aahh.......... shift.conf; totally forgot about that.
I think, it's okay on the disk06 (the new storage pool ):
[root@disk06 ~]# cat /etc/shift.conf
RFIOD TRUST serv02.hep.phy.cam.ac.uk disk06.hep.phy.cam.ac.uk
RFIOD WTRUST serv02.hep.phy.cam.ac.uk disk06.hep.phy.cam.ac.uk
RFIOD RTRUST serv02.hep.phy.cam.ac.uk disk06.hep.phy.cam.ac.uk
RFIOD XTRUST serv02.hep.phy.cam.ac.uk disk06.hep.phy.cam.ac.uk
RFIOD FTRUST serv02.hep.phy.cam.ac.uk disk06.hep.phy.cam.ac.uk
DPM PROTOCOLS rfio gsiftp
but wasn't okay on the head node, as disk06 wasn't listed in there:
Now added and glite-SE_dpm_disk ran happily ever after.
Question: Do I need to manually add the [new] hostname to the shift.conf
on the head node every time I add new disk server to the pool?
There is only one more thing to sort out, I think. Why it's pulling in
v1.6.0 in stead of v1.7.0 by default? My other pool node have v1.7.0
installed. This is the repo I'm using at the moment:
[root@disk06 ~]# cat /etc/yum.repos.d/gLite-dm.repo [ OK ]
[glite-dpm-disk]
name=glite DPM disk
baseurl=http://glitesoft.cern.ch/EGEE/gLite/R3.1/glite-SE_dpm_disk/sl4/$basearch
enabled=1
protect=1
Cheers,
Santanu
Andrew Elwell wrote:
>> dpm-qryconf: Permission denied
>> dpm-addfs dpmCam_2009 disk06.hep.phy.cam.ac.uk /dpm_data2: Permission denied
>> dpm-qryconf: Permission denied
>>
>
>
>
>> [root@serv02 ~]# dpm-qryconf
>> POOL dpmCam_2007 DEFSIZE 200.00M GC_START_THRESH 0 GC_STOP_THRESH 0
>> DEF_LIFETIME 7.0d DEFPINTIME 2.0h MAX_LIFETIME 1.0m MAXPINTIME 12.0h
>> FSS_POLICY maxfreespace GC_POLICY lru RS_POLICY fifo GIDS 0 S_TYPE -
>>
>
>
> hmm - odd that dpm-qryconf failed to begin with - whats the status of
> /etc/shift.conf on both the new storage pool and the headnode?
>
> They should each have the other listed
>
> I'd cut n paste an example but don't have a machine to hand. Oh, and
> make sure your hostnames are all lowercase ....
>
|