The drive I had the bulk of my Zimbra VM on has just decided it wants to develop some bad sectors. Funfunfun.
As quite a bit has happened since I last took a backup this morning [an upgrade and some important mail coming in] I'm trying to rescue the current system rather than rolling back to the last backup. Currently dd_rescude reports only 8Kb of bad reads so far [~30Gb of 60Gb partition copied so far], so if I'm very luck and those couple of blocks do not contain a critical bit of file-system structure I might get away with it.
After starting the VM in single user mode to run fsck on all the affected volumes, are there any useful commands I can run or other things to look at to double-check there are no serious detectable errors in the LDAP records the database, or other parts of Zimbra, before I let it try to start up properly?
I'll probably do a reinstall, restore from backup, and migrate missing data via imapsync later - but for now hopefully this will be "safe enough" and get access to stuff that cam in this afternoon...
Why do these things always happen on a gloriously pleasant summer day when I'd rather be anywhere but in front of a broken machine?!