hi Val,

All these tasks & transactions are for system cache, and don't know when
they're called. IIRC, they should be related with cache partitions exchange,
binary mapper and so on.

And when few nodes have issues like network or any other hardware issue, if
these tasks/transactions take too long time to fail, the whole cluster
hangs, so I want to find an efficient way to quickly identify the failed
nodes and isolate them to reduce the impact.

BTW, if one node encounters issue, e.g. OOM or something like this,  when do
the exchange with all the nodes, is there any efficient way to unblock the
cluster by isolating the failed node and recovering it to the previous
healthy status?

Thanks,
-Jason



--
View this message in context: 
http://apache-ignite-users.70518.x6.nabble.com/How-to-set-timeout-for-ComputeTask-and-GridCacheTransaction-tp6592p6609.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Reply via email to