I just thought I'd put this out here since I've seen plenty of other people who are experiencing segfaults in iPrint on OES1 or OES2 Linux - this may help on Netware also.

I recently did an upgrade of my iPrint Linux cluster that I described in the post entitled "SuSE9/OES1 iPrint cluster upgrade to SuSE10/OES2". The new OES2 cluster was patched and ended up with the April 22 iprint server rpm. Segfaulting started very soon after and updating to the June 13 debug code didn't help. All of the TID's and postings in this newsgroup indicated there must be other corruption with the iPrint db. I couldn't rebuild this thing by hand (300+ printers and growing daily). So grasping at straws to find the corruption I decided to have a look at the printmanager db backup files in the iPrint manager .psm directory. So I started paging through the csv and xml files with "less" - lo and behold I come across a print queue that had funny characters in the path pointing to the driver to be downloaded. I can't explain what they were - they just showed up in inverse/bold in less. I went back to iManager and deselected the driver for this printer and set it to "none". Amazingly my iPrint cluster has been running smoothly now for about two weeks and over 17000 print jobs. Before I did this iPrint was segfaulting several times a day.

If iPrint had detected a problem here I didn't see it mentioned in any of the logs - which I read repeatedly looking for something to give me a clue. Ultimately I figured if the db was corrupt somehow then the backup files would also be corrupt so I had a look. So far I am very happy!

I just thought this may help save someone some trouble and be a good place to start.

One related question - is the November 3 rpm the same as the June 13 debug code, only without all the debug stuff in? Should I move to the November rpm now that the June13 code seems to work good?

Wayne Fee
IT Enterprise Systems
Appalachian State University