We had an interesting experience last night. We have a RH Linux CVS server that holds several (hundred?)thousand repository files. A large number (>50) of the CVS files got corrupted (truncated) when (according to the history file) one of our users in Germany did a Tag request. The files affected were contained only within a few directories, but didn't seem to follow any pattern. This happened once before, last year. Same repository but a different machine (which we promptly executed and buried in the back lot). We now have a new superfast multi-processor raid-5 server. We would REALLY prefer this not happen again. One interesting coincidence was that most of the files needing to be restored were on the backup tape that was started just a few minutes before the corruption occurred. 10 were corrupted on this tape and we had to restore from the previous nights tape.
Is it possible that somehow our Veritas back up program is confusing CVS somehow? I can't see the Veritas backup daemon modifying the file, and our sysadmin doesn't think it should even be locking the file. Anyone have any other ideas? ===== --- Wayne Johnson, | There are two kinds of people: Those 3943 Penn Ave. N. | who say to God, "Thy will be done," Minneapolis, MN 55412-1908 | and those to whom God says, "All right, (612) 522-7003 | then, have it your way." --C.S. Lewis __________________________________________________ Do you Yahoo!? Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop! http://platinum.yahoo.com _______________________________________________ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
