[SERVER-12850] Interrupted index builds are not resumed on restart (via createIndexes or ensureIndex) Created: 23/Feb/14  Updated: 10/Dec/14  Resolved: 24/Feb/14

Status: Closed
Project: Core Server
Component/s: Index Maintenance
Affects Version/s: 2.6.0-rc0
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Kamran K. Assignee: Eric Milkie
Resolution: Done Votes: 0
Labels: 26qa
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: File create_indexes_shutdown.js     File ensure_index_shutdown.js    
Issue Links:
Related
Operating System: ALL
Participants:

 Description   

Index builds that are started via the createIndexes command or the ensureIndex shell helper do not seem to resume after the server is shut down. JS tests are attached for both ensureIndex and createIndexes.

Output:

[conn2] index build failed. spec: { v: 1, key: { foo: 1.0 }, name: "bg-shutdown-one", ns: "create.shutdown", background: true } error: 11600 interrupted at shutdown

I also noticed that there are a few IndexCatalog::createIndex() callers that use the default SHUTDOWN_CLEANUP behavior instead of specifying SHUTDOWN_LEAVE_DIRTY. Is that on purpose?


Version: d0135ad0ed90118ad2015283f22b0219c0df3301



 Comments   
Comment by Eric Milkie [ 24/Feb/14 ]

The code as it stands only ensures interrupted index builds are restarted on SECONDARY's as part of replication. This is the only situation where restarting is required to maintain data integrity between PRIMARY and SECONDARY.
The other functionality could be enabled but it currently does not work this way. We should change the documentation to reflect this – I'll take care of this.

Generated at Thu Feb 08 03:29:48 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.