Merged
Conversation
Member
|
@dperny doesn't build |
A simple but old error has recently become evident. Due to the fact that we read an object and then write it back across the boundaries of a transaction, it is possible for the task object to have changed in between transactions. This would cause the attempt to write out the old task to suffer an "Update out of sequence" error. This fix simply reads the latest version of the task back out within the boundary of a transaction to avoid the race. Signed-off-by: Drew Erny <drew.erny@docker.com>
6b571a4 to
d68ac46
Compare
Collaborator
Author
|
@tonistiigi it builds now, it's just failing on a known-flaky test (fixing this test is on my TODO list) |
tonistiigi
approved these changes
Jul 16, 2019
| // the task may have changed for some reason in the meantime | ||
| // since we read it out, so we need to get from the store again | ||
| // within the boundaries of a transaction | ||
| latestTask := store.GetTask(tx, t.ID) |
Contributor
There was a problem hiding this comment.
Two questions:
-
I forget, but is it possible for the task to change from this point on until the update on line 70?
-
If the update fails because of out of sequence, would it be better to simply retry?
thaJeztah
added a commit
to thaJeztah/docker
that referenced
this pull request
Sep 12, 2019
…3 branch) full diff: moby/swarmkit@4fb9e96...bbe3418 changes included: - moby/swarmkit#2889 [19.03 backport] Fix update out of sequence and increase max recv gRPC message size for nodes and secrets Which relates to - moby#39531 integration-cli: fix swarm tests flakiness - docker-archive#345 [19.03 backport] integration-cli: fix swarm tests flakiness And includes backports of - moby/swarmkit#2808 Fix flaky tests - moby/swarmkit#2866 Swap gometalinter for golangci-lint - moby/swarmkit#2869 Increase max recv gRPC message size to initialize connection broker - related / similar to moby#38103 / docker-archive#102 cluster: set bigger grpc limit for array requests - related / similar to moby#39306 Increase max recv gRPC message size for nodes and secrets - fixes moby/swarmkit#2733 Error generated when messages size is too big - moby/swarmkit#2870 Fix update out of sequence Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
docker-jenkins
pushed a commit
to docker-archive/docker-ce
that referenced
this pull request
Sep 12, 2019
…3 branch) full diff: moby/swarmkit@4fb9e96...bbe3418 changes included: - moby/swarmkit#2889 [19.03 backport] Fix update out of sequence and increase max recv gRPC message size for nodes and secrets Which relates to - moby/moby#39531 integration-cli: fix swarm tests flakiness - docker-archive/engine#345 [19.03 backport] integration-cli: fix swarm tests flakiness And includes backports of - moby/swarmkit#2808 Fix flaky tests - moby/swarmkit#2866 Swap gometalinter for golangci-lint - moby/swarmkit#2869 Increase max recv gRPC message size to initialize connection broker - related / similar to moby/moby#38103 / docker-archive/engine#102 cluster: set bigger grpc limit for array requests - related / similar to moby/moby#39306 Increase max recv gRPC message size for nodes and secrets - fixes moby/swarmkit#2733 Error generated when messages size is too big - moby/swarmkit#2870 Fix update out of sequence Signed-off-by: Sebastiaan van Stijn <github@gone.nl> Upstream-commit: f7dbee3eeaa1dd218116f85b8f60361acbd5b214 Component: engine
This was referenced Sep 25, 2019
thaJeztah
added a commit
to thaJeztah/docker
that referenced
this pull request
Oct 9, 2019
…v18.09) full diff: moby/swarmkit@142a737...5c86095 - moby/swarmkit#2892 [18.09 backport] Remove hardcoded IPAM config subnet value for ingress network - backport of moby/swarmkit#2890 Remove hardcoded IPAM config subnet value for ingress network - fixes [ENGORC-2651](https://docker.atlassian.net/browse/ENGORC-2651) - moby/swarmkit#2836 [18.09 backport] Switch to go 1.11 - backport of moby/swarmkit#2752 Switch to go 1.11 - moby/swarmkit#2901 [18.09 backport] Bump to golang 1.12.9 - backport of moby/swarmkit#2880 Bump to golang 1.12.9 - moby/swarmkit#2900 [18.09 backport] Fix update out of sequence and increase max recv gRPC message size for nodes and secrets - backport of moby/swarmkit#2762 Increased wait time on test utils WaitForCluster and WatchTaskCreate - backport of moby/swarmkit#2771 Allow using Configs as CredentialSpecs - **second commit only** (attempt to fix weirdly broken tests) - backport of moby/swarmkit#2808 Fix flaky tests - backport of moby/swarmkit#2866 Swap gometalinter for golangci-lint - backport of moby/swarmkit#2869 Increase max recv gRPC message size to initialize connection broker - related / similar to moby#38103 / docker-archive#102 cluster: set bigger grpc limit for array requests - related / similar to moby#39306 Increase max recv gRPC message size for nodes and secrets - fixes moby/swarmkit#2733 Error generated when messages size is too big - backport of moby/swarmkit#2870 Fix update out of sequence Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
docker-jenkins
pushed a commit
to docker-archive/docker-ce
that referenced
this pull request
Oct 23, 2019
…v18.09) full diff: moby/swarmkit@142a737...5c86095 - moby/swarmkit#2892 [18.09 backport] Remove hardcoded IPAM config subnet value for ingress network - backport of moby/swarmkit#2890 Remove hardcoded IPAM config subnet value for ingress network - fixes [ENGORC-2651](https://docker.atlassian.net/browse/ENGORC-2651) - moby/swarmkit#2836 [18.09 backport] Switch to go 1.11 - backport of moby/swarmkit#2752 Switch to go 1.11 - moby/swarmkit#2901 [18.09 backport] Bump to golang 1.12.9 - backport of moby/swarmkit#2880 Bump to golang 1.12.9 - moby/swarmkit#2900 [18.09 backport] Fix update out of sequence and increase max recv gRPC message size for nodes and secrets - backport of moby/swarmkit#2762 Increased wait time on test utils WaitForCluster and WatchTaskCreate - backport of moby/swarmkit#2771 Allow using Configs as CredentialSpecs - **second commit only** (attempt to fix weirdly broken tests) - backport of moby/swarmkit#2808 Fix flaky tests - backport of moby/swarmkit#2866 Swap gometalinter for golangci-lint - backport of moby/swarmkit#2869 Increase max recv gRPC message size to initialize connection broker - related / similar to moby/moby#38103 / docker-archive/engine#102 cluster: set bigger grpc limit for array requests - related / similar to moby/moby#39306 Increase max recv gRPC message size for nodes and secrets - fixes moby/swarmkit#2733 Error generated when messages size is too big - backport of moby/swarmkit#2870 Fix update out of sequence Signed-off-by: Sebastiaan van Stijn <github@gone.nl> Upstream-commit: e06f07ef337ab890f211397d6b408b75a2512dc5 Component: engine
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
- What I did
A simple but old error has recently become evident. Due to the fact that we read an object and then write it back across the boundaries of a transaction, it is possible for the task object to have changed in between transactions. This would cause the attempt to write out the old task to suffer an "Update out of sequence" error.
- How I did it
This fix simply reads the latest version of the task back out within the boundary of a transaction to avoid the race.
- How to test it
This was caught by swarm integration tests in the engine, but I'm unsure of how to write a test for this race condition that would have consistent behavior.