Package: trickle Version: 1.07-4 Followup-For: Bug #375173 I'm being bitten by the same bug: running something like NNTP (using port forwarding) over an SSH session managed by trickle results in 100% CPU utilization (mostly in system, rather than in user), regardless of the values of -t or -l that are applied to trickle. On a 1.5 GHz system, and with trickle set to limit traffic to under 20KB/sec, it seems like this is way out of line for the amount of traffic involved.
A regular SSH session, without the additional traffic, doesn't seem to incur the same overhead AFAICT. So, it seems like trickle may be spending too much time on the packet inspection process, although I'm not sure how to prove that. This is a very useful tool, and I'd really like to see it remain supported. Please let me know if I can help debug it further at the user level. -- System Information: Debian Release: 4.0 APT prefers stable APT policy: (990, 'stable'), (500, 'testing') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.18-5-686 Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Versions of packages trickle depends on: ii libc6 2.3.6.ds1-13etch2 GNU C Library: Shared libraries ii libevent1 1.1a-1 An asynchronous event notification trickle recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
