* Otto Wyss <[EMAIL PROTECTED]> [001130 15:11]:
> Sorry if I had known, that '[EMAIL PROTECTED]' gets relayed to
> debian-x, I wouldn't have sent it. Usually I send everything first at my
> home address and than out to the world.
Well, Branden forwards email sent to his own account to debian-x if he
thinks someone else stands a chance of handling it or if he thinks the
conversation archived would be useful to someone in the future.
> Unfortunatly for me, it's not that easy to find another employer.
Yes, I know; it was meant mostly in jest. :)
> There is absolute nothing in the log (disk), probably because the cache
> doesn't get flushed to disk. I guess there is nothing I could do to
> force a flush in short intervalls. Is there any easy way to redirect the
> output to another computer (running MacOS!) over the net?
Ok. I think we can defeat this one. If you don't mind playing with your
/etc/fstab file, you can change the partition that contains your
/var/log directory to be a synchronous mount. This of course will slow
things down a bit, but not by a terrible amount. (This is how the *BSDs
did their disk IO before Kirk McKusick invented SoftUpdates to get the
safety of synchronous writes with the speed of async mounts. :)
The other option is to have an endless loop of running sync(1). Someone
posted something similar not so long ago, but I am lousy at shell
scripting. The following C might do the job. :)
int main(void) {
while(1) {
if (system("/bin/sync") != 0)
break;
}
}
(Hmm. The system(3) manpage suggests this is very bad; however, since
the machine crashes anyway, it might not be horrible. Read the manpage
though and see if you can live with this before trying this version. :)
(A perl version wouldn't look much different, I don't think. If you
would rather use perl, I could dredge up my perl memories. :)
--
``Oh Lord; Ooh you are so big; So absolutely huge; Gosh we're all
really impressed down here, I can tell you.''
