> Please use for these works a node which you acquire
> with the help of easy. The use of dedicated nodes will
> disturb other users much less.
> If it is desired we could also establish one or two interactive
> nodes where you can login without a job submission.
> Best regards -- Med vänlig hälsning
> Michael Schliephake
More information about the Vagnekman-users
mailing list