[DOCS-1381] documentation for changing OpLog size suggests doing so on Primary first Created: 11/Apr/13 Updated: 28/Aug/13 Resolved: 16/Aug/13 |
|
| Status: | Closed |
| Project: | Documentation |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Andre de Frere | Assignee: | Bob Grabar |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Participants: | |
| Days since reply: | 10 years, 25 weeks ago |
| Description |
|
The documentation for changing the OpLog size mentions that you should bring down the current Primary then change the size. This is more intrusive than taking down the Secondaries one after the other and changing the oplog size there, as with the Rolling Index rebuild instructions. The docs also mention to shut down the node, rather than step down first. This could lead to a situation where the Secondaries are not up to date with the Primary before the Primary is shut down - if an rs.stepDown() is issued then there will at least be a warning about no nodes being within 10 seconds of the Primary. |
| Comments |
| Comment by auto [ 28/Aug/13 ] |
|
Author: {u'username': u'tychoish', u'name': u'Sam Kleinman', u'email': u'samk@10gen.com'}Message: |
| Comment by auto [ 28/Aug/13 ] |
|
Author: {u'username': u'bgrabar', u'name': u'Bob Grabar', u'email': u'bob.grabar@10gen.com'}Message: Signed-off-by: Sam Kleinman <samk@10gen.com> |
| Comment by Scott Hernandez (Inactive) [ 11/Apr/13 ] |
|
http://docs.mongodb.org/manual/tutorial/change-oplog-size/ These docs should also cover the ramifications of removing all the oplog entries, and provide some guidance and warnings. There is no difference with doing a stepDown nor shutdown as either way you could have rollback if no secondary can catch up in time. The stepdown is better since the window is smaller, but effectively can result in the same issue. |