[SERVER-6162] backport of targeted shard ranges instead of broadcast on all queries Created: 21/Jun/12 Updated: 11/Jul/16 Resolved: 26/Jun/12 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | 2.0.7 |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Greg Studer | Assignee: | Tad Marshall |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Operating System: | ALL |
| Participants: |
| Description |
|
We shouldn't ever send queries to shards where the collection does not exist - this can introduce version issues (since version 0 is "special") and cause problems when dropping shards (since the dropped shards are also targeted until refreshed). Proposing backport of f93918db (partial or full) - we previously didn't backport b/c potentially a bit dangerous, but has not caused any problems in master for several months, so it's probably burned-in enough. |
| Comments |
| Comment by auto [ 27/Jun/12 ] |
|
Author: {u'date': u'2012-06-27T10:41:22-07:00', u'email': u'tad@10gen.com', u'name': u'Tad Marshall'}Message: The test jstests/sharding/no_empty_reset.js was using some |
| Comment by Tad Marshall [ 26/Jun/12 ] |
|
Backported to version 2.0.7. |
| Comment by auto [ 26/Jun/12 ] |
|
Author: {u'date': u'2012-06-25T11:10:00-07:00', u'name': u'gregs', u'email': u'greg@10gen.com'}Message: test and fix for zero-version being sent to unneeded shards during scatter/gather Backport based on f93918db8ec15c9de7c884d0a560a185a8ef423d Signed-off-by: Tad Marshall <tad@10gen.com> |