Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Latency outlier detection #288

Open
2 of 3 tasks
mattklein123 opened this issue Dec 13, 2016 · 5 comments
Open
2 of 3 tasks

Latency outlier detection #288

mattklein123 opened this issue Dec 13, 2016 · 5 comments
Labels
area/outlier_detection enhancement Feature requests. Not bugs or questions. help wanted Needs help!

Comments

@mattklein123
Copy link
Member

mattklein123 commented Dec 13, 2016 •
edited
Loading

Overall tracking issue for major outlier detection features:

  • Consecutive 5xx
  • Periodic SR
  • Periodic latency
@mattklein123 mattklein123 added the enhancement Feature requests. Not bugs or questions. label Dec 17, 2016
@mattklein123 mattklein123 modified the milestone: 1.3.0 Feb 22, 2017
@mattklein123 mattklein123 modified the milestones: 1.3.0, 1.4.0 Apr 13, 2017
@mattklein123 mattklein123 modified the milestone: 1.4.0 Jun 12, 2017
@mattklein123 mattklein123 added this to the 1.6.0 milestone Nov 26, 2017
@mattklein123 mattklein123 removed this from the 1.6.0 milestone Jan 13, 2018
@mattklein123 mattklein123 added the help wanted Needs help! label May 18, 2018
@mattklein123 mattklein123 changed the title Outlier detection Latency outlier detection Nov 16, 2018
@rene-m-hernandez
Copy link

rene-m-hernandez commented Jul 17, 2019 •
edited
Loading

Discovered this by accident while searching through outlier issues (sorry to revive such an old issue).

Curious if Latency Outlier just fell by the wayside or if there were roadblocks that prevented it from happening?

In addition the latency outlier, I'd also be interested in the Periodic Success Rate; perhaps that refers to the success percentage based outlier that exists now? I was hoping for something along the lines of "cumulative 5xx over a specified time window" (not necessarily consecutive).

@mattklein123
Copy link
Member Author

@rene-m-hernandez just need someone to work on it.

@uromahn
Copy link

uromahn commented Nov 6, 2019

Wow, this is quiet "old".
Just adding a comment so the issue gets a freshness bump.
We are interested in this too and I might start working on it, provided I can free up enough time from my "normal" day job.

twghu referenced this issue in twghu/envoy Jan 5, 2021
@gargnupur @dio
log: Add "%_" to print message with escaped newlines (envoyproxy#13832)…
c658523 
… (#288)

This patch adds a new custom flag `%_` to the log pattern to print the
actual message to log, but with escaped newlines.

Signed-off-by: Dhi Aurrahman <dio@tetrate.io>

Co-authored-by: Dhi Aurrahman <dio@tetrate.io>
@mattklein123 mattklein123 mentioned this issue Apr 21, 2022
Open
@jtway
Copy link
Contributor

jtway commented Jun 17, 2022

Would upstream request time to first bytes be something the community would be open to as well? I get that this would be more effort and maybe even a follow up to leveraging response time from the upstream.

@slonka slonka mentioned this issue Sep 5, 2022
Closed
jpsim pushed a commit that referenced this issue Nov 28, 2022
@rebello95 @jpsim
swift: add conversion from Request to headers (#288)
ecfe5ea 
Adds converters to go from `Request` to a set of headers that upstream Envoy understands.

This will be utilized by the request logic in the `Envoy` class.

Previous consideration: envoyproxy/envoy-mobile#275

Signed-off-by: Michael Rebello <me@michaelrebello.com>
Signed-off-by: JP Simard <jp@jpsim.com>
jpsim pushed a commit that referenced this issue Nov 29, 2022
@rebello95 @jpsim
swift: add conversion from Request to headers (#288)
e995c5e 
Adds converters to go from `Request` to a set of headers that upstream Envoy understands.

This will be utilized by the request logic in the `Envoy` class.

Previous consideration: envoyproxy/envoy-mobile#275

Signed-off-by: Michael Rebello <me@michaelrebello.com>
Signed-off-by: JP Simard <jp@jpsim.com>
@holooooo
Copy link

Is must be a useful feature. In my company, the most of errors is client side timeout and it looks like a downstream disconnect which former outlier detection could not find and eject. I'd love to contribute it but I can't write c++...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
area/outlier_detection enhancement Feature requests. Not bugs or questions. help wanted Needs help!
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
7 participants
@uromahn @junr03 @mattklein123 @jtway @holooooo @rene-m-hernandez @cpakulski