Log connection failures at WARNING, with their cause (opal#711) - #60
Open
Stephen-Psaradellis wants to merge 1 commit into
Open
Stephen-Psaradellis wants to merge 1 commit into
Stephen-Psaradellis wants to merge 1 commit into
Conversation
A failed connect in both client handlers is re-raised for the retry wrapper and logged nowhere else, and the endpoint's catch-all around the receive loop was the only record of a client that died mid-session. All of them logged at INFO, so an operator running OPAL at WARNING or ERROR saw nothing when a client could not connect (permitio/opal#711, with the examples in opal#584, #516 and #588). Every connect failure now logs at WARNING and carries the exception text. The endpoint's bare `except:` becomes `except BaseException as err` - the same catch set - so the message names the cause; opal#584 traced a "Client connection failed" line to a special character in a password with nothing in the log to go on. Tests: tests/log_level_test.py - a refused client connect and a receive loop that dies with the RuntimeError the catch-all's own comment names; both fail on the current code (level 20) and pass with the change.
This branch has not been deployed
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes permitio/opal#711 (the examples there - opal#584, #516 and the #588 comment - all come from this library).
A failed connect in both client handlers is re-raised for the retry wrapper and logged nowhere else, and the endpoint's catch-all around the receive loop was the only record of a client that died mid-session. All of them logged at INFO, so an operator running OPAL at WARNING or ERROR saw nothing when a client could not connect.
Every connect failure now logs at WARNING and carries the exception text. The endpoint's bare
except:becomesexcept BaseException as err- the same catch set, so nothing about which exceptions are caught or how the disconnect is handled changes - and the message names the cause. opal#584 traced a bareClient connection failed - 52764 :: <id>line to a special character in a Postgres password, with nothing in the log to go on; that line now ends with thereprof what actually happened.Left at INFO on purpose:
Client disconnected(a normalWebSocketDisconnect),Trying server, and the reader'sConnection was terminated.- those are the lifecycle, not failures.Client connection failedon the endpoint is the only place a live session's death was recorded, so it is the one I promoted.Tests:
tests/log_level_test.py- a refused client connect, and an endpoint receive loop that dies with the exactRuntimeErrorthe catch-all's own comment names. Both fail onmaster(the records come out at level 20) and pass with the change. The full suite is 15 passed with the change; the existing 403 test now emits its rejection at WARNING, which is the point.flake8 --select=E9,F63,F7,F82is clean.Disclosure. An AI system running an autonomous software project found the issue, wrote the fix and the tests, and wrote this description; I'm the account holder submitting it and I have reviewed the patch. Take it, change it or drop it as you see fit.
This fix is offered at a fixed $300, permitio's own median completed bounty on Algora. If you want to pay for it, comment
/bounty $300on the issue before merging and Algora pays it on merge. If you would rather not pay, merge it anyway - the code is yours either way.