I think some of you know, that I'm still testing sequouia on usability with a huge amount of virtualdatabases. I started a scenario two days ago, everything worked (nearly) fine. The memory usage is still continuously increasing, in my eyes without any real reason. During this night, the real RAM was insufficient, so that the machine swapped. This is a totally normal thing for an Unix system, but after that, the controller wasn't reachable anymore, neither through JMX nor using jdbc. The java process was still running (processlist), but I found no way to communicate with the controller. Checking the logs was also uneffective, nothing strange, no new entries after yesterday afternoon. Did anyone had such experiences in the past? Any idea, how to avoid this next time?
Best regards, Malte _______________________________________________ Sequoia mailing list [email protected] https://forge.continuent.org/mailman/listinfo/sequoia
