-
Notifications
You must be signed in to change notification settings - Fork 879
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.
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
redis_exporter v1.0.0 #256
Conversation
Codecov Report
@@ Coverage Diff @@
## master #256 +/- ##
========================================
Coverage ? 74.2%
========================================
Files ? 2
Lines ? 663
Branches ? 0
========================================
Hits ? 492
Misses ? 143
Partials ? 28
Continue to review full report at Codecov.
|
6a7f840
to
e928651
Compare
bfd9d4b
to
636d56a
Compare
7fca386
to
feaa114
Compare
aac1be8
to
460e89b
Compare
This is available on docker as
|
6d6c689
to
36cdf48
Compare
@SuperQ - this PR addresses the issue you raised in #228 - do you mind having a look, at the feature in particular and maybe the PR in general? Could always some more feedback here, esp if the general direction makes sense. @brian-brazil - this PR addresses #31 that you raised a million years ago. Mind having a look if the general idea of the changes is somewhat sane? (cc @jkohen re: issue #178 ) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks better from a quick peek.
Thanks @brian-brazil - that's good feedback. |
365b467
to
038b87b
Compare
Let's do this. |
Github has issues with notifications right now, the exporter binaries and docker image for 1.0.0 will be available as soon as these have been resolved. |
That's very sad. We are running multiple redis containers on a single instance which means we now need to run multiple redis_exporter processes :( We use Prometheus and we do Service Discovery but we don't want to run multiple redis_exporters :( |
You don't need multiple exporters. The new version can still scrape multiple redis instances. It is configured more like the blackbox exporter. |
@roman-vynar - see here for more information how to configure Prometheus to scrape multiple hosts. |
Hi, Can anyone help me with redis exporter metrics prometheus config:
target is healthy under prometheus, but still data is not coming over grafana. When i see, the metrics curl http://DNS:9121/metrics | grep redis redis_exporter_build_info{build_date="2018-09-20-18:15:12",commit_sha="8bb0b841e9a70b0348f69483e58fea01d521c47a",golang_version="go1.10.4",version="v0.21.2"} 1 issue: only very few metrics are coming, because of which data is not coming over grafana. metrics required:
none of the data is coming, can anyone help. What am i missing Thanks in advance |
This PR introduces breaking changes and will be released as v1.0.0
Why
This project was started over four years ago and a lot has changed since. This is an effort to get the redis_exporter to more closely adhere to the Best Practices for Writing Exporters and clean up the code.
The biggest change is that the exporter no longer directly supports configuring (and scraping) multiple redis instances but pushes that responsibility to the Prometheus server. If this is how you currently use the exporter, then have a look at this section how to configure the Prometheus server to scrape multiple instances.
However, if you already use the exporter as intended by the Prometheus project and scrape only one instance (and e.g. use the
--redis.addr
and--redis.password
cmd line flags) then little to nothing will change for you.Breaking changes in this PR:
All this functionality should happen in the Prometheus server.
You can still scrape multiple redis instances with one exporter, see the README for an example Prometheus server configuration.
addr
andalias
labels are gone, use Prometheus server label rewriting for this if need/want them back./scrape
endpoint to allow scraping a particular instancetotal_system_memory_bytes
will not be included by default any longer, use a command line flag to include it anyway.latency_spike_milliseconds
-> latency_spike_seconds_seconds
last_scrape_duration
andscrape_error
metrics are now per-instance (if you scrape multiple instances) andscrape_error
includes the error as a label value--is-tile38
command line flag is setNew features:
rediss://
does not work #251 - added a flag to allow skipping cert validationTodo