prometheus / docs Goto Github PK

If I resize the browser until the content gets pushed down below the side menu, I can't click the links on the left side menu (#col-md-3) anymore. Setting z-index:100 fixes that but not sure where to put that and if there is a cleaner way.

Following the getting started tutorial, where one runs the example processes.
You could add people need to set up their GOPATH for go get -d and go build to work.

# Fetch the client library code and compile example.
git clone https://github.com/prometheus/client_golang.git
cd client_golang
export GOPATH=$(pwd)
cd examples/random
go get -d
go build

Also, when doing go get -d I got an error for Mercurial not being installed:

$ go get -d
go: missing Mercurial command. See https://golang.org/s/gogetcmd
package bitbucket.org/ww/goautoneg: exec: "hg": executable file not found in $PATH

Might be worth mentioning.

Currently we only mention in the FAQ that a clean requires sigterm or , that should be in the general docs too and also the http endpoint.

Hmm, should we mention that in the crash recovery error message?

See prometheus/prometheus#902

For some pages like the FAQ and SD it'd be good to have a ToC

Go to the expression browser and verify that Prometheus now has information about time series that these example endpoints expose, such as the rpc_durations_microseconds metric.

If the "expression browser" is located under the "Graph" tab, adding the 3 random servers only provides

• scrape_duration_seconds
• up

rpc_durations_microseconds is available on the respective metrics endpoints of the random services. It does not show up in the dropdown next to the blue "Execute" button.

prometheus/prometheus#1231

...once prometheus/prometheus#1019 is merged.

Following http://prometheus.io/docs/introduction/getting_started/.
Under ## Using the expression browser It says to navigate te the "Tabular" view under the "Graph" tab.
No such view is apparent.

Running prometheus 0.15.1 via the latest prom/prometheus docker image (ecebe0acff82)

I couldn't find the documentation about legend format filters. Maybe there are more promdash features which should be documented?

@stuartnelson3 @juliusv

I might have missed it, but I couldn't find the client format documentation. The format is currently documented in a google doc, but I'd vote to move it here.

https://docs.google.com/document/d/1ZjyKiKxZV83VI9ZKAXRGKaUKK2BIWCT7oiGBKDBpjEY/edit

honor_labels still needs to be documented (PR into next-release branch please).

Once prometheus/prometheus#905 is merged.

The HTML page title currently always says "Prometheus". This isn't helpful for SEO and search reasons. Add the title of each page to the html title attribute, something like $page | Prometheus e.g. Getting started | Prometheus.

We should document that if anyone needs temporary labels, they should use the __tmp_ prefix to ensure that their temporary label name will never conflict with any existing labels.

From prometheus/prometheus#745

prometheus/prometheus#1207

Perhaps next to the road map, add a "non-road-map" with the non-goals, i.e. things that sound like nice features but which we do not intend to implement for reasons to be stated in the new section.

Ideas so far:

• Rule templating language.
• Strings as sample values (and neither ints...).
• Support event logging / become a log db.

More?

If doing everything from 'getting started' page, one will find out that config is wrong:

marko@marko-desktop:~/tmp/prometheus (master) $ ./prometheus -config.file=prometheus.conf
prometheus, version 0.13.3 (branch: master, revision: 572b280)
  build user:       marko@marko-desktop
  build date:       20150520-18:45:13
  go version:       1.4.2
E0520 18:56:16.629990   18370 main.go:224] Couldn't load configuration (-config.file=prometheus.conf): yaml: line 3: did not find expected ',' or '}'
E0520 18:56:16.630068   18370 main.go:225] Note: The configuration format has changed with version 0.14, please check the documentation.

See prometheus/prometheus#1049

Those only have to cover the most basic scenarios to secure a Prometheus setup so people have a base for further customization.

When I try to load a file_sd_config file, I get no error, but prometheus doesn't show any of those hosts on the status page as candidate for scraping, no matter how long I wait. Nothing is logged, even with -log.level=debug, and promtool doesn't check that file.

included file:

targets:
  - host1.example.com:9103
  - host2.example.com:9103
  - host3.example.com:9103
  - host4.example.com:9103
  - host5.example.com:9103
labels:
  - job: zookeeper

I'm using this prometheus:

prometheus, version 0.15.0rc3 (branch: master, revision: d437fce)
  build user:       [email protected]
  build date:       20150720-16:25:16
  go version:       1.4.2

The possibility to query A records was added.

Document prometheus/prometheus#1201 (again)

See prometheus/prometheus#1051

prometheus/prometheus#1155

prometheus/prometheus#1299

• that it is required after a crash
• how long it might take
• what can go wrong (esp. mention the 'orphaned' directory)

See prometheus/prometheus#1050

prometheus/prometheus@9e05d82

See prometheus/prometheus#967

It's awesome that we have a blog now, but it's not possible to subscribe to updates. While RSS/atom is not super popular anymore, it would be nice to provide a feed so that people can use tools to get notified immediately about new blog posts.

http://prometheus.io/docs/alerting/alertmanager/ still documents the old/deprecated alertmanager.

Introduced by prometheus/prometheus#921

See prometheus/prometheus#998

On the metrics naming docs it references a "Prometheus Domain". However, I can't find any more docs on this. I just want some clarity that the naming of the etcd metrics are alright; for example should these be prefixed with etcd_?

• rafthttp_message_sent_latency_microseconds
• rafthttp_message_sent_failed_total

So far, we only list exporters, but there is so much more...

https://github.com/google/cadvisor
https://github.com/GoogleCloudPlatform/kubernetes
https://github.com/armon/go-metrics
https://github.com/coreos/etcd
https://github.com/mesosphere/kubernetes-mesos
https://github.com/peterbourgon/gokit

Anything else?

In "Recording rules" the rule_checker tool is referenced but according to CHANGELOG the tool has been replaced and is not longer go get-able.

We've lots of terms, and I expect some of them (like bridge vs. collector vs exporter) are only in a few people's heads. We should have a glossary.

Title says it all. It should probably be added to the basics section.

We need to clearly document at various places that prometheus.io is based on HEAD and represents the latest releases of the various other repos, while unreleased features (in the HEAD of the various repos) are in the next-release branch of docs.

Currently, we document something completely wrong on http://prometheus.io/docs/introduction/install/ :
"Note that this documentation (as published on prometheus.io) refers to the latest production release. The head of the prometheus/docs GitHub repository refers to the (possibly not yet released) head of the prometheus/prometheus (and other) repositories."

From jaredwiltshire via prometheus-developers:

"The InfluxDB comparison highlights that Prometheus uses much less disk space than Prometheus due to its data storage model. However this seems to no longer be the case - https://influxdb.com/blog/2015/10/07/the_new_influxdb_storage_engine_a_time_structured_merge_tree.html
It also references a Github issue for InfluxDB regarding indexing which has since been closed."

I also believe we might want to include a few other monitoring and event-logging systems.

We should document the backticks for the next release.