Skip to content

Latest commit

 

History

History
48 lines (42 loc) · 2.66 KB

TODO.md

File metadata and controls

48 lines (42 loc) · 2.66 KB
Raw

Improvements and fixes to be done

  • P2 Labeling GKE Nodes (VM Instances) and Disks

    • Today, this is not done because it forces recreation of resources.
    • It can be done through specific mechanisms for GKE clusters.

  • P2 PubSub push endpoint security: Note: The token which is now used is not very secure, though it is an improvement on earlier versions of Iris, and Google has at times recommended it.

    Either

    • Replace the PUBSUB_VERIFICATION_TOKEN with random value in deploy.sh
    • Or better: Use JWT

  • P2 Use Cloud Tasks instead of PubSub

    • to trigger label_one. This will allow a delay of 10 minutes for Cloud SQL, so allowing allow the labeling on creation (rather than just cron) for Cloud SQL
    • to trigger do_label from schedule(), with a random delay, so minimizing the number of App Engine instances that are created.

  • P3 In integration_test.sh

    • Test more labels (in addition to iris3_name)
    • Test the copying of labels from the project.
    • Support testing of the cron-based labeling, which would also allow testing of Cloud SQL (otherwise difficult because it takes 10 minutes to initialize). In this test: 1. Modify cron to run 1 minute after the deploy launches (and restore it at the end of the test.) 1. Call deploy.sh using with the -c switch to disable event-based labeling 1. Wait 1.5 minutes after deploy before checking that the labels are there.

  • P3 Address the error "Labels fingerprint either invalid or resource labels have changed", printed in _batch_callback, which occurs intermittently, especially with disks. Solutions: - Retry - Ignore and let the cron do it - Implement Cloud Task with a delay. (Not clear if that will help.)

  • P3 Rethink the need for title case in class names. This is clumsy for Bigtable and Cloudsql.

  • P3 See the TODO in the code re not labeling GKE nodes.

  • P4 Implement new labels, for example using ideas from the GCP Auto Tag project But don't add too many: There are a lot of fields on resources.

    • Add a label with the email of the creator of the resource
    • For disks, add a label with the name of the instance to which they are attached.
    • Immediately label boot disks that are created with their instance.
      • (This seems not to work now in Iris3, though that needs to be double-checked.)
      • As shown in GCP Auto Tag, do this by pulling a list of disks from the information about the instance.