Comment 2 for bug 2007643
Revision history for this message
| Alistair Coles (alistair-coles) wrote | #2 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
This bug will manifest predictably in the following circumstances:
1. An object's data file has been created with a timestamp offset. This happens when an object is versioned (the offset if 1) or migrated between policies by the reconciler (the offset is 2).
2. The object subsequently has POST request that creates a meta file (for example a POST to set X-Delete-At)
It would appear that in these circumstances, if that object is included in the scope of a reconstructor job, but the object is already in sync, the ssync protocol will always fail with a 409 and the reconstructor will therefore fail to process the partition.