Solved [help] syslogd does not start

[Mjölnir]

uname -aU: FreeBSD t450s.local.lan 12.2-RELEASE-p4 FreeBSD 12.2-RELEASE-p4 GENERIC amd64 1202000

To debug this, I added sysrc syslogd_flags+=" -d". When I service syslogd start, the error message is: logmsg: pri 53, flags 0, from , msg getaddrinfo failed for syslog: Service was not recognized for socket type

Where do I have to dig deeper? Thx.

 

[zirias@]

Judging from this message, maybe here:

$ grep syslog /etc/services
syslog        514/udp
syslog-conn    601/tcp    #Reliable Syslog Service
syslog-conn    601/udp    #Reliable Syslog Service

edit: see also services_mkdb(8).

 

[Mjölnir]

This is a typical shoot.self.foot(into)

syslogd(8) is a very basic system service, and even more basic are the varios name services (user/id/group/password, /etc/services names/#, DNS) queried via NSSWITCH/nsdispatch(3).

So the story goes like this:

1. Greenhorn has issues with nsswitch.conf(5)/nscd(8), see this post over there.
   (nscd(8)'s caching does not work; nasty error messages from NSSWITCH(_nsdispatch) written by syslogd(8) into /var/log/debug.log).
2. GH changes /etc/nsswitch.conf -> nasty error messages don't come up any more
3. Conclusion: these changes in /etc/nsswitch.conf must have been good, right?! Plain simple logic, right?! RIGHT???!!! Or not?
4. Some time after reboot, GH begins to wonder: "why are the /var/log/messages not updated properly?"
5. GH checks service syslogd status: none, niente, garnix.
6. GH removes the show-stopper from /etc/nsswitch.conf (NIS, who run that in a SOHO? but it's in the default config...), sysrc syslogd_flags-=" -d", service syslogd start, service syslogd status: syslogd is running as pid 9254.

The lesson is clear: syslogd(8) needs nsdispatch(3). Don't pull that carpet from under it's feet.

 

[zirias@]

Ah, sure, if it's not the database itself, it's the way to access it – makes sense.

This is a typical shoot.self.foot(into)

good hunting ?