[SERVER-4687] need to check draining and potentially other shard status before auto-migrates Created: 16/Jan/12 Updated: 10/Dec/14 Resolved: 12/Aug/14 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Greg Studer | Assignee: | Randolph Tan |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Operating System: | ALL | ||||||||
| Participants: | |||||||||
| Description |
|
While we reload shard information before running the balancer, we don't do the same for auto-migrates, and the logic for the target shard is different from the balancer policy. There's also a secondary race condition in that it's possible for shard status to change after shard info reload and before the migration is triggered, but this is a much rarer event and will generally be caught while waiting for draining. |
| Comments |
| Comment by Randolph Tan [ 12/Aug/14 ] |
|
|