search

thoth-station / thoth-application

Thoth-Station ArgoCD Applications
GNU General Public License v3.0
12 stars 22 forks source link

Some results of container image analyses fail to sync #2243

Closed fridex closed 2 years ago

fridex commented 2 years ago

Describe the bug

Some results of container image analysis results fail to sync. The reason behind it is exceeded deadline as the result of container images computed is pretty big

To Reproduce Steps to reproduce the behavior:

  1. Go to the stage environment
  2. Try to analyze quay.io/thoth-station/s2i-pytorch-notebook:v0.0.2 and sync results to the database
  3. Observe graph-sync failure caused by deadline exceeded

Expected behavior

Graph sync should succeed.

codificat commented 2 years ago

/triage accepted /assign @harshad16 /priority critical-urgent

fridex commented 2 years ago

The issue here might be deadline on workflow level or on graph-sync task level (both have defaults taken). The goal here is to identify if the graph-sync task is killed based on the workflow or task deadline and increase the corresponding deadline.

fridex commented 2 years ago

/sig knowledge-graph

harshad16 commented 2 years ago

The package extraction for the mentioned image was tested. s2i-pytorch image was successful extract and synced to db of stage cluster: Results from thoth container images pytorch https://stage.thoth-station.ninja/api/v1/container-images?page=0&per_page=25&os_name=ubi

Result from successful graph sync run of the package-extract: s2i-pytorch cli

It is now also verified on the prod cluster as well: https://khemenu.thoth-station.ninja/api/v1/container-images?page=0&per_page=25

The graph-sync active deadline min is set to 50, and the graph sync for such container images seems to be taking 25mins. https://github.com/thoth-station/thoth-application/blob/5bba76b4ac23f677fc4aa61eb13cd60954ac14db/graph-sync/base/openshift-templates/graph-sync.yaml#L41

As there were no changes made, I think earlier it failed due to an issue on the database side, which seems to be fixed with the updated storage module and database schema. Please let me know if we should look at some other images.

harshad16 commented 2 years ago

closing this issue as the image is being successfully analysed Thanks everyone for the effort :100: