[SERVER-34322] Best effort ssv to recipient shard will never succeed for the first time Created: 04/Apr/18 Updated: 29/Oct/23 Resolved: 11/Jan/22 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | 3.7.3 |
| Fix Version/s: | 5.3.0 |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Randolph Tan | Assignee: | [DO NOT USE] Backlog - Sharding EMEA |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Assigned Teams: |
Sharding EMEA
|
||||||||
| Backwards Compatibility: | Fully Compatible | ||||||||
| Operating System: | ALL | ||||||||
| Participants: | |||||||||
| Description |
|
if the recipient shard has no version initialized for the namespace. This is because the ssv request sent has authoritative: false and will trigger this error on the recipient shard. Another bug is that it is being sent with the collection version instead of the shard version. |
| Comments |
| Comment by Sergi Mateo Bellido [ 11/Jan/22 ] |
|
Closing it after internal discussion. The issue is already fixed on versions >= 5.3. |
| Comment by Sergi Mateo Bellido [ 11/Jan/22 ] |
|
As Kal mentioned above, we don't have this problem in the new migration protocol: we send a command to the recipient shard to refresh the filtering metadata and release the critical section during the commit phase of the moveChunk. |
| Comment by Kaloian Manassiev [ 12/Oct/21 ] |
|
I believe that this ticket will no longer be necessary after the work in this Epic. |