2021, 03/29 13:18 (1344day) Posted
| 1343day Updated
The problem that had been occurring since the morning of the 27th, where when acquiring a broadcast slot and starting viewing in HD mode, an error message would appear indicating that there were no slots available, was resolved at 6:00 am on the 30th.
The cause was a failure on the cloud provider's side, but because it was an unprecedented type of failure, the alarm did not work and discovery was delayed, and mitigation measures were not as effective as expected.
Additionally, after the cloud provider began work on the service after the morning of the 29th, some other network functionality failed, although the initial outage had been resolved.
The following measures have been added to prevent recurrence.
・Added a function to constantly check the interconnectivity of all nodes in the cluster ・Improved the failure detection function of the automatic node increase/decrease function ・Prepared more spare resources than before (in case of failure)
We apologize for the inconvenience caused for a long time.
https://magical.kuku.lu/?3613
ツイート