The problem with the queue system is that things will hang up in there and there really is no way to get rid of them that doesn't require a Windchill restart (typically a cache clearing as well). A stuck queue is always at the "EXECUTING" state. In WC10 there is no way to remove this without the restart process. Its all a large pain. I do a couple of things to help alleviate these issues.
1- Setup multiple publishing queues (1 and 2).
a. Redundancy can help if one queue gets stuck
b. Doesn't just transfer everything from the stuck queue to the active but at least SOME stuff will get through
2- Republish Schedule
a. I have a publishalllatestepmdocumentnorepresentation job that runs very early in the AM after backups
b. I also do a cleanup job once a week
3- Worker machine reboot
a. This is a necessity as the worker machine can get hungup in a number of ways where even just restarting the worker services don't do the trick
b. I have a remote restart command of the windows type in my nightly backup script.
i. My backup script turns off Windchill, clears out the caches and restarts the worker machine all before doing the actual backup command and finally bringing everything back online
c. When used in conjunction with item #2 this helps keep everything published, the worker running smoothly and the queues in decent shape.
[cid:image001.gif@01CD64C6.383DD150]
Steve Vinyard
Application Engineer
