[SERVER-26848] Exit catchup mode when not syncing more data Created: 31/Oct/16  Updated: 10/May/18  Resolved: 21/Apr/17

Status: Closed
Project: Core Server
Component/s: Replication
Affects Version/s: None
Fix Version/s: 3.4.6, 3.5.7

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

Issue Links:
Backports
Documented
Related
related to SERVER-27499 Race condition in SyncSourceResolver ... Backlog
related to SERVER-28751 Allow stepDown command to work agains... Closed
is related to SERVER-28969 Wait until the voter has the latest o... Closed
is related to SERVER-26847 Tie threshold of how caught up we mus... Closed
Backwards Compatibility: Minor Change
Backport Requested:
v3.4
Sprint: Repl 2017-01-23, Repl 2017-02-13, Repl 2017-03-06, Repl 2017-03-27, Repl 2017-04-17, Repl 2017-05-08
Participants:
Linked BF Score: 0

 Description   

Currently we expose one parameter to the replica set config - catchUpTimeoutMillis (defaults to 2 seconds) - which controls how long to stay in catchup mode after winning an election.

We should exit catchup mode when the primary finds itself the most up-to-date after refreshing heaertbeats.



 Comments   
Comment by Githook User [ 16/Jun/17 ]

Author:

{u'username': u'visualzhou', u'name': u'Siyuan Zhou', u'email': u'siyuan.zhou@mongodb.com'}

Message: SERVER-26848 Remove dead code of primary catch-up.

(cherry picked from commit e4f20f24ddbb9cea2255df6fe0cfd34ddeb0263e)
Branch: v3.4
https://github.com/mongodb/mongo/commit/133978ed116d3852bb1b8557cc8603b7f6045947

Comment by Githook User [ 16/Jun/17 ]

Author:

{u'username': u'visualzhou', u'name': u'Siyuan Zhou', u'email': u'siyuan.zhou@mongodb.com'}

Message: SERVER-26848 Exit catchup mode when not syncing more data.

(cherry picked from commit 4680351e3fe6f8f47c04440f1c5d1232a4ab7b2d)
Branch: v3.4
https://github.com/mongodb/mongo/commit/7f509b09e3b17a4e6627fdbd9a46d12611745941

Comment by Githook User [ 21/Apr/17 ]

Author:

{u'username': u'visualzhou', u'name': u'Siyuan Zhou', u'email': u'siyuan.zhou@mongodb.com'}

Message: SERVER-26848 Remove dead code of primary catch-up.

This reverts commit fac33fe5a6814169c9c6131d80f1b325c74647da.
Branch: master
https://github.com/mongodb/mongo/commit/e4f20f24ddbb9cea2255df6fe0cfd34ddeb0263e

Comment by Githook User [ 21/Apr/17 ]

Author:

{u'username': u'visualzhou', u'name': u'Siyuan Zhou', u'email': u'siyuan.zhou@mongodb.com'}

Message: SERVER-26848 Exit catchup mode when not syncing more data.

This reverts commit c08590a6ac9dc54c9d910822d47ea17140b56f89.
Branch: master
https://github.com/mongodb/mongo/commit/4680351e3fe6f8f47c04440f1c5d1232a4ab7b2d

Comment by Githook User [ 20/Apr/17 ]

Author:

{u'username': u'guoyr', u'name': u'Robert Guo', u'email': u'robert.guo@10gen.com'}

Message: Revert "SERVER-26848 Exit catchup mode when not syncing more data."

This reverts commit d0c851e2f4bfea514e22c97af1838640d2849a8c.
Branch: master
https://github.com/mongodb/mongo/commit/c08590a6ac9dc54c9d910822d47ea17140b56f89

Comment by Githook User [ 20/Apr/17 ]

Author:

{u'username': u'guoyr', u'name': u'Robert Guo', u'email': u'robert.guo@10gen.com'}

Message: Revert "SERVER-26848 Remove dead code of primary catch-up."

This reverts commit 7109d453e5a264ba77093b9b068d1eaa056ec837.
Branch: master
https://github.com/mongodb/mongo/commit/fac33fe5a6814169c9c6131d80f1b325c74647da

Comment by Githook User [ 19/Apr/17 ]

Author:

{u'username': u'visualzhou', u'name': u'Siyuan Zhou', u'email': u'siyuan.zhou@mongodb.com'}

Message: SERVER-26848 Remove dead code of primary catch-up.
Branch: master
https://github.com/mongodb/mongo/commit/7109d453e5a264ba77093b9b068d1eaa056ec837

Comment by Githook User [ 19/Apr/17 ]

Author:

{u'username': u'visualzhou', u'name': u'Siyuan Zhou', u'email': u'siyuan.zhou@mongodb.com'}

Message: SERVER-26848 Exit catchup mode when not syncing more data.
Branch: master
https://github.com/mongodb/mongo/commit/d0c851e2f4bfea514e22c97af1838640d2849a8c

Comment by Spencer Brody (Inactive) [ 13/Dec/16 ]

If we make the time spent applying operations during catchup infinite, we need a plan for what to do if we lose connectivity to the sync source for any reason. I propose that in that case, we spend the node selection timeout amount of time again re-searching for a node to sync from and continue where we left off.

Generated at Thu Feb 08 04:13:23 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.