Skip to content
This repository has been archived by the owner on Feb 18, 2021. It is now read-only.

Log metadata queries with high latency #317

Merged
merged 1 commit into from Nov 1, 2017
Merged

Log metadata queries with high latency #317

merged 1 commit into from Nov 1, 2017

Conversation

kirg
Copy link
Contributor

@kirg kirg commented Oct 30, 2017

No description provided.

@kirg kirg requested review from kobeyang and datoug October 30, 2017 21:37
@coveralls
Copy link

coveralls commented Oct 31, 2017

Coverage Status

Coverage decreased (-0.5%) to 66.845% when pulling 0c67f45 on metametrics-log into 0752a0c on master.

@@ -29,6 +32,11 @@ import (
"github.com/uber/tchannel-go/thrift"
)

const (
pointQueryHighLatencyThreshold = 3 * time.Second
listQueryHighLatencyThreshold = 8 * time.Second
Copy link
Contributor

Choose a reason for hiding this comment

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

Seems too high. How about 1 second for point queries and 5 seconds for list queries?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

as discussed offline .. we'll go with these values for now; they are 'high' thresholds, but this is to prevent swamped with logs when there is an ongoing metadata issue. we will experiment to see whether we need to further refine this with one of the following:1

  1. rate-limit the logs, so we don't log too much
  2. make the timeouts dynamically configurable
  3. emit these with destination/cg tags, and perhaps avoid logs all together

@kirg kirg merged commit 81b1e01 into master Nov 1, 2017
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants