- Le plus récent
- Le plus de votes
- La plupart des commentaires
Hello there,
Our MSK cluster complies with all three recommendations; however, we observed that the MSK endpoint was not accessible for a period of 1 hour.
This is unexpected behaviour. This needs to be investigated further. Please check the ActiveControllerCount metric to see how much time the broker went down. Also, please check the partitionCount metric to see which broker went down for how much time. Also, you can check the broker logs if enabled. If needs, please raise a support case to investigate further and to avoid this issue in future.
Is it expected that both endpoints may go down during rolling update?
No, during any update and rolling restart only one broker goes down at a time..
Also, we currently do not have support plan, AWS still does not provide a way from dashboard to re-schedule the updates during non-production hours.
This can't be done from your end. You need to reach out to our Support team to request the reschedule of your maintenance window to non-production hours. Also, you can ask us to reschedule it to particular time and day of every month.
Ex: Every 1st Sunday at 02:00AM UTC
Contenus pertinents
- demandé il y a 9 mois
- demandé il y a 8 mois
- demandé il y a 2 mois
- AWS OFFICIELA mis à jour il y a un an
- AWS OFFICIELA mis à jour il y a un an
- AWS OFFICIELA mis à jour il y a un an