I recently started using the 2.2.18pre series because of instability with the 2.2.16 kernel, although the 2.2.18pre17 I am using is much more stable, I did notice one problem, although to be honest i have not tested this on an older kernel. I want to use rdate -s to keep my machines in time sync with my file server, all running the 2.2.18pre17 kernel, but when I did this I noticed that all of the connections from rdate to the file server left the connection in CLOSE_WAIT, and even after days the connections remained so. A netstat -e shows that there is one byte left in the queue: Proto Recv-Q Send-Q Local Address Foreign Address State User Inode tcp 0 0 router.home.pc:ssh mutt.home.pc:1023 ESTABLISHED root 4817 tcp 1 0 router.home.pc:time mutt.home.pc:1036 CLOSE_WAIT root 4816 This looks like a DOS to me if it happens when any bad client connects and disconnects and data is left in the queues. I hope this is the appropriate mailing list for this problem, I do find the MAINTAINERS file hard to read sometimes. I do have a spare machine I can regression test if that would be helpful. -Shawn -- Shawn Houston mhouston@gci.net - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org
