Last modified 12 years ago Last modified on 09/01/09 18:03:27

dCache Tier I meeting September 1, 2009

[part of a series of meetings]

Present, IN2P3(), Sara(Onno), Triumf(Simon), BNL(Pedro), NDGF(Gerd), PIC(), GridKa(Doris, Silke), Fermi(), CERN(Andrea)


Site reports


  • Strange problems with one of the poools. Pool Usage Page : Code 204 : File could not be opened. Please check the file system. File system is XFS. File system seems to be ok. The reported PNFS was ok. Checksum was ok. After restart of the pool it seemed to be ok. But there are still problems with some files. Reading with SRM reports timeout. No entry in the log files for the requested pnfsid's. Pools stays online. Files are not in the restore queue. Try : try to find the file in the PinManager, try to read the file with dcap, and read other files from the pool.
  • User forum message : usage of remove orphant file script. Access to the ssh interface is not reliable. Possibly should be replaced by JPython.


Nothing to report.


  • resilient dcache reduction times out #5064. Will be discussed at the developers phone conference.
  • resilient dcache on space token area #5050. Right now Pedro is happy using the RM in a manual fashion. This is of course not a good solution. Fermilab would need to decide if they want to put some more effort into the RM in order to enable it support this feature.
  • dcap solaris library not available w/o preload. We make it available for BNL. Talk to Ofer about this.


  • upgrade 1.9.5 : no new features enabled yet. (tape protection acls)


  • Info : gridKa had a problem with gridftp movers getting stuck. Active movers were blocking. Not observer any more. BUT active dCap movers stuck. Paul got all the information from Doris yesterday. No timeout configured for dcap as file are supposed to stay open for days.

To Do List

  • Onno needs information on JPhyton
  • Solaris version to Pedro and Ofer.


Proposed: same time, next week.