This repository has been archived by the owner on Jul 1, 2023. It is now read-only.
Check for "active" status #233
Merged
Conversation
This file contains 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
wadells
commented
May 28, 2020
a-palchikov
reviewed
May 29, 2020
r0mant
reviewed
May 29, 2020
This was referenced Jun 2, 2020
wadells
added a commit
to wadells/robotest
that referenced
this pull request
Jun 3, 2020
Per @a-palchikov's advice here: gravitational#233 (comment)
wadells
commented
Jun 3, 2020
wadells
commented
Jun 3, 2020
wadells
force-pushed
the
active-status
branch
2 times, most recently
from
ee5a6bc
to
67c8aeb
Compare
a-palchikov
reviewed
Jun 4, 2020
|
Testing updated with an install that includes 0809da1. |
a-palchikov
approved these changes
Jun 4, 2020
These do not need the lenient 30 minute status timeout. Splitting them out allows independent tuning and opens the status timeout up for refactoring without breaking these code paths.
Previously the code shared one context for two separate ssh command invocations. Using independent contexts gives more control as to how long each command should take.
Previously, these operations all shared the 30 minute status timeout which is far too much time (especially when we want to fail fast during debugging). These new independent timeouts are my best estimate as a overly cautious window for the operations.
Fixes gravitational#225. This will help robotest catch issues when Gravity is not 'degraded' but is also not healthy, such as gravitational/gravity#1523.
This allows unit testing of the status validation function as well as plugging in different status validation functions (e.g. if a test case wants to expect that a cluster is degraded). Contributes to gravitational#225.
This is consistent with the language Gravity uses.
During testing I discovered that robotest was proceeding with upgrades before the cluster was ready. I captured the status json, added it as a test case, and updated the check to mark this status as invalid. Contributes to gravitational#225.
Per @a-palchikov's advice here: gravitational#233 (comment)
As @a-palchikov suggested here: https://github.com/gravitational/robotest/pull/233/files/67c8aeb172d284a8c0439f489c48ae32c4ff1581#r435196137 Using a strongly typed value channel and errgroup makes this code easier to understand and considerably more robust.
This was referenced Jun 8, 2020
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Description
This change introduces more rigorous status checking in all suite tests cases. Specifically, we go from
state != degraded(30m timeout) tostate = active && system_status = Running(5m timeout).Type of change
I mention breaking because the 30m -> 5m timeout may break some of the tests cases we don't run presently. It should be safe for all currently run test cases, because they wait for operations to complete before querying status.
Linked tickets and other PRs
Fixes #225.
Heavily rewrites the code introduced in #136.
TODOs
Manual testing
Testing done
Added unit tests for status validations.
Logs from end-to-end runs: