Re: [HTCondor-users] Blast job crash !

[Romain <nuelromain@xxxxxxxxx>]

Cotton, Benjamin J <bcotton@...> writes:

> It does sound like the NAS can't handle the load. If it were a
> configuration issue, that shouldn't affect ls.
> 
> Shameless plug: You may also be interested in using the BLASTer tool
> available on the DiaGrid hub:
> http://diagrid.org/resources/blastgui
> 
> Thanks,
> BC
> 

Thanks for your response.

You're right, it's a problem with the NAS or more precisely with transfer to 
the NAS I guess.
I try change mount system from NFS to CIFS and jobs run correctly with no 
problem (I launch more than 700 jobs in one submission and execute with no 
problem)

But there's a lil problem, CIFS is slower than NFS...
So for a simple and quick job there's no differences or a very little but 
for 1 000 or 5 000 or 10 000 jobs it can be very very slower I think.

-I use this (in /etc/fstab) for mount the NAS:
# NFS
IP_ADDRESS:/NAS   /SHARED_SPACE_NAME    nfs         exec,rw 0
# CIFS
//IP_ADDRESS/NAS/  /SHARED_SPACE_NAME   cifs    
exec,file_mode=0777,dir_mode=0777,uid=xxxx,gid=xxxx,username=yyyy,password=z
zzz,iocharset=utf8 0 0

What can I do to solve the problem ? can I use NFS with others options mount 
for better performances ?

Thanks
Regards.

--Romain--