While it is true file corruption doesn't happen very often it still happens. There is no fix that will prevent file corruption from ever occurring although a ticket was assigned to Steve Traylen asking him to put some kind of check for file corruption in the standard setup instruction. (He hasn't done it yet)
RAL has not hit corruptions problems because it has been running cvmfs_fsck checks for a long time. I think RAL have had two or three corrupt files ever and these have always been caught quickly. SARA ran into problems as it did not have checks running and so a few WN became blackholes during re-processing. You should definitely be running some kind of check.
Alastair
On 19 Oct 2011, at 14:56, Alessandra Forti wrote:
> My understanding is that the desire to run cvmfs_fsck on a regular basis actually comes from problems with corruption that have now been fixed. The Tier-1 hit them and ran cvmfs_fsck on a regular basis. Manchester also ran into the problem and filed a ticket. Once I'd fixed QMUL's StoRM problems I noticed lots of failed analysis jobs on a subset of nodes - sending the dodgy files to Jakob resulted in a fixed cvmfs a couple of weeks later. I've not seen problems since (Adding regular cvmfs_fsck is still on my todo list). Chris
>
> yes, the problem if in the ticket I opened and it is corrected in version 2.0.3
>
> http://northgrid-tech.blogspot.com/2011/09/cvmfs-upgrade-to-203.html
>
> I updated the wiki
>
> https://www.gridpp.ac.uk/wiki/UK_CVMFS_Deployment#cvmfs_fsck_setup
|