[SERVER-20626] downgrading and upgrading protocolVersion without shutting down perf issue Created: 24/Sep/15 Updated: 07/Oct/15 Resolved: 07/Oct/15 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Replication |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Eric Milkie | Assignee: | Siyuan Zhou |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Operating System: | ALL |
| Sprint: | RPL A (10/09/15) |
| Participants: |
| Description |
|
If you reconfig a replica set from protocolVersion 1 to protocolVersion 0 and then reconfig again from protocolVersion 0 to protocolVersion 1 without shutting down nodes in the interim, w:majority writes start to fail. This may be due to updatePosition commands failing. |
| Comments |
| Comment by Siyuan Zhou [ 07/Oct/15 ] |
|
Adding another upgrade test case after downgrade in the downgrade/upgrade JS test doesn't reproduce this issue. Perhaps, before the second upgrade, not all nodes have observed the previous downgrade? If that's the case, we should add a helper in shell or pre-check in mongod reconfig to guarantee it never happens. |