We are having trouble with our directory browser app which is driven by Sun One 5.2 (using IIS connecting to a servlet in Tomcat via jk2).
With a very low number of concurrent users, the CPU hits 100% and the app either hangs or slows down considerably before recovering.
In my directory log, I see a very large number of concurrent entries like this when the hang occurs:
[29/Nov/2004:14:34:30 +0000] conn=302783 op=-1 msgId=-1 - closing - A1
[29/Nov/2004:14:34:30 +0000] conn=302834 op=-1 msgId=-1 - closing - A1
[29/Nov/2004:14:34:30 +0000] conn=302835 op=-1 msgId=-1 - closing - A1
I understand this is an internally issued request, indicating that the client is aborting the connection? Is that correct?
Could this message indicate a cause of the problem, or is it a symptom of a problem elsewhere?
If anyone has seen this behaviour before I'd be very grateful for any advice.
Thanks alot in advance for any help,