Investigate failing of jobs in write MM services right before the release. It might be timeouts. (osparc 50% failure, s4l.io and science <0-10% failure, for exactly same jobs) (Voyager)

[wvangeit]

• Task is not yet ready error in MM task #1952
• Client lost the connection to the scheduler error in MM tasks #1953

[wvangeit]

For now the hypothesis for osparc.io is that this is caused by:
https://github.com/ITISFoundation/private-issues/issues/225#issuecomment-3027111099

[matusdrobuliak66]

MD Notes:

• The Osparc frontend calls the function-permissions endpoint, and based on the response, it displays the function browser

  • The WEBSERVER_FUNCTIONS env var needs to be enabled

• Testing osparc.io with simple function and 40 samples - 100% success rate

• Testing the official example, issue with permissions

  • SQL query for permissions: Create a function-job-collection to be used in the MM SuMo tutorial and turn it read-only for all #1910

• Testing the official example, already getting some RPC timeouts https://monitoring.osparc.io/graylog/messages/graylog_223/4d0c7d41-8328-11f0-938f-0242c0a80112