Highlighted Key items/actions
Confluent Kafka
Description: new sets of unknown issues (other than with VM Kafka) may emerge.
Action: Peter Kmec (Unlicensed) Ondřej Wantula (Unlicensed) Aleksandr Kanaev (Unlicensed) - write down any new issues or challenges being experienced with Confluent Kafka.
When: Continuously during the switch and post-switch activities.
Euronet VPN
Description: TBD
Action: Ondřej Wantula (Unlicensed) - document the proxy workaround for the future - usable in case Euronet reacts slowly.
When: TBD
SSL Certificates
Action: Document the Observation on Certificates Limis (Denis Galajda (Unlicensed) )
Action: Aleksandr Kanaev (Unlicensed) Change SSL Certificates on the old Stage in January-2023 at latest.
Default nodepools
Description: GCP is creating its own nodepools (for an unknown reason), potentially resulting in breaking some services. Workaround to fix broken services - deleting the new nodepool.
Action: Ondřej Wantula (Unlicensed) to investigate GCP auto-scaling to figure out the root cause.
Communication
Action: Remind tech-leads to read #sre-announcements (Denis Galajda (Unlicensed) Peter Kmec (Unlicensed) )
We will continue to announce in advance: what we’re going to do; when this can be expected; what are the impacts/implications (e.g. downtime)
Action: Share #sre-announcements to POs & PMs channels (Denis Galajda (Unlicensed) )
Recommendation: Set-up a bot to remind app teams to perform needed actions.
GitHub Pipelines
Action: Aleksandr Kanaev (Unlicensed) to share templates for documentation purpose
Terraform structure
Action: Peter Kmec (Unlicensed) to share the new structure (existing example with readme)
Dedicated Channel
Action: Create a temporary channel dedicated to issues and queries related to major DevOps change to create a Point of Contacts for other teams.
Key Observations
TBD
Future Recommendations
“Don’t to everything at the same time“ - but rather in sequence
TBD