The dashboard_engine on a particular host is failing to listen on a port. I've turned up the log level to 5, but I don't see any error messages.
What are some things I can do to fix this?
Does it have a pid I think a java process may be hub you could stop wasp,de,dap and see if you still have java process running.
Neal has a good point. If the probe was somehow started twice, you could end up with an entry in the probe list that does not have a port number listed. Otherwise, I think the dashboard_engine and a few other probes (like the wasp) only get a port number after they finish initializing. I would think that if the dashboard_engine were not initializing completely, you would see something of interest in the log. If I recall correctly, you should see a log message that says something like "started up in X seconds" after the dashboard_engine initialization is complete.
I would recommend trying to verify you have only once instance of the probe running. If that checks out and the log is not giving you any clues, you will probably need to open a support case. Feel free to post a copy of the log here if you would not mind us looking through it.
Thanks for the replies guys.
ntimm: Yes, the dashboard_engine has a PID, but it doesn't have a port. I deactivated the various java based probes, except for dashboard_engine, and java.exe was still running. Then after I deactivated the dashboard_engine, the java.exe process was no longer around. I then restarted all the probes, but no luck.
kruepke: I set the log level to 5 and looked through it, but everything looked successful. I'll take your advice and open a ticket.
I'll open a ticket, but in the meantime, if you have any suggestions, feel free to post them.
I just did a quick test on a server running an older version of the dashboard_engine (2.60). Here is what I saw at loglevel 0 during the startup of the probe:
Jan 31 16:20:10:885 [main, dashboard_engine] Starting dashboard engine...Jan 31 16:20:10:885 [main, dashboard_engine] Dashboard Engine Version : [2.60]Jan 31 16:20:14:696 [main, dashboard_engine] Dashboard Engine started!!Jan 31 16:20:14:712 [main, dashboard_engine] Dashboard engine started in 3811 milliseconds.Jan 31 16:20:14:712 [main, dashboard_engine] ****************[ Starting ]****************
From my experience with the Java SDK, I know that last line (which contains "**[ Starting ]**") is logged by the invocation of the event loop in the SDK. So in a Java probe, you would normally start by setting up the various parts of the probe (logging, callbacks, etc.) and optionally log some stuff related to the startup. Some of the SDK methods used for setup might automatically add lines to the log. Then you start the event loop, which is very similar to the SDKs for the other languages as well. I think this is the point at which the probe starts listening on the port because this is when it can actually accept callbacks.
I am not completely sure if newer versions of the probe keep the logging more or less the same, but I would be curious to find out which lines you see in your log when you stop and start the probe.
Those are pretty much the lines I see when DE starts. I know I had some issues related to nis server and DE, we force DE to use group server and all appeared good. I had to put a entry in via raw config to fix until I got a dev version of DE 3.15D and now it works fine. I know when it was happening the entries we would see in the log where attempting to fetch groups with nis_server or something of that order then nothing. after that would just stay there and never start. I believe that happened shortly after a upgrade to 2.0UMS. You may want to check if yours is attempting to use the nis_server.
kruepke: log level 3 only shows this:
Jan 31 16:07:42:568 [main, dashboard_engine] Starting dashboard engine...
Jan 31 16:07:42:568 [main, dashboard_engine] Dashboard Engine Version : [3.00], Required db Schema version : 3.00
Jan 31 16:07:42:568 [main, dashboard_engine] Instance for DashboardConfiguration has been created
Jan 31 16:07:42:584 [main, dashboard_engine] Configuration file has been loaded:dashboard_engine.cfg
ntimm: I don't see anything related to the NIS server in the dashboard_engine raw config.
You should see something like below in the logs. Even at log_level 2
Feb 01 09:01:33:076 [main, dashboard_engine] ****************[ Stopped ]****************Feb 01 09:01:33:076 [main, dashboard_engine] Dashboard engine stopped...Feb 01 09:01:34:435 [main, dashboard_engine] Starting dashboard engine...Feb 01 09:01:34:435 [main, dashboard_engine] Dashboard Engine Version : [3.15D], Required db Schema version : 3.00Feb 01 09:01:34:450 [main, dashboard_engine] Instance for DashboardConfiguration has been createdFeb 01 09:01:34:450 [main, dashboard_engine] Configuration file has been loaded:dashboard_engine.cfgFeb 01 09:01:34:450 [main, dashboard_engine] deactivate acl condition is: 0Feb 01 09:01:34:466 [main, dashboard_engine] Database Provider: mssql detected.....Feb 01 09:01:35:465 [main, dashboard_engine] Memory Status : Max Limit = 1820.0 MB, Current Allocated = 981.0 MB, Free = 930.0 MB, Used = 51.0 MBFeb 01 09:01:36:512 [main, dashboard_engine] Building Dynamic views tree using group server......Feb 01 09:01:37:230 [main, dashboard_engine] Dynamic views tree build in ms.
Retrieving data ...