Author Topic: Cache Changes (Read 4194 times)

~~GnarlySnarly~~ · « **Reply #20 on:** March 31, 2006, 12:38:06 pm »

Quote from: KM on March 31, 2006, 11:41:57 am

***
the initial tests seemed fine, which leads me to conclude that the bug is somewhere with connection testing (it seems to be running out of handles - so probably a bug with closing sockets properly when running tests

which reminds me.. about excessive handle usage in the past...

While i was tinkering with firewall rules... I got an alert of too many handles in use.. I looked at my ABPmon graph and it showed a definite rampup of handle use over a 20 minute period, until it rose to the level needed to trigger the alert warning. I then looked at the handles column in TaskInfo and WCS was using nearly 1000 handles by itself. A quick look at the list showed the vast majority were for sockets.

Seems the firewall rule was blocking some WCS traffic.. but even so, shouldn't WCS be able to close the blocked sockets after xx sec/min?

spam:
ABPmon and TaskInfo are some net system monitoring tools!
www.iarsn.com

KM · « **Reply #21 on:** March 31, 2006, 01:25:22 pm »

i found the problem with closing sockets in connection testing, it was where i experimented with closing one socket before testing with another socket - then changed back to closing both sockets at the end instead... except i left it set to forget about the first connection before running the test...

would make more sense if you could see the code, lol

~~GnarlySnarly~~ · « **Reply #22 on:** April 01, 2006, 12:32:51 pm »

Quote from: KM on March 31, 2006, 01:25:22 pm

i found the problem with closing sockets in connection testing, it was where i experimented with closing one socket before testing with another socket - then changed back to closing both sockets at the end instead... except i left it set to forget about the first connection before running the test...

would make more sense if you could see the code, lol

was that in WCS code [my problem] or in the new peer cache software?