-
Notifications
You must be signed in to change notification settings - Fork 762
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
Default /metrics route does not support specifying auth_module #815
Comments
Hey @OscarVanL, Can you share your scrape config ? I too have similar issue, where |
I modified my prometheus scrape config to look like this:
Specifically adding the |
Hey @OscarVanL , i tried this and still it didn't work. Attaching one more issue link below, please point me where i'm making the mistake ? |
I couldn't be sure @Sam-Sundar. I'm not an expert with this project, and your setup looks far more complicated than mine. |
Hello @OscarVanL , |
Addresses issue 815: prometheus#815 Like the author of the issue, I also used the example in the README file to setup scraping with Prometheus only to discover that by default the /metrics end-point is scraped instead. Signed-off-by: Thomas Venieris <thomas.venieris@gmail.com>
thanks |
mysqld_exporter version: output of
mysqld_exporter --version
v0.15.1
MySQL server version
MySQL v5.7, v8.0, also MariaDB 10.7
mysqld_exporter command line flags
--config.my-cnf=/etc/mysql/conf.d/mysqld-exporter.cnf
What did you do that produced an error?
I have followed the docs to setup multi-target support.
The use-case is that I have lots of MySQL/MariaDB servers and do not want to reuse the same username/password across them all.
To achieve this, I have created multiple auth modules in the config file, then created multiple prometheus jobs each using a different
auth_module
for all the target(s) where those credentials apply.However, it does not operate as the docs suggest and continues to only use the default
client
auth_module for all targets.As a result, authentication fails to my instances.
What did you expect to see?
The specified
auth_module
should be utilised on the/metrics
route for the target database instance.What did you see instead?
There appears to be one of the following:
/metrics
path that does not exist in the/probe
pathEither the documentation does not mention that the the value
metrics_path: /probe
must be set on the prometheus config (overriding the default/metrics
), or there is a bug that/metrics
does not utilise the providedauth_module
param.The text was updated successfully, but these errors were encountered: