Comment 7 for bug 1583144
Revision history for this message
| clayg (clay-gerrard) wrote Re: object-reconstructor index out of range | #7 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
35a32c5
(Get the code!)
OIC, yeah you should have wiped the disks before deploying new rings. This sort of thing happens in test/dev.
Now, you're looking at a dark data problem. If you clean up those parts so they don't disrupt replication things should be able to move forward. The remaining data in the overlapping parts won't *hurt* anything - but there's a good chance that data is not useful and might not get cleaned up.
It really depends on your situation. You could either write a script to troll the objects datadirs and validate the objects are accessible and listed in their containers, then decide what to do with the data that doesn't match. Or you could wipe the whole thing and start over (reloading any data that you want in the cluster from their sources).
GL!