Question
Last activity: 25 Apr 2017 5:54 EDT
What could be the reasons as to why a node goes offline-unknown for few minutes and comes back online-normal
In a DNode cluster for a customer, a node went offline-unknown for 15 minutes and came back on its own to online-normal. From the PegaRULES log for node, here is the log snippet -
2015-09-02 13:47:33 with Cassandra ERROR; hazelcast Timeout
2015-09-02 13:55:48 D-Node exited the cluster
2015-09-02 14:10:32 D-Node joined the cluster
Thank you in advance for your help and insights.
***Updated by moderator: Lochan to add Categories***
Accepted Solution
Offline-unknown means that the PRPC server is offline (for example, server restart), or seen as offline (for example, network issues), but the node itself is still recognized as a D-Node. The fact that it comes back to online-normal after about 15 minutes seems to point to a restart of some sort (PRPC application, server or application server, for example), although I would not rule out network issues.
Pegasystems Inc.
NL
One possible explanation is that "Offline-unknown" node is overloaded with some other work and unable to respond.
The fact that one of the nodes goes "Offline" temporary shouldn't break the system, it will function properly.
Can you attach the real log for the outage period? Maybe we could get some more information from it.