I had 2 hard drives fail this morning on a RedHat 7.2 server. (Yes, I know, ancient.) The users are froathing at the mouth about this old thing.

I've replaced one of he drives, managed to run fsck on / and everything else, repaired about a million files, corrected a ton of inodes, etc. I got into the sh prompt and restored an old version of the passwd file. I think I've cleared up most stuff... BUT, when I try to boot it gets all the way to 'Starting opcagt:' and fails hardcore. The error "Assertion failure in journal_forget_Rsmp_726ff55a() at transaction.c:1227: "!jh->b_commited_data"

Is there any way to recover from this? If I can just get the system up and semi-functional I can do a restore from tape.

Do you happen to use OpenView in your environment? I've seen that opcagt thing in conjunction with OV before. If so, you can just remove the startup statements for it, and it should continue to boot.

Another alternative is to build a new system, with an up-to-date distro, and do the file restore to it, solving two problems (having not only an old system, but a dead one too).

Thank you!! I removed the openview from startup and it booted the rest of the way. Now here's hoping the system is functional for a restore to kick off!

You're very welcome...glad it helped. What do you use to back up? If you're using Tivoli, you should be able to just do a bare load of the client, then modify the dsm.opt and .sys files, giving the new system the same name as the old. The client should just fire up, and let you restore. I don't know about other network-based backup solutions, but it should be about the same thing.

Good luck.....