| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
| |
with theirs (change the -I option to -s (but leave -I in for backwards compat.)
Also eliminate an make sane some magic numbers, and fix a small bug where we'd
send to an unopened socket.
Reviewed by: wollman
Obtained from: NetBSD
|
| |
|
|
| |
the dI in `.Op Fl dI' for the name of an internal mdoc command).
|
| |
|
|
|
|
|
|
|
|
| |
`interesting' feature of syslogd turned up. It calls getbootfile() for each
log entry. Since the kernel makefile now changes kern.bootfile when doing
a 'make install', it's quite startling to see the syslog lines change.
This change makes it call getbootfile() once at startup and cache it,
saving a syscall per loop, and keeping something a little more asthetically
pleasing in /var/log/messages...
|
| |
|
|
|
|
|
| |
Document `-d' and `-I'. Add a BUGS section noting that
logging from UDP is an unauthenticated remote disk-filling service,
and probably should be disabled by default in the absence of some sort
of authentication.
|
| |
|
|
|
|
|
|
|
| |
so that it only unlinks the file if syslogd knows it created it.
If the path specified for the socket already exists then syslogd
will now exit with an "address already in use" error which is more
sensible than blindly unlinking the existing filename. This stops
syslogd -d foo/bar from unlinking foo/bar if it's a real file.
|
| | |
|
| | |
|
| |
|
|
| |
Submitted by: pds
|
| |
|
|
|
|
| |
Use getbootfile() to determine the kernel name.
Submitted by: Peter da Silva
|
| |
|
|
| |
(and rename a few in the process).
|
| | |
|
| |
|
