ivakegg opened a new issue, #3157:
URL: https://github.com/apache/accumulo/issues/3157
In 1.10 we set master.startup.tserver.avail.min.count to 174 and
master.startup.tserver.avail.max.wait to 5m. At one point the system was taken
down and upon restart we only had enough nodes to start 157 tservers. We
expected the system to continue startup after 5m but instead it continued
blocking. The last master info message we saw was:
Blocking for tserver availability - need to reach 174 servers. Have 157
Time spent blocking 1575 sec.
That would be 26.25 minutes and we were still blocked.
It appears that the master.startup.tserver.avail.max.wait is not working as
we expected. We had to manually modify the
master.startup.tserver.avail.min.count in zookeeper to be able to start the
system. Either we need to update the master to break out of the loop once
max.wait is met, or add a master.start.tserver.avail.max.iterations property or
something like that.
- Affects 1.10 but I believe this code is unchanged in 2.1
- OS: CentOS 7.3
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
