[SERVER-48537] addShard is not idempotent for retries Created: 02/Jun/20 Updated: 26/Oct/23 |
|
| Status: | Backlog |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Pierlauro Sciarelli | Assignee: | Backlog - Catalog and Routing |
| Resolution: | Unresolved | Votes: | 0 |
| Labels: | oldshardingemea, sharding-csrs-stepdown-only | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||
| Assigned Teams: |
Catalog and Routing
|
||||||||||||
| Operating System: | ALL | ||||||||||||
| Participants: | |||||||||||||
| Linked BF Score: | 14 | ||||||||||||
| Description |
|
From a high level point of view, the addShard method triggered by a _configsvrAddShard command is executing the following steps: 1) Check if the shard exists in config.shards (if yes, return). If there is an interruption between steps 2 and 3, any addShard retry will not execute step 3 resulting in config.databases potentially presenting an inconsistent/incomplete state. |
| Comments |
| Comment by Pierlauro Sciarelli [ 24/Jul/20 ] |
|
I repurposed the description to give some more context as the original text was very sketchy. Here are some possible ways for solving the problem:
|