Testing of New dCache Storage Node Build

Here is the testing result for a new storage node (summary: it works). After running \x91install_dcache.sh\x92 on UMFS16 and starting \x91dcache\x92, I logged into the head01 dcache admin interface. The pools already show up:

ssh -l admin localhost -p 22223 -c blowfish -1

[head01.aglt2.org] (PoolManager) admin > psu ls pool
\x85
umfs09_3
umfs16_2
umfs16_1
umfs16_4
umfs16_3
\x85

Now add the \x91umfs16_1\x92 pool to AGLT2_SCRATCHDISK:

[head01.aglt2.org] (PoolManager) admin > psu addto pgroup AscratchPools umfs16_1

[head01.aglt2.org] (PoolManager) admin > psu ls -a pgroup AscratchPools
AscratchPools
 linkList :
   ascratch-link  (pref=10/10/-1/10;;ugroups=3;pools=1)
 poolList :
   msufs06_4  (enabled=true;active=1;rdOnly=false;links=0;pgroups=1;hsm=[];mode=enabled)
   thor01_4  (enabled=true;active=22;rdOnly=false;links=0;pgroups=1;hsm=[];mode=enabled)
   umfs16_1  (enabled=true;active=0;rdOnly=false;links=0;pgroups=1;hsm=[];mode=enabled)

So the pool added without problems.

Let\x92s try a \x93read-write\x94 test into the storage. As myself I get VOMS ATLAS production role:

smckee@umt3int02 ~$ source /afs/atlas.umich.edu/opt/WLCG-client/setup.sh
smckee@umt3int02 ~$ voms-proxy-init -voms atlas:/atlas/Role=production
Enter GRID pass phrase:
Your identity: /DC=org/DC=doegrids/OU=People/CN=Shawn McKee 83467
Creating temporary proxy ....................................................................... Done
Contacting  voms.cern.ch:15001 [/DC=ch/DC=cern/OU=computers/CN=voms.cern.ch] "atlas" Done
Creating proxy ....................................................... Done
Your proxy is valid until Thu Feb 11 02:23:42 2010
smckee@umt3int02 ~$ voms-proxy-info -all
subject   : /DC=org/DC=doegrids/OU=People/CN=Shawn McKee 83467/CN=proxy
issuer    : /DC=org/DC=doegrids/OU=People/CN=Shawn McKee 83467
identity  : /DC=org/DC=doegrids/OU=People/CN=Shawn McKee 83467
type      : proxy
strength  : 1024 bits
path      : /tmp/x509up_u10048
timeleft  : 11:49:12
=== VO atlas extension information ===
VO        : atlas
subject   : /DC=org/DC=doegrids/OU=People/CN=Shawn McKee 83467
issuer    : /DC=org/DC=doegrids/OU=Services/CN=vo.racf.bnl.gov
attribute : /atlas/Role=production/Capability=NULL
attribute : /atlas/Role=NULL/Capability=NULL
attribute : /atlas/lcg1/Role=NULL/Capability=NULL
attribute : /atlas/usatlas/Role=NULL/Capability=NULL
timeleft  : 11:49:12
uri       : vo02.racf.bnl.gov:15003

Now I can try a copy into AGLT2_SCRATCH (NOTE: you must use the space-token description via \x96S flag!):

smckee@umt3int02 ~$ lcg-cp -v -b -D srmv2 -S ATLASSCRATCHDISK file:////etc/hosts srm://head01.aglt2.org:8443/srm/managerv2?SFN=/pnfs/aglt2.org/atlasscratchdisk/user/hosts
Using grid catalog type: UNKNOWN
Using grid catalog : (null)
VO name: atlas
Checksum type: None
Destination SE type: SRMv2
Destination SRM Request Token: -2086041495
Source URL: file:/etc/hosts
File size: 107
Source URL for copy: file:/etc/hosts
Destination URL: gsiftp://umfs11.aglt2.org:2811//pnfs/aglt2.org/atlasscratchdisk/user/hosts
# streams: 1
          107 bytes      0.61 KB/sec avg      0.61 KB/sec inst
Transfer took 1000 ms
smckee@umt3int02 ~$ ls /pnfs/aglt2.org/atlasscratchdisk/user
hosts  ueda

Worked. Let\x92s disable ALL pools but the new UMFS16_1 pool for writes:

[head01.aglt2.org] (local) admin > cd msufs06_4
[head01.aglt2.org] (msufs06_4) admin > pool disable -rdonly
Pool msufs06_4 disabled(store,stage,p2p-client)
[head01.aglt2.org] (msufs06_4) admin > ..
[head01.aglt2.org] (local) admin > cd thor01_4
[head01.aglt2.org] (thor01_4) admin > pool disable -rdonly
Pool thor01_4 disabled(store,stage,p2p-client)

Then we can retry a copy:

smckee@umt3int02 ~$ lcg-cp -v -b -D srmv2 -S ATLASSCRATCHDISK file:////etc/hosts srm://head01.aglt2.org:8443/srm/managerv2?SFN=/pnfs/aglt2.org/atlasscratchdisk/user/hosts2
Using grid catalog type: UNKNOWN
Using grid catalog : (null)
VO name: atlas
Checksum type: None
Destination SE type: SRMv2
Destination SRM Request Token: -2086040697
Source URL: file:/etc/hosts
File size: 107
Source URL for copy: file:/etc/hosts
Destination URL: gsiftp://umfs07.aglt2.org:2811//pnfs/aglt2.org/atlasscratchdisk/user/hosts2
# streams: 1
          107 bytes      0.34 KB/sec avg      0.34 KB/sec inst
Transfer took 1000 ms
smckee@umt3int02 ~$ ls /pnfs/aglt2.org/atlasscratchdisk/user                    hosts  hosts2  ueda

Worked! Test is successful.

Undoing changes and removing UMFS16 from the production setup\x85

(re-enable pools, delete files from /pnfs\x85)

[head01.aglt2.org] (PoolManager) admin > psu removefrom pgroup AscratchPools umfs16_1
[head01.aglt2.org] (PoolManager) admin > psu ls -a pgroup AscratchPools
AscratchPools
 linkList :
   ascratch-link  (pref=10/10/-1/10;;ugroups=3;pools=1)
 poolList :
   msufs06_4  (enabled=true;active=1;rdOnly=false;links=0;pgroups=1;hsm=[];mode=enabled)
   thor01_4  (enabled=true;active=20;rdOnly=false;links=0;pgroups=1;hsm=[];mode=enabled)

-- ShawnMcKee - 10 Feb 2010

• install_dcache.sh: The script used to put dCache on a storage node

• dCacheSetup: The stock AGLT2 dCacheSetup config file

• node_config.storage: The stock AGLT2 node_config file

• srm_setup.env: Stock AGLT2 srm_setup.env file