<!-- keep this as a security measure: #uncomment if the subject should only be modifiable by the listed groups * Set ALLOWTOPICCHANGE = Main.TWikiAdminGroup,Main.CMSAdminGroup * Set ALLOWTOPICRENAME = Main.TWikiAdminGroup,Main.CMSAdminGroup #uncomment this if you want the page only be viewable by the listed groups # * Set ALLOWTOPICVIEW = Main.TWikiAdminGroup,Main.CMSAdminGroup --> %TOC% %ICON{arrowleft}% Go to [[CMSTier3LogXX][previous page]] / [[CMSTier3LogXX][next page]] of Tier3 site log %M% ---+ 16. 11. 2012 Restarted pnfsd <pre> Thanks for the info! M. From: Fabio Martinelli <fabio.martinelli@psi.ch<mailto:fabio.martinelli@psi.ch>> Organization: Paul Scherrer Institut Date: Friday, November 16, 2012 10:28 AM To: "cms-tier3@lists.psi.ch<mailto:cms-tier3@lists.psi.ch>" <cms-tier3@lists.psi.ch<mailto:cms-tier3@lists.psi.ch>> Cc: Pablo Fernandez <pablo.fernandez@cscs.ch<mailto:pablo.fernandez@cscs.ch>>, Miguel Gila <miguel.gila@cscs.ch<mailto:miguel.gila@cscs.ch>> Subject: Re: pnfsDump error on t3dcachedb01 Hi guys maybe an experience that can be recycled at CSCS too. I was searching for a way to get in pause mode dCache because of our pnfsDump error; eventually I ran : * [t3se01.psi.ch] (PoolManager) admin > rc ls # to find an idle moment * [t3se01.psi.ch] (PoolManager) admin > rc suspend on -all # to stop * raised to 16 the shmclients and shmservers and restarted pnfsd * [t3se01.psi.ch] (PoolManager) admin > rc suspend off -all # to turn it on again * [t3se01.psi.ch] (PoolManager) admin > rc retry * -force-all # to retry the tens and tens of requests that in the meantime were intentionally suspended so now I see: /opt/pnfs/tools/shmcom stat 1122|egrep '^(Client|Server)' Clients 16 Servers 16 and pnfsDump can go ahead; me with him. Following the pnfsd restart My Nagios can retrieve by SRM a file from each pool, like it usually does. An SRM write worked too. perhaps it can be a generic trick to use when you want to change something on the fly. ciao, Fabio On 11/15/12 5:41 PM, Fabio Martinelli wrote: Hi Derek today during my migration tests I got this error, that according to [1] seems because a little number of shmclients; please do you have experience about that ? I would augment it but it needs a pnfsd restart; if you don't then I have to restart pnfs in a moment where the T3 is unused; so far I'm stuck in my previous tests it never happened. ciao, Fabio ****************** + /opt/pnfs/tools/pnfsDump -r 000000000000000000001080 -vv -d10 -o chimera /var/CHIMERA_MIGRATION/pnfs.sql -2 -p 0000E0C7D9D3ABCC4B518BDC7B16EE78FFF5 -o verify /var/CHIMERA_MIGRATION/verify.md5 -r -o files /var/CHIMERA_MIGRATION/files.lst -f Unable to initialise shared-memory interface: sclClientOpen: Unknown(0) (-330) shmget: Identifier removed # /opt/pnfs/tools/shmcom stat 1122|egrep '^(Client|Server)' Clients 8 Servers 8 # cat /usr/etc/pnfsSetup shmkey=1122 shmclients=8 shmservers=8 pnfs=/opt/pnfs ... [1] http://svn.dcache.org/pnfs/trunk/docs/README.pnfsDump </pre> -- Main.FabioMartinelli - 2012-11-16 ---------------- %ICON{arrowleft}% Go to [[CMSTier3LogXX][previous page]] / [[CMSTier3LogXX][next page]] of Tier3 site log %M%
This topic: CmsTier3
>
WebHome
>
CMSTier3Log
>
CMSTier3Log30
Topic revision: r1 - 2012-11-16 - FabioMartinelli
Copyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki?
Send feedback