[SERVER-2116] Indexes don't always get carried over to all shards Created: 18/Nov/10 Updated: 17/Mar/11 Resolved: 16/Jan/11 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | 1.6.2 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Giri Iyengar | Assignee: | Unassigned |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
Centos 64 bit, Amazon EC2 |
||
| Operating System: | Linux |
| Participants: |
| Description |
|
In a sharded configuration, we have noticed that indexes don't always get carried over to all the shards. This seems to happen more frequently when we do a remove on a collection that has chunks on multiple shards and then start filling up the collection again. When the chunk size gets larger than what the primary shard can handle, the data gets sent to the secondary shard but it doesn't seem to have the index. Workaround is to manually build the indexes on all shards upon discovery excessive CPU usage. |
| Comments |
| Comment by Eliot Horowitz (Inactive) [ 16/Jan/11 ] |
|
We can't reproduce this. |