We are running into a timeout with TestHistogramConcurrency on our Jenkins box. I noticed in the stack trace for the timeout this block.
goroutine 2348 [chan send]:
github.comcast.com/ventris/kober/vnd/github.com/prometheus/client_golang/prometheus.(*goCollector).Collect(0xc20801e8e0, 0xc20800a7e0)
/var/lib/jenkins/jobs/Kober/workspace/src/github.comcast.com/ventris/kober/vnd/github.com/prometheus/client_golang/prometheus/go_collector.go:49 +0x6dd
github.comcast.com/ventris/kober/vnd/github.com/prometheus/client_golang/prometheus.func·028()
/var/lib/jenkins/jobs/Kober/workspace/src/github.comcast.com/ventris/kober/vnd/github.com/prometheus/client_golang/prometheus/go_collector_test.go:27 +0x11a
created by github.comcast.com/ventris/kober/vnd/github.com/prometheus/client_golang/prometheus.TestGoCollector
/var/lib/jenkins/jobs/Kober/workspace/src/github.comcast.com/ventris/kober/vnd/github.com/prometheus/client_golang/prometheus/go_collector_test.go:28 +0x35e
This suggested that even though the TestGoCollector test was finished, a goroutine was still hanging around. I traced it back to the call to c.Collect always sending twice of the provided channel. This change receives that second value and allows the goroutine to finish with the test.
Still can't figure out why TestHistogramConcurrency is timing out after 2 minutes :(
- Clarify documentation about sorting requirements.
- Add missing histogram support in consistency check.
- Add label sorting to consistency check.
- Improve error messages when reporting a metric.
(Previously, the metric name was not printed.)
InstrumentHandler provides proxy object over given http.ResponseWriter
- responseWriterDelegator that only implements the bare minimum
required of an http.ResponseWriter and doesn’t implement any interface
upgrades (http.Flusher, http.Hijacker, etc.). This commit fixes it by
providing fancyResponseWriterDelegator with all the fancy bells and
whistles if standard library http.ResponseWriter is detected.
Heavily inspired by Goji's middleware.
https://avtok.com/2014/11/05/interface-upgrades.html#the-proxy-problem
The cgo dependency in the "procfs" package is being removed in:
https://github.com/prometheus/procfs/pull/4
So now it should be fine to always import the "procfs" package and have
it determine dynamically at runtime whether the proc filesystem is
actually there.
This fixes a problem for users which were vendoring client_golang on
MacOS X, but building the result on Linux (the procfs package was
missing in the vendored packages).
If a metric family returned by the injection hook already exists (with
the same name), then its metrics are simply merged into that metric
family. With enabled collect-time checks, even uniqueness is checked,
but in general, things stay the same that the caller is responsible to
ensure metric consistency.
This fixes https://github.com/prometheus/pushgateway/issues/27 .
The new vendoring was produced by running:
godep save -r ./examples/... ./prometheus/... ./text/... ./model/... ./extraction/...
Two things to note:
- "extraction/processor0_0_{1,2}_test.go" imported a package from
"github.com/prometheus/prometheus", all for just one tiny testing
function. To not have to deal with a circular vendoring dependency, I
simply replaced the usage of the function by some in-line logic.
- godep grouped the rewritten imports slightly differently for some
reason, but at least the standard library imports are still in a
separate section. Not sure if it's worth manually keeping our old
import grouping scheme or if we should simply use that godep-generated
one.
As it is expected that the process collector can fail under certain
conditions (proc information for a process only readable by root or
other user for example) and as there is currently no option to configure
the error behavior of the client, this change reverts the error
reporting introduced in 159e96f. This effectively means that errors are
simply ignored and there won't be any samples for the process_* metrics
in case of an error.
Once a user can control how to behave in case of errors returned by
collectors, this change should probably be reverted.
The procfs package has a cgo dependency (necessary to calculate system
times). As procfs is not available under windows, darwin and supposely
all newer BSD systems, this change remove the procfs dependency on these
systems.
If a given pidFn for a process collector can't determine the pid, this
had to be signaled with an invalid pid so far (pid <= 0). In order to
make the error interface clearer for users, this change introduces an
explicit error parameter.
Fabian Reinartz