[SERVER-27297] DatabaseCloner should log CollectionCloner errors on failure Created: 06/Dec/16  Updated: 05/Apr/17  Resolved: 07/Feb/17

Status: Closed
Project: Core Server
Component/s: Replication
Affects Version/s: None
Fix Version/s: 3.5.3

Type: Improvement Priority: Major - P3
Reporter: Judah Schvimer Assignee: Judah Schvimer
Resolution: Done Votes: 0
Labels: neweng
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
Backwards Compatibility: Fully Compatible
Sprint: Repl 2017-01-23, Repl 2017-02-13
Participants:
Linked BF Score: 0

 Description   

When the CollectionCloner fails, the status is placed into _failedNamespaces here. However, we never do anything with _failedNamespaces besides check and print its size. We should print some, if not all, of the errors in the CollectionCloner, especially now that we don't do any collection cloning in parallel.

Another solution would be to print CollectionCloner errors at a log level greater than 1: https://github.com/mongodb/mongo/blob/726cafd713c7333640f8458ec9808ed4f678e3a7/src/mongo/db/repl/collection_cloner.cpp#L542



 Comments   
Comment by Githook User [ 07/Feb/17 ]

Author:

{u'username': u'judahschvimer', u'name': u'Judah Schvimer', u'email': u'judah@mongodb.com'}

Message: SERVER-27297 always log CollectionCloner failure statuses
Branch: master
https://github.com/mongodb/mongo/commit/ad9f558ae842b34a5d4a38a87ac603f1aabab560

Comment by Benety Goh [ 06/Dec/16 ]

_failedNamespaces is a leftover from an earlier version of DatabaseCloner where DatabaseCloner would attempt clone every collection in the database before aggregating the results for the caller.

Generated at Thu Feb 08 04:14:46 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.