Even if there is an exception somewhere, it might be swallowed by the container. Try installing a custom
exception handler (forum parser doesn't like links with a hash inside...):
Another thing you could do is set the Quartz log level to debug or trace to see if your job is being triggered at
As I said the quartz is being triggered and working fine for couple of weeks and then just stops running. Jobs are triggered as 'run forever' and there is no rule when it is going to stop.
I will go with the fine logging and wait for it to happen again:(
I was working with quartz for a long time but never experienced this kind of issue. This happened when quartz was run within the seam container.
I had similar issue on our production servers and was hard to debug the problem. It turns out that Quartz had an issue with 1.6.4 release (and we were using older version) where threads
dieout for a jvm error. It could be possible that threads just died (which is bad). I'm currently using the latest 1.7.1 release and not anticipating any issues.
Seems its maybe safer to not using in memory but database storage for quartz.
Well probably but we didn't have the need for rerunning jobs if they failed.
And I forgot to reply back on this. After a lot of time trying to figure out what was happening we moved to quartz servlet initialization which works excellent. The class which is invoked by quartz programatically initializes seam context and thats it.
Quartz is reading from its xml configuration file and is watching on it so schedulers can be easily redefined at runtime.