search

yugabyte / yugabyte-db

YugabyteDB - the cloud native distributed SQL database for mission-critical applications.
https://www.yugabyte.com
Other
8.68k stars 1.04k forks source link

[docdb] Master self-healing #1375

Open rao-vasireddy opened 5 years ago

rao-vasireddy commented 5 years ago

Jira Link: DB-2191 Master process should move to a new node if a node is available when the current node goes down. In large deployments, you are likely to have other nodes available when the node where a master is running goes down. In such scenarios, we should start a master process on one of the other nodes so that master does not stay under replicated

bmatican commented 5 years ago

cc @ramkumarvs Maybe at the very

I think the short term view for this is that we have a YugaWare decommission flow to take care of this. The long term plan should be #1316 and having YB be able to heal itself in the face of master failures, but we'd still need to consider the implications on YugaWare if we dynamically move masters underneath...