We determined that one of the three nodes in the cluster was not connected to the cluster for about 15 hours today. While all data was correctly written to the database, the two nodes in the cluster were showing different content than the third node which wasn't connected. We resolved this issue around 6pm Pacific time on December 3, and we are taking steps to ensure it doesn't happen again.
There may be some additional delay before the caches are 100% in sync again, and I will report status tomorrow. Meanwhile, we continue to watch the application closely so we can respond quickly if any other problems arise.
Thank you for your patience. Regards, Robert
Update December 4, 11:21am Pacific: It appears that node synchronization problem is resolved, but we are still seeing high load on the servers and some errors reported by VMware Communities participants. We are actively working with Jive, and we have some active areas of investigation that we expect will lead to improvements. Thanks again for your patience.
Update December 5, 3:55pm Pacific: We made a configuration change early this morning that has decreased server load significantly.
Corey_R
JohnTroyer
ericni
RDellimmagine
Badsah