#50 Issue closed: missing initrd rebuild in RHEL5¶

Labels: bug

wdpoorte opened issue at 2012-03-30 11:49:¶

When recovering on different hardware with a different scsi controller, the recoverd system will not find the root volume group and stop with a kernel panic.

Probably a rebuild of the initrd is needed.

jhoekx commented at 2012-03-30 12:05:¶

This is planned for a future release and tracked in issue #45.

For now, you can add a custom script to rebuild it in your local.conf. See this excerpt from default.conf.

# NOTE: The scripts can be defined as an array to better handly spaces in parameters. The
# scripts are called like this: eval "${PRE_RECOVERY_SCRIPT[@]}"
# call this after rear did everything in the recover workflow. /mnt/local refers to the recovered system
POST_RECOVERY_SCRIPT=

gdha commented at 2012-04-04 14:49:¶

we added the code into rear to fix this on RHEL. However, it is still untested. We hope to get this tested tomorrow...

dagwieers commented at 2012-04-05 14:06:¶

We need more feedback on this if we want to get v1.13 out of the door.

wdpoorte commented at 2012-04-06 12:02:¶

For me it works on RHEL 5.8. No more kernel panics after recovery.

dagwieers commented at 2012-04-06 19:08:¶

Ok, let's close this one :-) Thanks for the feedback.

If someone notices a problem with the implementation, please re-open this ticket.

---

[Export of Github issue for rear/rear.]