From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; nb-NO; rv:1.7.10) Gecko/20050720 Fedora/1.0.6-1.1.fc4 Firefox/1.0.6 Description of problem: In testing of a High Availability database we have see kernel related problems with shared memory. The database is a heavy user of SVr4 shared memory. In the process of taking down the database and freeing shared memory we has seen complete hangs of the system. On 4-ways (2 Xeon w/HT) systems we see two processes each eating up one CPU. The commands $ ipcs -s $ ipcrm then just hangs, however $ ipcs -m works. The two processes eating CPU can't be killed and we has to reboot to recover the machine. On two-way systems we see more complete hangs as is't not possible to login, however we get answer from ping. We has to reboot to get the machine in shape. We wonder if this is a timing/race problem in the shared memory kernel code as it's not possible to reproduce the problem at will, but happens from time to time. These machines are running RHEL AS 3 U4 with kernel-smp-2.4.21-27.EL, the hardware is Xeon 2.8 GHz and 1 GB RAM. BTW: We have not seen these problems on Solaris 9 and 10 and Windows 2003 Enterprise Edition. Version-Release number of selected component (if applicable): kernel-smp-2.4.21-27.EL How reproducible: Sometimes Steps to Reproduce: See Description. Additional info:
Obviously we'll need more information to go on. Preferably, can you set the machine up with netdump and/or diskdump, and then forcibly crash the machine with alt-sysrq-c when the hang occurs? Short of that, when it's in that state, can you capture the output of alt-sysrq-w?
This bug is filed against RHEL 3, which is in maintenance phase. During the maintenance phase, only security errata and select mission critical bug fixes will be released for enterprise products. Since this bug does not meet that criteria, it is now being closed. For more information of the RHEL errata support policy, please visit: http://www.redhat.com/security/updates/errata/ If you feel this bug is indeed mission critical, please contact your support representative. You may be asked to provide detailed information on how this bug is affecting you.