Taylor R Campbell <[email protected]> wrote: > > Well, that requires to crash a mail server in busy hours on purpose. I > > will collect if it happens again by accident. > > Despite the confusing name, you can run crash(8) on a running system > without interrupting it -- it lets you non-intrusively examine the > kernel and get stack traces as if with ddb.
Sure, I know about crash(8), but my point is that producing the problematic situation means crashing the machine: it gets so slow that only a reboot from ddb can fix it. -- Emmanuel Dreyfus http://hcpnet.free.fr/pubz [email protected]
