config: Validate timeout and interval of the scrape configuration #275
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
|
|
zecke
force-pushed
the
zecke/profile-timeout
branch
from
55f6038
to
9f7ed78
Compare
brancz
approved these changes
Oct 11, 2021
|
Looks like there's a test that needs to be adapted. |
zecke
force-pushed
the
zecke/profile-timeout
branch
from
9f7ed78
to
e63b648
Compare
metalmatze
reviewed
Oct 11, 2021
pkg/config/config.go
Outdated
Comment on lines
34
to
42
| const ( | ||
| pprofProcessCpu string = "process_cpu" | ||
| ) |
There was a problem hiding this comment.
If we want constants for this, I'd say let's go for constants for all of these strings. :)
There was a problem hiding this comment.
Any preference? I didn't want to hardcode process_cpu in two places (given the names already changed once).
There was a problem hiding this comment.
Yep, that makes sense. Let's make all of them a constant, as you've already started too. 👍
In general we want to stick to the configured interval and don't overrun because of slow or timing out scraping. Go's /debug/pprof/profile endpoint will start profiling when being invoked, has no support for concurrency and the timeout can only be configured in seconds. The existing code alternated between using "Seconds", "Delta"+ScrapeInterval but none of these work correctly when enforcing timeout <= interval. Make sure that "seconds" < ScrapeTimeout <= ScrapeInterval and adjust the default configuration to honor this. In the long run we should consider treating this endpoint entirely different to the others (e.g. jitter scraping instead of sticking to interval, profile for a longer duration). Signed-off-by: Holger Hans Peter Freyther <holger@moiji-mobile.com>
zecke
force-pushed
the
zecke/profile-timeout
branch
from
e63b648
to
07024bb
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
In general we want to stick to the configured interval and don't
overrun because of slow or timing out scraping.
Go's /debug/pprof/profile endpoint will start profiling when being
invoked, has no support for concurrency and the timeout can only
be configured in seconds. The existing code alternated between using
"Seconds", "Delta"+ScrapeInterval but none of these work correctly
when enforcing timeout <= interval.
Make sure that "seconds" < ScrapeTimeout <= ScrapeInterval and adjust
the default configuration to honor this. In the long run we should
consider treating this endpoint entirely different to the others (e.g.
jitter scraping instead of sticking to interval, profile for a longer
duration).