Patrick Proniewski:
[ Charset ISO-8859-1 unsupported, converting... ]
> Hi all,
>
> On my internal mail gateway, I've found in the log some "watchdog timeout".
> These last days, it's about one or two a day, sometimes none:
>
> Oct 7 11:55:50 ru postfix-mailgw/smtp[96909]: fatal: watchdog timeout
> Oct 7 11:55:51 ru postfix-mailgw/master[1167]: warning: process
> /usr/local/libexec/postfix/smtp pid 96909 exit status 1
> Oct 7 21:21:23 ru postfix-mailgw/smtp[76389]: fatal: watchdog timeout
> Oct 7 21:21:24 ru postfix-mailgw/master[1167]: warning: process
> /usr/local/libexec/postfix/smtp pid 76389 exit status 1
> Oct 5 05:32:02 ru postfix-mailgw/smtp[30650]: fatal: watchdog timeout
> Oct 5 05:32:03 ru postfix-mailgw/master[1167]: warning: process
> /usr/local/libexec/postfix/smtp pid 30650 exit status 1
>
> It seems the smtp process does its job, then is not used for few hours, then
> dies:
Postfix daemons terminate voluntarily when they receive no request
for $max_idle seconds. The watchdog timer is a safety mechanism;
it terminates a process when it appears to hang.
The $max_idle timer uses select(2), poll(2), epoll(3) or /dev/poll
depending on the operating system type and version. Failure of this
timer is usually a symptom of virtualization bugs.
The watchdog timer uses alarm(3).
Wietse
