Skip to content
This repository was archived by the owner on Feb 1, 2021. It is now read-only.

[experimental] Simple container rescheduling on node failure#1578

Merged
abronan merged 5 commits intodocker-archive:masterfrom
aluzzardi:rescheduling
Jan 12, 2016
Merged

[experimental] Simple container rescheduling on node failure#1578
abronan merged 5 commits intodocker-archive:masterfrom
aluzzardi:rescheduling

Conversation

@vieux
Copy link
Copy Markdown
Contributor

@vieux vieux commented Jan 4, 2016

See #1488 for details

When a node goes down, swarm will try to reschedule container on another machine.

Depends on moby/moby#19001 for IP stability.
Depends on #1569 for proper cleanup of old containers.

@vieux vieux added the area/API label Jan 4, 2016
@vieux vieux self-assigned this Jan 4, 2016
@vieux vieux added this to the 1.1.0 milestone Jan 4, 2016
dongluochen
dongluochen reviewed Jan 7, 2016
View reviewed changes
cluster/config.go Outdated
Copy link
Copy Markdown
Contributor

@dongluochen dongluochen Jan 7, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: update comment.

@vieux vieux force-pushed the rescheduling branch 2 times, most recently from ea7afd3 to b72d27f Compare January 8, 2016 22:17
dongluochen
dongluochen reviewed Jan 8, 2016
View reviewed changes
cluster/watchdog.go Outdated
Copy link
Copy Markdown
Contributor

@dongluochen dongluochen Jan 8, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Any reason this trigger whole cluster reschedule, instead of just this engine? I think there could be a lot of events for a big cluster. Isn't just checking this container enough?

@vieux vieux force-pushed the rescheduling branch 2 times, most recently from 9352993 to 0220f36 Compare January 8, 2016 23:02
@mavenugo mavenugo mentioned this pull request Jan 9, 2016
Merged
@mavenugo
Copy link
Copy Markdown
Contributor

mavenugo commented Jan 9, 2016

@vieux we are working on the forced cleanup fix in moby/libnetwork#862.
you can use my private docker branch (https://github.com/mavenugo/docker/tree/epcleanup), where this is integrated fully with docker engine API which swarm can use to reschedule container reclaiming the locked resources. Would you like to give it a try in parallel as we make progress in getting this PRs merged in multiple repos ?

@mavenugo mavenugo mentioned this pull request Jan 9, 2016
Merged
aluzzardi and others added 3 commits January 11, 2016 15:59
@aluzzardi @vieux
cluster: Support multiple event handlers.
56941d0 
Signed-off-by: Andrea Luzzardi <aluzzardi@gmail.com>
@aluzzardi @vieux
Add support for container rescheduling on node failure.
13f6021 
Add rescheduling integration tests.

Signed-off-by: Andrea Luzzardi <aluzzardi@gmail.com>
@vieux
add doc
78008f4 
fix tests and keep swarm id
remove duplicate on node reconnect
explicit failure

Signed-off-by: Victor Vieux <vieux@docker.com>
@vieux
improve eventHandlers locking
a2018c1 
Signed-off-by: Victor Vieux <vieux@docker.com>
@vieux vieux changed the title [experimental] [WIP] Container Rescheduling on node failure [experimental] Simple container rescheduling on node failure Jan 12, 2016
moxiegirl
moxiegirl reviewed Jan 12, 2016
View reviewed changes
experimental/README.md~ Outdated
Copy link
Copy Markdown

@moxiegirl moxiegirl Jan 12, 2016

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@vieux This looks like it shouldn't be here it is a backup file

@vieux
move doc to experimental/
74dfe8b 
Signed-off-by: Victor Vieux <vieux@docker.com>
@moxiegirl
Copy link
Copy Markdown

moxiegirl commented Jan 12, 2016

LGTM

@dongluochen
Copy link
Copy Markdown
Contributor

dongluochen commented Jan 12, 2016

LGTM.

@abronan
Copy link
Copy Markdown
Contributor

abronan commented Jan 12, 2016

LGTM

abronan added a commit that referenced this pull request Jan 12, 2016
@abronan
Merge pull request #1578 from aluzzardi/rescheduling
e121338 
[experimental] Simple container rescheduling on node failure
@abronan abronan merged commit e121338 into docker-archive:master Jan 12, 2016
@vieux vieux deleted the rescheduling branch January 12, 2016 23:00
@abronan abronan mentioned this pull request Jan 12, 2016
Closed
ChristianKniep pushed a commit to ChristianKniep/swarm that referenced this pull request Jul 27, 2017
@abronan
Merge pull request docker-archive#1578 from aluzzardi/rescheduling
15a3e1e 
[experimental] Simple container rescheduling on node failure
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Reviewers

No reviews

Assignees

@vieux vieux

Labels

area/API priority/P2 status/1-design-review

Projects

None yet

Milestone

1.1.0

Development

Successfully merging this pull request may close these issues.

8 participants

@vieux @mavenugo @moxiegirl @dongluochen @abronan @claudiu-coman @GordonTheTurtle @aluzzardi