9f5628e55b
Instances can bounce between hosts many times in a short interval, especially if the CSVs go down (as much as 20 times in less than 2 minutes). We're not handling this properly. The failover handling logic is prone to race conditions, as multiple hosts may attempt to claim the instance, which will end up in an inconsistent state. We're introducing distributed locks, preventing races between hosts. At the same time, we're validating the events, as the instances can move again by the time we process the event. The distributed lock backend will have to be configured. At the same time, we're now waiting for "pending" cluster groups, which may not even be registered in Hyper-V, so any action we take on the VM would fail. Closes-Bug: #1795299 Closes-Bug: #1796673 Change-Id: I3dbdcf208bb7a96bd516b41e4725a5fcb37280d6 |
||
---|---|---|
.. | ||
notes | ||
source |