This repository has been archived by the owner on May 26, 2020. It is now read-only.
Add Application metrics Section in User Docs #119
Closed
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
This uses a much simpler example application, with the minimum possible kubernetes configuration. Some outdated values have been corrected (e.g. the hostname at which the deployed application can be accessed).
This may save the reader the hassle of following the link to the ECR section and then coming back.
The user isn't configuring the cluster, just their namespace. Similarly, we don't need to specify that we're deploying the application 'to the cluster'
sablumiah
changed the title
Update the 'deploying an app.' walkthrough
This commit adds a brief explanation of how to remove unneeded cluster resources that the user created while working through sections of the guide. The 'Cleaning Up' section is added as the last part of the 'Deploying an app' part of the user docs.
jasonBirchall
suggested changes
Mar 5, 2019
| @@ -93,6 +93,111 @@ kubectl -n <namespace> describe prometheusrules prometheus-custom-rules-<applica | |||
| ## PrometheusRule examples | |||
| If you're struggling for ideas on how and which alerts to setup, please see some examples [here](https://github.com/ministryofjustice/cloud-platform-infrastructure/blob/master/terraform/cloud-platform-components/resources/prometheusrule-examples/application-alerts.yaml). | |||
|
|
|||
| ## Application Metrics | |||
There was a problem hiding this comment.
Due to the volume of this entry, I would suggest this needs its own document.
|
I think it's worth including a mention of an important consideration: if you're using a pre-forking web server (like unicorn or puma for Ruby, or gunicorn for Python) and have it configured to use multiple processes, then you need to use a Prometheus client library which supports exporting metrics from multiple processes. Not all the official clients do that - for example prometheus/client_ruby#9 has been open for a long time, although prometheus/client_ruby#95 looks promising so 🤞 If you don't use a library which supports this, then requests to |
Our policy for the environments repo is for developers to create branches with the changes they want, not forks.
Add a 'Cleaning Up' section to the user guide
* Add 'Step 4' (since we use 'Step 1-3' earlier) * Use imperative tense, like the other steps
Describe the process of using `kubectl delete` with a directory full of yaml files to delete all named resources from the user's namespace using a single command.
Explain `kubectl delete --filename` to delete
Make the document headings consistent
As I've had to run this a few times and have to keep googling it... Note: the example text decodes to `Aladdin:open sesame` and is not an actual secret
…mple Add an example showing use of base64 to decode ECR secrets
For rails applications with force_ssl = true, probes need extra http headers. Adding this explanation here will help developers avoid problems with failing probes, without our help.
Rails 5 sends a 307, not a 301. We don't really care which redirect we're avoiding.
Add description of using https headers for http liveness/readiness probes, where the application only responds to https.
If users are not in the MoJ/webops github team, they won't be able to clone the helloworld demo app. via SSH.
Amend instructions to git clone via https, not ssh
People who are familiar with github pages will know how to go from https://ministryofjustice.github.io/cloud-platform-user-docs ...to https://github.com/ministryofjustice/cloud-platform-user-docs For those who don't, this change provides instructions on how to find the source repo and contribute to the guide.
This makes things less confusing when using a development jekyll server, while editing the document. It also protects against the documentation repository being moved, in the future.
This requires approval from the cloud platform team, and takes a while to apply once approved. So, it makes sense to tell the user to do this first.
If the user tags their docker images with a different version number (using the makefile in the demo app), they will need to update the image reference in the kubernetes deployment yaml files.
It's quite likely that someone will already have deployed an instance of the demo app at https://multi-container-demo.apps.cloud-platform-live-0.k8s.integration.dsd.io/ In that case, creating the ingress on the cluster will fail. This warns the user to expect that error.
…-app Add a guide for deploying a multi-container app
Add instructions on how to contribute to the guide
This is a temporary measure to forewarn teams that live services will be hosted on the live-1 cluster (although it's not quite ready yet).
…ce/cloud-platform-user-docs into live0-deprecation-notice
Add a deprecation warning for the live-0 cluster
…ud-platform-user-docs into applicationmetrics
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
WHAT
Add section on how to create application metrics
WHY
This will enable users to create an application metrics endpoint via instrumentation or using an exporter, and then create alerts based on those metrics using Prometheus