[SERVER-30544] fsynclock results in corrupt disk snapshot Created: 07/Aug/17 Updated: 09/Oct/17 Resolved: 15/Sep/17 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | WiredTiger |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Dharshan Rangegowda | Assignee: | Mark Agarunov |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Attachments: |
|
| Operating System: | ALL |
| Participants: |
| Description |
|
I am using server 3.2.12/3.4.3. FsyncLock() commands appear to corrupt the disk data such that a snapshot taken at this time does not work. The data files exist but wiredtiger ignores them - so it might be some issue with the on disk catalog I am using latest version of mongodb java driver so it is unrelated to I am able to reproduce the issue with reasonable consistently (doesn't happen every time). Here are my steps The lock and unlock steps appear to work and trace in the log I have attached two files The database "test.test" does not show up in fail.zip. The underlying file fpr this collection is collection-18--3756087474573173486.wt which is present in both the snapshots Using the wt command line tool directly confirms that wiredtiger does not see this table even though the file exists |
| Comments |
| Comment by Mark Agarunov [ 15/Sep/17 ] |
|
Hello dharshanr@scalegrid.net, Thank you for the detailed report. Unfortunately I have not been able to reproduce this behavior following the steps you outlined. If more information comes to light, please let me know and we can reopen this ticket. Thanks, |