ls segmentation fault

Confirmed that this happens after a reboot. Did not happen after a umount and remount.
Confirmed that running ls -l before running ls -latr means no segmentation faults.
Confirmed another program that segfaults in the backgroud.

kernel: [ 3690.913237] cron[4023]: segfault at 00002b96e84cd3b8 rip 00002b96e888d26f rsp 00007fffc2652a40 error 4

(I believe that it segfaulted while I was running to test tar. It segfaulted while running cron.hourly which has no jobs to run)

Also noted corruption of postfix mail spool:

postfix/pickup[4024]: fatal: scan_dir_push: open directory maildrop: Permission denied
postfix/master[3836]: warning: process /usr/lib/postfix/pickup pid 4024 exit status 1
postfix/master[3836]: warning: /usr/lib/postfix/pickup: bad command startup -- throttling

Once again I believe that this was happening during the test tar. tar has not segfaulted again, however there does seem to be some instability with other programs. I was able to fix the postfix issue with postfix set-permissions. It would be nice to know what is causing these issues.

Thank you for any input.

The above problems only seem to manifest with edgy. I have now installed dapper and do not have the ls -latr problem, or the cron segfaults, and postfix problem.

However tar crashes with the 4500+ directory:
Command run: cd /user # directory has 4500+ directories with 250Gs of data
Command run: tar cf /dev/null .
Error: *** glibc detected *** free(): invalid pointer: 0x00002aaaaaeff800 ***

Any input would be appreciated.

From your description, it would seem that ls/coreutils is not to blame, and xfs/kernel may be. The format of the error messages look like kernel complaints. However, I'm not confident to confirm it for linux-source, either, as it could conceivably be a glibc bug.

Were these error messages shown on the terminal, or did they come from a log file? Did you notice any related logs in the log files?

I would recommend opening a separate bug for tar, as it seems its failure may not be related to (presumed) xfs's.

We are closing this bug report as it lacks the information, described in the previous comments, we need to investigate the problem further. However, please reopen it if you can give us the missing information and don't hesitate to submit bug reports in the future.