Description of problem: Machine panicked during seemingly normal operation, but I can't be sure since this is a general login server and had ~100 users logged in at the time. The panic appears to be related to autofs. This particular machine relies heavily on autofs, mounting several hundreds of filesystems from a large number of NFS servers around campus, mostly Solaris, HP-UX and Linux. The output (grabbed from serial console) is attached. Version-Release number of selected component (if applicable): 2.6.9-11.ELsmp How reproducible: Unknown. It's the first time I've seen this.
Created attachment 116591 [details] Console messages during panic
Looks like a recursive lookup of some sort. Is this problem reproducable at all? Could you please provide the information requested under the section "Filing bug reports" on the following web page? http://people.redhat.com/jmoyer Thanks.
Sure, her it is: - autofs version: autofs-4.1.3-131 - kernel version: 2.6.9-11.ELsmp I'll attach the contents of auto.master and the maps. Debug logging with syslog seems to create large files very fast, so I've disabled that for now. I don't know how to reproduce the panic anyway..
Created attachment 117118 [details] /etc/auto.master
Created attachment 117119 [details] Contents of /etc/auto.d/
This looks very much like a problem that was encountered when doing bind mounts. Are there any bind mounts in your setup? Can you post the output of 'mount' for your setup? I have a patch that should address this issue. Is this easily reproducable?
I'm still waiting for answers to the above questions. Could you provide an update, please?
Sorry for the late reply. No, there aren't any bind mounts, and I'll attach the output from 'mount'. I have seen this problem only once, and have no idea how to reproduce it, since I don't know what triggered it in the first place.
Created attachment 120305 [details] Output from mount. Note that the autofs setup has been changed slightly since I attached the autofs maps, as new disks are mounted.
Have you seen this problem again? I really think that it was fixed in kernel version 2.6.9-22.6. Without debug logs, I'll be unable to diagnose this exactly. Could you at least try the U3 kernel (2.6.9-34.EL) and let me know if it resolves the problem for you? Thanks!
I have not seen the problem again. Only once. But I'll install the latest kernel anyway and let you know if it happens again. I think it's safe to close this bug for now. I'll reopen it if I ever see this problem again. Thanks Jeff.
OK, thanks!