[SERVER-20889] Introduce means to disable sharding minOpTime recovery Created: 12/Oct/15  Updated: 03/Apr/19  Resolved: 22/Oct/15

Status: Closed
Project: Core Server
Component/s: Sharding
Affects Version/s: None
Fix Version/s: 3.2.0-rc1

Type: Bug Priority: Major - P3
Reporter: Kaloian Manassiev Assignee: Spencer Brody (Inactive)
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
depends on SERVER-20943 Write procedures for restoring a CSRS... Closed
Documented
is documented by DOCS-6435 Document how to disable sharding stat... Closed
Related
is related to SERVER-19934 Ill-timed crash at end of chunk migra... Closed
is related to DOCS-6435 Document how to disable sharding stat... Closed
is related to SERVER-20824 Test for sharding state recovery Closed
Backwards Compatibility: Fully Compatible
Operating System: ALL
Sprint: Sharding B (10/30/15)
Participants:

 Description   

With the introduction of Config Servers as Replica Sets, shard hosts will checkpoint the latest config server opTime after each metadata operation they perform and if necessary on startup will contact the config server's primary in order to recover the minOpTime that they should be using.

This poses problems with restoring a host, which was previously backed up with this recovery information, because it will attempt to connect a potentially defunct config server.

For this reason, we need to either have a separate argument, which allows the recovery information to be ignored or we should start requiring --shardSvr to be specified for all shard servers.



 Comments   
Comment by Andy Schwerin [ 20/Nov/15 ]

Most important documentation change is to update the restore procedures for sharded clusters and single shards, which is DOCS-6597.

Comment by Spencer Brody (Inactive) [ 22/Oct/15 ]

Sharding state recovery can be disabled by starting mongod with --setParameter=recoverShardingState=false

Comment by Githook User [ 22/Oct/15 ]

Author:

{u'username': u'stbrody', u'name': u'Spencer T Brody', u'email': u'spencer@mongodb.com'}

Message: SERVER-20889 Add server parameter for disabling sharding state recovery
Branch: master
https://github.com/mongodb/mongo/commit/14172523b2afcb4b1c3837cd777a5373a3be1441

Comment by Cailin Nelson [ 14/Oct/15 ]

Ah, understood. Thanks.

Comment by Kaloian Manassiev [ 14/Oct/15 ]

If the minOpTime recovery information is not present, this means that this particular shard never executed any metadata changing operation (i.e., move chunk). Don't think of it as cache, but instead as a progress checkpoint.

So this information must be backed up if the node will ever be used as a shard again and patched up on restore to point to the correct config server.

Generated at Thu Feb 08 03:55:36 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.