Re: [task #5490] syslog

[olafBuddenhagen]

Hi,

On Tue, Jan 30, 2007 at 09:30:32PM +0100, Neal H. Walfield wrote:

> The syslog facility could be implemented as follows:
> 
> The program which starts a translator opens /dev/klog on the
> translator's behalf and inserts the file descriptor into the
> translator's stdout and stderr.
[...]

An interesting idea, though it doesn't really help with the problem at
hand: This thread (and I think the task it refers to) isn't about
logging from normal translators (which wouldn't need any special
facility to use syslog() in the canonical way), but from the low-level
Hurd servers.

For logging from normal translators, this approach is certainly worth
considering. Note however that the same would work with traditional UNIX
daemons -- and yet, it hasn't been done this way; syslog() instead has
been introduced as a completely different facility.

Of course, this doesn't necessarily mean that the idea is bad and the
syslog() approach must be better. But at least one needs to wonder, why
it has been done differently... Any ideas?

-antrik-