Mehul Ved put forth on 12/23/2009 2:02 AM: > On Mon, Dec 21, 2009 at 9:35 PM, Wietse Venema <wie...@porcupine.org> wrote: >> I think your system is in a state of serious overload. There are >> tons of "connection refused" for internal services, and watchdog timeout >> and other timeouts that should NEVER EVER happen. >> >> Reduce default_process_limit to 50, do "postfix reload" and see >> if the errors go away. > > This is working well so far. No more timeouts and whole queue is clear > as of now. > > Thank you to everyone who replied.
Glad to hear it's working well once again. Say, since this appears to have been an excessive load issue, would you mind posting some basic specs for this server? Such as processor type and qty, num cores, total system memory, disk subsystem (num of disks, raid card, software raid, etc?), and internal/external message load? Thanks. -- Stan
