Description of problem: Version-Release number of selected component (if applicable): slocate-2.7-13.el4.6 How reproducible: updatedb Steps to Reproduce: 1. updatedb Actual results: *** glibc detected *** double free or corruption (top): 0x08553d60 *** Expected results: nothing Additional info:
Thanks for your report. If you are a RHEL customer and have an active support entitlement, please contact official Red Hat Support at https://www.redhat.com/apps/support/ to allow correct prioritization of this issue. Otherwise, please install the valgrind and slocate-debuginfo packages, run the following commands: chmod g-s /usr/bin/updatedb valgrind --log-file=updatedb.log updatedb chmod g+s /usr/bin/updatedb and attach the generated updatedb.log* files.
Created attachment 144891 [details] valgrind --log-file=/root/updatedb.log updatedb
Thanks. Please install http://people.redhat.com/mitr/221406/slocate-2.7-13.el4.6.0.mitr.test221406.1.i386.rpm and: - run (updatedb 2>&1 |tee log), and attach at least the last 100 lines of the generated log output. This should crash similarly to slocate-2.7-13.el4.6. - run (TEST_FIX=1 updatedb 2>&1 | tee log2). This should not crash; if it does, please attach at least the last 100 lines of the generated log2 output. This is only a debugging version of the package, please return to 2.7-13.el4.6 after performing the tests Please attach the contents of your /etc/mtab as well.
Created attachment 145143 [details] updatedb 2>&1 |tee log
Created attachment 145144 [details] TEST_FIX=1 updatedb 2>&1 | tee log2
Created attachment 145145 [details] /etc/mtab
Thank, I run the supplied version and it still crashes (even with TEST_FIX=1): see attachments. Matteo
Thanks. This demonstrates that the root cause of the problems is that the inode number of /usr/sepp/Old/pack/openssl-0.9.8c-lp is changing while updatedb is running; this can't be fixed within slocate. http://www.openafs.org/pipermail/openafs-devel/2006-June/013901.html seems to suggest this is a bug in OpenAFS. Nevertheless, updatedb shouldn't crash, and I can't figure out why TEST_FIX=1 doesn't work. Please install http://people.redhat.com/mitr/221406/slocate-2.7-13.el4.6.0.mitr.test221406.1.i386.rpm , run (updatedb 2>&1 |tee log), and attach the generated log file.
Created attachment 145246 [details] updatedb 2>&1 |tee log
Thanks again, OK. I'll wait to try to fix AFS so that we can reproduce the bug. Matteo
I'm really sorry, I intended to point you to http://people.redhat.com/mitr/221406/slocate-2.7-13.el4.6.0.mitr.test221406.2.i386.rpm (not ...test221406.1.i386.rpm). Can you attach the log from this version, please?
Created attachment 145331 [details] updatedb 2>&1 |tee log4 slocate-2.7-13.el4.6.0.mitr.test221406.2.i386.rpm
Thanks. This version didn't crash, did it?
Sorry I forgot to comment last run. It didn't crash (although it seems to stop where the problem occurs, i.e., the last processed entry seems the same). Matteo
Thank you. Stopping at the problem directory is expected, there is no practical way to recover from the inode mismatch.
Created attachment 145333 [details] Make sure fts_cur is always valid after return from fts_read ()
This request was evaluated by Red Hat Product Management for inclusion, but this component is not scheduled to be updated in the current Red Hat Enterprise Linux release. If you would like this request to be reviewed for the next minor release, ask your support representative to set the next rhel-x.y flag to "?".
Thank you for submitting this issue for consideration in Red Hat Enterprise Linux. The release for which you requested us to review is now End of Life. Please See https://access.redhat.com/support/policy/updates/errata/ If you would like Red Hat to re-consider your feature request for an active release, please re-open the request via appropriate support channels and provide additional supporting details about the importance of this issue.