Projection refresh taking too long

Projection refresh that used to take ~6 hours was running for 44 hours before we canceled it. The only change we made prior to that was enabling local segmentation (with scaling factor = 16). We restarted the cluster and started the refresh again, but are afraid that we will bump into the same problem. We ran select * from projection_refreshes; to figure out the session_id and then queried: SELECT * FROM execution_engine_profiles WHERE session_id = '?' which now returns an empty row. Is this expected behaviour? How can we check that projection refresh is running smoothly and get an estimate of time it will need to complete?

Comments

  • Hi, after we restarted the cluster, the projection refresh finished successfully. Still, it would be good to know how to check projection refresh progress while it is running. We can see time executing, but not time left or any other indicator of how much was already done.
  • Hi Grega, glad you were able to solve the problem. If you'd like to see the time left or a status bar for this type of projection, feel free to post it in the ideas section of our community. This way, our product team can review it and consider it for future releases.

Leave a Comment

BoldItalicStrikethroughOrdered listUnordered list
Emoji
Image
Align leftAlign centerAlign rightToggle HTML viewToggle full pageToggle lights
Drop image/file