[DOCS-12096] Primaries journal much more frequently Created: 28/Sep/18 Updated: 29/Oct/23 Resolved: 12/Nov/19 |
|
| Status: | Closed |
| Project: | Documentation |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Kelsey Schubert | Assignee: | Kay Kim (Inactive) |
| Resolution: | Fixed | Votes: | 1 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||
| Participants: | |||||||||||||||||
| Days since reply: | 4 years, 13 weeks, 1 day ago | ||||||||||||||||
| Epic Link: | DOCSP-1769 | ||||||||||||||||
| Description |
Descriptionhttps://docs.mongodb.com/manual/core/journaling/#journaling-process currently suggests that we journal every 50 milliseconds. On a primary this does not hold true, as secondaries are only allowed to read writes that have been journaled on the Primary. As a result, its reasonable to expect that the journal constantly be flushed to disk. cc geert.bosch Scope of changesImpact to Other DocsMVP (Work and Date)Resources (Scope or Design Docs, Invision, etc.) |
| Comments |
| Comment by Githook User [ 12/Nov/19 ] |
|
Author: {'name': 'Kay Kim', 'username': 'kay-kim', 'email': 'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 12/Nov/19 ] |
|
Author: {'name': 'Kay Kim', 'username': 'kay-kim', 'email': 'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 12/Nov/19 ] |
|
Author: {'username': 'kay-kim', 'email': 'kay.kim@10gen.com', 'name': 'Kay Kim'}Message: |
| Comment by Githook User [ 12/Nov/19 ] |
|
Author: {'name': 'Kay Kim', 'username': 'kay-kim', 'email': 'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 12/Nov/19 ] |
|
Author: {'name': 'Kay Kim', 'username': 'kay-kim', 'email': 'kay.kim@10gen.com'}Message: |
| Comment by Githook User [ 12/Nov/19 ] |
|
Author: {'name': 'Kay Kim', 'username': 'kay-kim', 'email': 'kay.kim@10gen.com'}Message: |
| Comment by Eric Milkie [ 13/May/19 ] |
|
Note that the Replicate before Journal project PM-1274 will change this logic. |
| Comment by Geert Bosch [ 20/Dec/18 ] |
|
When a secondary is actively waiting for new oplog entries (it has an active getMore on a tailable cursor on the oplog collection), it will cause immediate flushes. Since The journal frequency is given by the journalCommitInterval server parameter, and defaults to 100ms. However, on replica sets, both the regular journal flushing and the oplog manager's journal flushing happen every 100ms. So you could expect to see journal writes on average at least once every 50ms, even if there are no oplog readers. Now, this is an implementation detail that I hope will go away soon (we want to support secondaries reading the oplog before it has been flushed), so I think it's fine to say every 100ms. |