You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@ignite.apache.org by szj <sz...@juraszczyk.com> on 2018/06/06 06:17:57 UTC

Re: [ANNOUNCE] Apache Ignite 2.5.0 Released

Hi

While scaling from 100s to 1000s nodes have you by chance changed the way
how nodes can be restarted, too, possibly without mentioning that in the
docs? I happened to start my adventure with Ignite right after 2.5 was
released and could not get this simple thing working properly (details in
the post titled "Ignite 2.5 nodes do not rejoin the cluster after restart").
The problem did not exist when I rebuilt my dead simple PoC cluster using
2.4 so it can't (exclusively) be because I'm just a complete moron.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: [ANNOUNCE] Apache Ignite 2.5.0 Released

Posted by szj <sz...@juraszczyk.com>.

In order to reproduce it you need to start ignitevisorcmd.sh on any node and
connect it to the grid. Then restarting any node (including the one on which
you start ignitevisorcmd.sh) will fail with "Node with BaselineTopology
cannot join mixed cluster running in compatibility mode". Reproducible every
time for me. When ignitevisorcmd.sh is not connected then the node restart
works.



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/