Hi, I am upgrading to Ignite 2.16 (from 2.12) and after 20+ hours, I get reconnect errors and have to restart the whole cluster (servers and clients)? The above error message is logged in the client log file when it tries to reconnect to the cluster??? All Ignite nodes are running on RHEL 8, running Java 17.
I did not have any of these reconnect issues in test. Now that I am trying to deploy this system to operations, I am running into issues. Also, as expected, the operational system is running under much heavier load. The Thick clients are (re)connecting via an IgniteConfiguration Spring bean. Upon (re)connect, the following pseudo code is followed: 1 - lookup the bean via the name [*IgniteConfiguration igniteCfg = AplicationContext.getBean(<String>)*] 2 - (re)establish a connect to the cluster [*Ignite cluster = Ignition.start(igniteCfg)*] I just recently found that *Ignition *has a *getOrStart() *method. Should I now use that now?? Was that just added with the recent releases? Also, I noticed the IgniteSpringBean and IgniteClientSpringBean - what is the purpose of these? Again, the clients are not "starting" the cluster. The clients are only allowed to connect to the cluster after it has been activated. The client connections are blocked via firewalld. Since I am behind schedule and trying to deploy to OPS, I would prefer to fix this via Spring configuration, if possible. Thanks in advance for any assistance/thoughts about this configuration. Greg
