Hello fungalnet.

ntp and openntpd can not be installed at the same time, this is the normal behavior.

In detail.
There was no ntp package on obarun repository until now. When I started to work on ntpd-66serv, I had to correct some mistakes on openntpd-66serv
Next I did some optimization. Now, configuration file for NTP can be found in /etc/ntp/ and configuration file for OPENNTPD can be found in /etc/openntpd/

So, if you had ntp software installed and you want to install openntpd, you have to uninstall first ntpd-66serv, and next install openntpd-66serv who will install openntpd.
if you had openntpd software installed and you want to install ntp, you have to uninstall first openntpd-66serv, and next install ntpd-66serv who will install ntp.

Do not forget during the service migration to use the -C option with 66-enable.
So if you have ntpd-66serv installed before the update the dependency from openntp to ntp is switched, that's all, right.
I can't remember, openntrp-66serv is a new pkg? ntpd is giving me the wrong time again, remember I had a custom (Eric revision) ntpclient working for me on my main installation, ntpd-66serv here.
update, new ntpclient-66serv available
ntpclient-66serv is a skarnet script, nothing to do with ntp and ntpd-66serv.

If you made personnal changes on your ntpclient-66serv script, backup it before installing the new pkgbuild.
i take advantage of this thread to explain again how to keep a personal change on a frontend file.

ntpclient-66serv upstream service (service provided by pacman) go to /usr/lib/66/service/nptclient. If you do change at this service you will loose it when the package is updated. If you want to avoid that, make a copy of /usr/lib/66/service/nptclient to /etc/66/service/ntpclient.
Now even if the upstream is updated your own /etc/66/service/nptclient service will NEVER be touched.

so 
# pacman -S nptclient-66serv
# cp /usr/lib/66/service/nptclient /etc/66/service
---- make your change at /etc/66/service/nptclient ---
# 66-enable ntpclient
if you want to simply change the environment variable of the daemon you don't need to touch the frontend file itself
In your case the ntpclient frontend file define the [environment] section like this 
[environment]
cmd_args=!0.pool.ntp.org
you want to change the value of the cmd_args variable, simply do(obviously the service should be already enabled): 
# 66-env -r  "cmd_args=!1.pool.ntp.org" ntpclient
# 66-enable -f -c -S ntpclient
now your service use the new value of your cmd_args variable.
Well now what's happen if the upstream change the service ?
Nothing. The environment variable is never touched if you don't specify it especially.
For example you do: 
# 66-env -L nptclient <- just display the current environment used by the service
# pacman -Sy ntpclient-66serv
# 66-enable -f -S nptclient
# 66-env -L nptclient <- check if the environment have changed
the frontend file will be updated but your environment variable rest the same.
now if you do 
# 66-env -L  
# pacman -Sy ntpclient-66serv
# 66-enable -S -f -C nptclient
# 66-env -L
the fontrend file will be updated AND you loose your change at the environment variable. the -C option say:"overwrite my envrionment variable by the the one coming from the upstream."
Ok, it makes perfect sense (is this part of posix compliance?) but I have not yet understood what /etc/66/conf does, why would it have an env variable different than what I have in /etc/66/service ?
a concret example can be explain here with your dhcpcd trouble (https://forum.obarun.org/viewtopic.php?id=843).

Actually in the frontend file of dhpcd the command line is: 
dhcpcd ${cmd_args} -f ${conf_dir}/${conf_file}
at the execution time 66 will replace the variable by the given one at /etc/66/conf/dhcpcd, meaning 
cmd_args=!-B
cmd_args_stop=!-x
conf_dir=!/etc
conf_file=!dhcpcd.conf
So the command line become 
dhcpcd -B -f /etc/dhcpcd.conf
i guess this part is clear for you. Just i explain here for other people.
So, now you want to specify the interface to use. To do so , you need to pass the name of the interface into the dhcpcd command line.
Well do 
# 66-env -r 'cmd_args=!-B eth0' dhcpcd
now you /etc/66/conf/dhcpcd will contain: 
cmd_args=!-B eth0
cmd_args_stop=!-x
conf_dir=!/etc
conf_file=!dhcpcd.conf
to appy this change at the daemon run 
# 66-start -r dhcpcd
now the daemon use as command line 
dhcpcd -B eth0 -f /etc/dhcpcd.conf
So you have modified the command line of the daemon without the need to change the entire frontend file.
In case of update of the daemon from upstream you don't loose your change except if you ask it to 66.
a command like 
# 66-enable -f -S dhcpcd
will not overwrite the /etc/66/conf/dhcpcd file.
So you use the new script from the upstream but with your environment variable set previously
A command like 
# 66-enable -C -f -S dhcpcd
will overwrite the /etc/66/conf/dhcpcd file and so you loose your change made previously
https://forum.obarun.org/viewtopic.php?pid=3537# p3537 
@ 400000005d35b7ec0a3b8f6c s6-clockadd: fatal: unable to read 16 bytes from stdin: Success
With the custom ntpclient script I had from the above comment on 66-0.2 it is not working and throwing the above error.
So much trying to save scripts and moving them over. 2 days and I still have a mess.
Trying the new ntpd I get this:
2019-05-12 10:10:52.712366500  reply from 94.130.184.193: offset 0.001380 delay 0.084872, next query 31s
2019-05-12 10:10:54.807924500  reply from 129.70.132.37: offset 0.000360 delay 0.093331, next query 31s
2019-05-12 10:10:54.815070500  reply from 144.76.96.7: offset 0.007651 delay 0.100528, next query 30s
2019-05-12 10:11:17.937410500  reply from 80.151.151.109: offset 0.001106 delay 0.097398, next query 33s
2019-05-12 10:11:17.952087500  reply from 51.148.141.63: offset -0.003074 delay 0.113905, next query 30s
2019-05-12 10:11:18.035936500  reply from 212.18.3.19: offset 0.006090 delay 0.098384, next query 34s
2019-05-12 10:11:18.565757500  pipe write error (from main): Success
2019-05-12 10:11:18.565769500  ntp engine exiting
2019-05-30 07:28:41.030055500  adjtimex returns frequency of 0.000000ppm
2019-05-30 07:28:41.030216500  /var/lib/openntpd/db/ntpd.drift is empty
2019-05-30 07:28:41.071949500  ntp engine ready
2019-05-30 07:28:41.135307500  could not parse "pool.ntp.org": System error
2019-05-30 07:28:41.135437500  could not parse "0.europe.pool.ntp.org": System error
2019-05-30 07:28:41.135554500  could not parse "1.europe.pool.ntp.org": System error
2019-05-30 07:28:41.135666500  could not parse "2.europe.pool.ntp.org": System error
2019-05-30 07:28:41.135779500  could not parse "3.europe.pool.ntp.org": System error
2019-05-30 07:28:56.045220500  no reply received in time, skipping initial time setting
2019-05-30 07:30:50.310429500  pipe write error (from main): No such file or directory
2019-05-30 07:30:50.310457500  ntp engine exiting
you mean with the -f option at the end of the script? Yes, I tried, I tried the pool.ntp and given it different local or not so local domains, no change.
To tell you the truth going back to my main installation with 66 -0.1 and looking at logs I can't tell much but it syncs to milliseconds. The other I threw it intentionally out from +9s to hours, or days, see if it will adjust and nothing.
jm has changed the compilation of openntpd, maybe this cause the trouble when you try to use the nptd-66serv.
If it the case a downgrade of the openntpd package should resolve it.
But ntpd never worked for me, only ntpclient worked and only with the custom script above. Now it doesn't work either.

Powered by Obarun