[SERVER-4071] mongod doesn't detect file system disappearance Created: 14/Oct/11 Updated: 06/Dec/22 Resolved: 15/Nov/16 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Storage |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Minor - P4 |
| Reporter: | Richard Kreuter (Inactive) | Assignee: | Backlog - Storage Execution Team |
| Resolution: | Cannot Reproduce | Votes: | 1 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
Linux (Ubuntu Maverick x86_64). |
||
| Assigned Teams: |
Storage Execution
|
| Participants: |
| Description |
|
It appears that at least under some versions of Linux, if the file system disappears out from under mongod, the server will continue to operate to some degree, with the consequence that data might get written to the wrong locations and so probably lost later. Probably the server ought to abort in these cases, if it's possible to detect them. Reproduction case: – p=30000 ); db.c.find().forEach(printjson)' ); db.c.find().forEach(printjson)' sudo mount -t tmpfs -o size=512M tmpfs "$d"
|
| Comments |
| Comment by Eric Milkie [ 15/Nov/16 ] |
|
Can no longer reproduce this issue. |
| Comment by Richard Kreuter (Inactive) [ 15/Oct/11 ] |
|
No journaling. The reproduction case above leaves no files behind. |
| Comment by Eliot Horowitz (Inactive) [ 15/Oct/11 ] |
|
Is this with journalling? |