Hi,
The queue gets removed because by default there are no backups and when you
kill the node, you potentially lose part of your data. To fix this provide
the CollectionConfiguration specifying the correct number of backups like
this:
CollectionConfiguration colCfg = new CollectionConfiguration();
colCfg.setBackups(1);
IgniteQueue<String> queue = ignite.queue("MyQueue", 0, colCfg);
Having one backup guarantees that you won't lose data when one node fails.
If you assume that there can be more failures at the same time, you will
need to have more backups.
I'm not sure why the log is growing. Most likely your service continues to
try polling from the queue even when it's not available anymore, so each
iteration results in exception. So probably you should change your code to
stop the service or recreate the queue after the first failure.
-Val
--
View this message in context:
http://apache-ignite-users.70518.x6.nabble.com/Lot-s-of-repeated-error-logs-when-a-node-is-killed-in-cluster-1GB-minute-tp1143p1150.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.
