[SOLVED] Help...My Life is a Country Song and I've Broken My Zimbra :(

[jholder]

Chewie,
You're an NE customer. Can you please Pvt me login creditials and add me to your IM list? Yahoo: z_jholder

I'll take a look and see if I can recover it.

john

[mmorse]

Ya, I was afraid of that (the RHEL5 install killing the db)
Check your through your .saveconfig, any ldap.bak's, openldap-data/ldap.bak, etc (hopefully there's a valid one somewhere *crosses fingers*)

[jholder]

Originally Posted by mmorse

Ya, I was afraid of that (the RHEL5 install killing the db)
Check your through your .saveconfig, any ldap.bak's, openldap-data/ldap.bak, etc (hopefully there's a valid one somewhere *crosses fingers*)

I disagree

The FP error is on slapcat, which is the binary. I bet his data is fine. We just need to figure out why slapcat is failing.

Of course it's possible that slapcat is dying because of the data file, but I don't think so.

[mmorse]

John's the man

[Chewie71]

Originally Posted by mmorse

John's the man

Certainly is!! I gave him access to my server and he got it straightened out. Had to move some binaries around but he finally got it going.

KUDOS John!! Thanks for taking time out of your schedule to give me a hand.

You Zimbra guys are awesome. Easily THE best company I deal with on a regular basis.

Thanks,
Matt

[mmorse]

Very happy to mark this thread solved!
Bet you'll never forget a backup again eh?
For FOSS I'm a fan of rsnapshot, though I've listed some other methods here: Snapshots - Open Source Edition Backup Procedure - Zimbra :: Wiki
Hope everything falls in to place (not literally!) with the rest of your week

[routerguy]

Wish the fix were documented here, as I've gone and done the *exact* same thing! I'm a NE customer so I guess I'll open a case, unless somebody has the fix handy?

[rrust]

I did the exact same thing. Then to make matters worse, I did the restore from an older backup. 300 accounts, 120gb, took a while. then had to go back and mailbox restore from the newer backup

anyhow, restoring from backup was my solution...then running the update (the correct centos4 instead of centos5)

If I didnt love centos so much I would bite the bullet and pay $1k a year for rh...