[SERVER-366] btree exent freeing Created: 16/Oct/09 Updated: 12/Jul/16 Resolved: 29/Jun/10 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Index Maintenance |
| Affects Version/s: | None |
| Fix Version/s: | 1.5.4 |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Eliot Horowitz (Inactive) | Assignee: | Aaron Staple |
| Resolution: | Done | Votes: | 13 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Participants: |
| Description |
|
| Comments |
| Comment by auto [ 08/Jul/10 ] |
|
Author: {'login': 'astaple', 'name': 'Aaron', 'email': 'aaron@10gen.com'}Message: |
| Comment by auto [ 08/Jul/10 ] |
|
Author: {'login': 'astaple', 'name': 'Aaron', 'email': 'aaron@10gen.com'}Message: |
| Comment by Aaron Staple [ 29/Jun/10 ] |
|
I've implemented everything in the description. Let me know if we need anything more. |
| Comment by auto [ 29/Jun/10 ] |
|
Author: {'login': 'astaple', 'name': 'Aaron', 'email': 'aaron@10gen.com'}Message: |
| Comment by Aaron Staple [ 15/Jun/10 ] |
|
FYI looks like there is a higher level guard in place to prevent the the background operation/dropping issue I described above. |
| Comment by auto [ 15/Jun/10 ] |
|
Author: {'login': 'astaple', 'name': 'Aaron', 'email': 'aaron@10gen.com'}Message: |
| Comment by auto [ 15/Jun/10 ] |
|
Author: {'login': 'astaple', 'name': 'Aaron', 'email': 'aaron@10gen.com'}Message: |
| Comment by Eliot Horowitz (Inactive) [ 15/Jun/10 ] |
|
You can make a separate case for that. |
| Comment by auto [ 14/Jun/10 ] |
|
Author: {'login': 'astaple', 'name': 'Aaron', 'email': 'aaron@10gen.com'}Message: |
| Comment by Aaron Staple [ 14/Jun/10 ] |
|
One of the requirements is to ensure the extent is freed on dropIndex. Extent freeing happens now in general, but if there is a background operation in progress it looks like the index will be dropped but its extents won't be freed. Is fixing this something I should work on now? |
| Comment by Aaron Staple [ 04/Mar/10 ] |
|
It might be worthwhile to also spending some time studying how balanced our btrees are under various pathological operation sequences. |
| Comment by Aaron Staple [ 08/Feb/10 ] |
|
We should be careful when doing this to preserve the invariant expected in BtreeBucket::advance() that every bucket has at least one key. (Or alternatively we could make the invariant more flexible). |