[log] [metric] Log Setting API, More TRACE Logs, and Gatherer Support #201
Conversation
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
…patibility Signed-off-by: hfuss <hayden.fuss@kaleido.io>
…for logging Signed-off-by: hfuss <hayden.fuss@kaleido.io>
…lementation Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
…amically via monitoring server; all dbsql logs to trace Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
EnriqueL8
reviewed
Dec 1, 2025
Contributor
EnriqueL8
left a comment
EnriqueL8
left a comment
There was a problem hiding this comment.
Thanks @onelapahead - I agree with moving those logs to Trace, I've felt the same level of pain debugging logs and being polluted by that.
I have a few questions on the way the log level is set and some confusions there
onelapahead
commented
Dec 1, 2025
onelapahead
commented
Dec 1, 2025
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
… name w/ comment Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Co-authored-by: Peter Broadhurst <peter.broadhurst@kaleido.io> Signed-off-by: hayden <haydenfuss@gmail.com>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
| // Wrap the request itself in a log wrapper, that gives minimal request/response and timing info | ||
| l := log.L(ctx) | ||
| l.Infof("--> %s %s", req.Method, req.URL.Path) | ||
| l.Logf(hs.apiEntryLoggingLevel, "--> %s %s", req.Method, req.URL.Path) |
Contributor
There was a problem hiding this comment.
Was thinking you'd grab the local/remote fields back out the context to log here.
peterbroadhurst
approved these changes
Dec 1, 2025
Contributor
peterbroadhurst
left a comment
peterbroadhurst
left a comment
There was a problem hiding this comment.
Thanks for working through this. Important stuff 👍
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
Signed-off-by: hfuss <hayden.fuss@kaleido.io>
EnriqueL8
approved these changes
Dec 1, 2025
Contributor
EnriqueL8
left a comment
EnriqueL8
left a comment
There was a problem hiding this comment.
Looks great - thanks @onelapahead for working through all the comments 🙇🏼
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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 is a slightly philosophical PR - but working on codebases using ff-common, I've found I'm often failing to find the log message I need, and often fighting to filter out all the noise of the logs I rarely need in production (but do find helpful in development). Adapted from #200 originally.
In short - I think we log too much at the
debuglevel, especially in these libraries which should mostly logtrace, and not enough at theinfolevel. I worry we depend too much ondebugto "see everything", and we can't guarantee users will actually havedebugon when they hit a new problem. We need to make sureinfohas "enough" (which is very hard to known) to debug know issues / bugs / edge cases, and compensate additionally with metrics and tracing. For unknown issues / networking / database-related issues - thats wheretracecan be enabled temporarily to help triage.Between human SREs, log aggregation systems, and agentic AI - there is a cost to every byte of waste in terms of storage, processing, and context. A "less is more" mindset is necessary. Especially as we consider adopting more modern, performant logging frameworks like https://github.com/uber-go/zap which encourage structured logging and sampling.
All that said - this PR proposes two significant changes:
ffapi,ffresty,fftls,andtodbsqltraceto avoid logging 10-100s of lines per API request (especially when TLS and databases are in play)PUTAPI on the monitoring server:tracelogs referred to above, or go frominfotodebugin a less expensive way, w/o continually over logging.Additionally, exposes the
prometheus.Gatherwithin the metrics managers' Prometheus registry to allow for custom metrics exporting and filtering.