ntpd not starting after upgrade to 12.0-RELEASE

after upgrading from 11.2-RELEASE to 12.0-RELEASE ntpd(8) won't start.
The error message in /var/log/messages is:
Code:
Dec 16 14:55:41 ludger-thinkpad root[2017]: /etc/rc.d/ntpd: WARNING: failed to start ntpd
When I execute /etc/rc.d/ntpd start manually I get
Code:
Starting ntpd.
su: unknown login: ntpd
/etc/rc.d/ntpd: WARNING: failed to start ntpd
The ntpd user is included in /etc/passwd
Code:
ntpd:*:123:123:NTP Daemon:/var/db/ntp:/usr/sbin/nologin
Any idea what I'm missing?
 
Did you carefully run mergemaster -p? The ntpd user is new. Mine look like this:

Code:
ntpd:*:123:123::0:0:NTP Daemon:/var/db/ntp:/usr/sbin/nologin

Btw, I don't have how to test it because I don't have ntpd(8) installed.
 
I had the same issue. Here, mergemaster added the ntpd user to /etc/passwd, but for some reason spoiled the password database. I simply ran /usr/sbin/pwd_mkdb -p /etc/master.passwd and this resolved the issue.

After updating another system, the old ntpd.pid file owened by the 11.2 root user was for some reason left in place and prevented the new ntpd starting up, because the user ntpd had no write permissions to the old .pid file (owned by root). I needed to delete said old .pid file manually and that resolved that issue.
 
Last edited:
I have a similar situation to that of the original poster.

Prior to the upgrade, I had:

ntpd_enable="YES"

in /etc/rc.conf .

I upgraded my system to 12.0 with the following:

freebsd-update fetch
freebsd-update install
freebsd-update -r 12.0-RELEASE upgrade

I didn't run mergemaster at any time, nor do I recall any mention of it during the upgrading procedure. I also re-installed all the packages and ports as required by the process.

Can I still get ntpd to run?
 
...
I upgraded my system to 12.0 with the following:
freebsd-update fetch
freebsd-update install
freebsd-update -r 12.0-RELEASE upgrade
...

After the upgrade command, there should have been more 3 install commands interlaced with reboots and before the final install one, rebuilding of all 3rd party software. Did you forgot to inform these steps, or were these simply not executed yet?

I wrote down a brief upgrade log which worked for all of the systems which I maintain:
 
After the upgrade command, there should have been more 3 install commands interlaced with reboots and before the final install one, rebuilding of all 3rd party software. Did you forgot to inform these steps, or were these simply not executed yet?

I wrote down a brief upgrade log which worked for all of the systems which I maintain:
Yes, there were a number of times in which I was required to reboot the machine and then run:

/usr/sbin/freebsd-update install

Towards the end, all the packages and, I believe, ports were rebuilt.

The command:

pwd_mkdb -p /etc/master.passwd ,

however, doesn't ring a bell with me. Maybe I wasn't paying attention or something like that. Would that have been the step that I missed? If so, can I run it and fix my problem with ntpd?

Thanks.
 
Yes, there were a number of times in which I was required to reboot the machine and then run:

/usr/sbin/freebsd-update install

Towards the end, all the packages and, I believe, ports were rebuilt.

The command:

pwd_mkdb -p /etc/master.passwd ,

however, doesn't ring a bell with me. Maybe I wasn't paying attention or something like that. Would that have been the step that I missed? If so, can I run it and fix my problem with ntpd?

Thanks.
It looks like I might have fixed the problem. I ran:

pwd_mkdb -p /etc/master.passwd

and, afterward, changed the ownership of:

/var/ntp/ntpd.pid

to ntpd.

I set my watch to the WWV time signal and checked it against the computer's clock. Both are synchronized.

As it turns out, this problem is a bug with freebsd-update upgrade, though not, it seems, with a fresh installation of FreeBSD 12.0.
 
After the upgrade command, there should have been more 3 install commands interlaced with reboots and before the final install one, rebuilding of all 3rd party software. Did you forgot to inform these steps, or were these simply not executed yet?

I wrote down a brief upgrade log which worked for all of the systems which I maintain:
Your link to your upgrade log was quite helpful. Thank you.
 
Thanks for the tips everyone. I ran into the same issue and I have ntpd running again now.

However, it appeared that for some reason the /var/db/ntp/ntpd.pid file does not contain the correct PID:

$ ps aux | grep [n]tpd
ntpd 24350 0.0 0.2 15856 15928 - Ss 15:10 0:00.18 /usr/sbin/ntpd -p /var/db/ntp/ntpd.pid -c /etc/ntp.conf -f /var/db/ntp/ntpd.drift


$ cat /var/db/ntp/ntpd.pid
39781


Which also leads to an incorrect status:

$ service ntpd status
ntpd is not running.


The ntpd.pid and ntpd.drift appear to have the correct permissions, and the ntpd.pid file is updated when I restart the service -- just not with the PID that is actually running.

I killed the running ntpd process and deleted the ntpd.pid file and started ntpd again, e.g.:

$ sudo service ntpd start

Now the PID matches, status is reported correctly, etc. Just thought I'd mention it in case anyone else is running into the same thing.
 
FreeBSD 13.0-Relase, after ntpd enabled, the permission of /var/db/ntp/ntpd.pid is root:ntpd" instead of ntpd:ntpd, this cause that the service always said "ntpd is not running", but it actually running.
Using this command to fix it:

# chown ntpd:ntpd /var/db/ntp/ntpd.pid
 
Back
Top