This repository has been archived by the owner on Nov 1, 2022. It is now read-only.
Implement configurable timeout for RPC operations #2171
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
|
Adding a context to the RPCs is 💯 I think this will need to go further though. The problem being addressed (in large part) is that Kubernetes API requests pile up, once they start needing to wait for rate limiting. To be able to abandon those when a request times out, the context must be propagated to those calls (i.e., to the cluster.Cluster methods, then to the Kubernetes client methods). Sadly, the client-go API doesn't support contexts. What I suggest is, for the meantime, is that a context is passed through, and checked before each Kubernetes API call (or anything that occurs in a loop, at least). |
squaremo
reviewed
Jun 20, 2019
hiddeco
force-pushed
the
enhancement/rpc-timeout
branch
from
6d4f77c to
f2402e2
Compare
squaremo
approved these changes
Jun 24, 2019
This assures that in a scenario where for example the Kubernetes API is temporary rate limited, requests that are taking too long get abandoned, rather than piling up. As we have observed excessive amount of goroutines getting spawned in a short amount of time, sometimes resulting in a shutdown of the Flux daemon due to the spike in CPU (and RAM) that comes with it.
hiddeco
force-pushed
the
enhancement/rpc-timeout
branch
from
f2402e2 to
bfdfe31
Compare
squaremo
pushed a commit
that referenced
this pull request
Jun 27, 2019
Implement configurable timeout for RPC operations
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Outdated suggestions cannot be applied.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.
This assures that in a scenario where for example the Kubernetes API
is temporary rate limited, requests that are taking too long get
abandoned, rather than piling up.
As we have observed excessive amount of goroutines getting spawned
in a short amount of time, sometimes resulting in a shutdown of the
Flux daemon due to the spike in CPU (and RAM) that comes with it.