Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | BUMP 2.2.1r2.2.1 | Dan Pasette | 2012-10-29 | 3 | -3/+3 |
| | |||||
* | post 2.2.1-rc1 | Eric Milkie | 2012-10-19 | 2 | -2/+2 |
| | |||||
* | BUMP 2.2.1-rc1r2.2.1-rc1r2.1.1-rc1 | Eric Milkie | 2012-10-19 | 2 | -2/+2 |
| | |||||
* | SERVER-7374 mrShardedOutput.js make failure much more informative | Greg Studer | 2012-10-19 | 1 | -3/+46 |
| | |||||
* | buildbot auth_slaveok_routing.js need to wait for primary status to avoid ↵ | Greg Studer | 2012-10-19 | 1 | -0/+6 |
| | | | | host spontaneously becoming primary | ||||
* | buildbot sharding_balance4.js turn off balancer during initial load so that ↵ | Greg Studer | 2012-10-19 | 1 | -1/+11 |
| | | | | we are sure to have a chunk diff | ||||
* | buildbot limit_push.js stop the balancer if we're doing manual moves | Greg Studer | 2012-10-19 | 1 | -0/+3 |
| | |||||
* | SERVER-7376 make migration not wait for deletes, no change to default delete ↵ | Greg Studer | 2012-10-18 | 1 | -3/+12 |
| | | | | behavior | ||||
* | Revert "SERVER-7376 make all migration cleanup async unless explicitly ↵ | Greg Studer | 2012-10-18 | 7 | -95/+33 |
| | | | | | | specified otherwise" This reverts commit dfb8e2604409e2879ea975c4c49ba6ffde90d795. | ||||
* | buildbot sharding_migrate_cursor1.js wait for cleanup after data moves. | Greg Studer | 2012-10-17 | 1 | -0/+5 |
| | |||||
* | buildbot shard6.js use separate config server if counting connections to shards | Greg Studer | 2012-10-17 | 1 | -1/+1 |
| | |||||
* | SERVER-7376 make all migration cleanup async unless explicitly specified ↵ | Greg Studer | 2012-10-17 | 7 | -33/+95 |
| | | | | otherwise | ||||
* | SERVER-7376 test fixes for 2.2 | Greg Studer | 2012-10-17 | 15 | -24/+25 |
| | |||||
* | SERVER-7186 Fixed addToSet logging for an element that exists and is not at ↵ | Alberto Lerner | 2012-10-17 | 2 | -6/+10 |
| | | | | the end of the array. | ||||
* | SERVER-7376 better logging in moveChunk, don't swallow error reported to mongos | gregs | 2012-10-17 | 1 | -3/+12 |
| | |||||
* | SERVER-7376 more logging around removeRange | gregs | 2012-10-17 | 1 | -1/+9 |
| | |||||
* | SERVER-4781 Fix style issues. | Alberto Lerner | 2012-10-16 | 3 | -19/+30 |
| | |||||
* | SERVER-7371 Fix same corner case as the last one, but now when applied to ↵ | Alberto Lerner | 2012-10-16 | 2 | -1/+30 |
| | | | | secondary. | ||||
* | SERVER-7371 Remove old comment. | Alberto Lerner | 2012-10-16 | 1 | -1/+0 |
| | |||||
* | SERVER-7371 Fix corner case where an update mod is applied twice. | Alberto Lerner | 2012-10-16 | 2 | -1/+23 |
| | |||||
* | SERVER-4781 If in initial replication, allow a non-strict mod applying policy. | Alberto Lerner | 2012-10-15 | 7 | -30/+129 |
| | |||||
* | SERVER-6925 use the size of the queue in bytes rather than number of objects ↵ | Eric Milkie | 2012-10-12 | 1 | -1/+1 |
| | | | | when calculating cap | ||||
* | SERVER-7186 Fixed unittest impacted by previous commit. | Alberto Lerner | 2012-10-12 | 1 | -1/+1 |
| | |||||
* | SERVER-7186 Fixed the case where logging addToSet would fail dollar sign ↵ | Alberto Lerner | 2012-10-12 | 2 | -8/+31 |
| | | | | verification. | ||||
* | SERVER-7186 Optimized logging of push and addToSet update modifiers. | Alberto Lerner | 2012-10-12 | 4 | -10/+58 |
| | |||||
* | SERVER-7186 Fixed the case where a no-op turned into an object replace on ↵ | Alberto Lerner | 2012-10-12 | 2 | -8/+59 |
| | | | | the secondary. | ||||
* | SERVER-7186 Addressed idempotency violations in update operators. | Alberto Lerner | 2012-10-12 | 5 | -103/+632 |
| | |||||
* | SERVER-6671 end batch early if oplog version change is detected | Eric Milkie | 2012-10-05 | 3 | -2/+43 |
| | |||||
* | SERVER-6671 split lines for easier debugging | Eric Milkie | 2012-10-04 | 2 | -7/+15 |
| | |||||
* | SERVER-6671 add a 'v' version field to each oplog document | Eric Milkie | 2012-10-04 | 5 | -30/+53 |
| | | | | | | | | The version field will allow us to detect the primary's version. We need to know which version because only newer primary oplog streams should prevent a secondary from enforcing unique index constraints in initial sync or recovering states. The version field will also be useful in the future when we want to make schema changes in the oplog. | ||||
* | SERVER-7199 Bump minvalid correctly when batch goes past minvalid | Kristina | 2012-10-04 | 3 | -16/+44 |
| | |||||
* | SERVER-7199 Keep minvalid up-to-date and only write initial op once | Kristina | 2012-10-04 | 2 | -26/+11 |
| | |||||
* | SERVER-7199 Don't reclone docs on 2nd and 3rd oplog applications | Kristina | 2012-10-04 | 4 | -20/+43 |
| | |||||
* | SERVER-7199 Replay oplog a second time before initial syncing indexes | Kristina | 2012-10-04 | 1 | -1/+30 |
| | |||||
* | SERVER-7198 Prevent rollback in inconsistent state | Kristina | 2012-10-03 | 2 | -0/+87 |
| | |||||
* | SERVER-7181 fix build - assert code conflict | Eric Milkie | 2012-10-03 | 1 | -1/+1 |
| | |||||
* | SERVER-7181 Fix error reporting in mongorestore during index build | Randolph Tan | 2012-10-03 | 1 | -5/+13 |
| | |||||
* | SERVER-7238 fix power of 2 allocation with documents > 8mb | Eliot Horowitz | 2012-10-02 | 2 | -4/+11 |
| | |||||
* | SERVER-6671 do not ignore special _id unique index | Eric Milkie | 2012-10-02 | 4 | -13/+24 |
| | |||||
* | SERVER-6671 ignore dups for multikey index insertions during certain ↵ | Eric Milkie | 2012-10-02 | 1 | -2/+3 |
| | | | | replication states | ||||
* | SERVER-6671 add ROLLBACK to the list of states where we suppress duplicate ↵ | Eric Milkie | 2012-10-02 | 1 | -1/+2 |
| | | | | | | | | key enforcement During rollback, there is a stage where we apply all the docs we fetched to undo all the rolled-back operations. This stage may encounter duplicate key errors, since we are fetching the docs from the future and not from the time when they were recorded in the oplog. | ||||
* | SERVER-6671 suppress unique index enforcement during STARTUP2 or RECOVERING ↵ | Eric Milkie | 2012-10-02 | 6 | -15/+30 |
| | | | | | | secondary states Operations involving unique indexes may be repeated during a recovery or initial sync. We need to ignore unique key violations in this situation. This is okay because by the time we reach the end of the oplog stream (and are finally caught up), all the violations should be resolved (as we should end up with an exact copy of the primary). | ||||
* | SERVER-7182 Fix broken compile for mongos | Spencer T Brody | 2012-10-02 | 1 | -1/+1 |
| | |||||
* | SERVER-7182 Call getLastError on the correct database in mongorestore | Spencer T Brody | 2012-10-02 | 6 | -16/+85 |
| | |||||
* | SERVER-7160 Full mongorestore with auth should auth only against the admin db | Spencer T Brody | 2012-10-02 | 3 | -5/+45 |
| | |||||
* | SERVER-7146 buildbot writeback_bulk_insert.js use correct collection for ↵ | Greg Studer | 2012-10-01 | 1 | -6/+7 |
| | | | | last writeback and disable autosplit | ||||
* | SERVER-7146 buildbot writeback_bulk_insert.js fix test for v8 | Greg Studer | 2012-10-01 | 1 | -3/+3 |
| | |||||
* | SERVER-6911 don't run TTL monitor if not in a readable state | Kevin Matulef | 2012-09-27 | 1 | -1/+5 |
| | |||||
* | post 2.2.1-rc0r2.2.1-rc0 | Dan Pasette | 2012-09-22 | 2 | -2/+2 |
| | |||||
* | BUMP 2.2.1-rc0 | Dan Pasette | 2012-09-19 | 2 | -2/+2 |
| |