Poking at this a bit more, I'm wondering if it's Debian/Ubuntu specific. 
The /etc/init.d/weewx script calls /usr/bin/weewxd in Debian systems, which 
is a shell that somehow calls the "real" weewxd in /usr/share/weewx. I'm 
not 100% sure how it does that, as nothing actually refers to that location 
- just /usr/bin, /etc/weewx and /run. I changed the 
/etc/default/weewx-multi to change WEEWXD_BINDIR to /usr/share/weewx and 
once I did calling "/etc/init.d/weewx stop vp2" (or tempest) both worked 
correctly to stop and show status. However, they fail to start it back up - 
import configobj error. Added WEEWX_BIN=/usr/bin/weewxd and left BINDIR at 
/usr/share/weewx and same behavior.

Started everything manually - "python3 /usr/share/weewx/weewxd 
--pidfile=/run/weewx-vp2.pid /etc/weewx/vp2.conf" and no errors - just a 
string of loop packets. Digging in the archives here I found references to 
mis-matched python version causing this error, and I have both python 2.7 
and 3.8 installed. Changed link for python from python2 to python3 and a 
new error - unable to locate /home/weewx/bin/weewxd. I changed the 
/etc/init.d/weewx script - commented out the original variables for 
WEEWX_BINDIR wec. and pasted in the correct ones from 
/etc/defaults/weex-multi but same error... There's now no reference to 
/home in /etc/init.d/weewx, /etc/defaults/weewx-multi or /usr/bin/weewxd 
and yet that's the error.

Again, confusion reigns supreme. Time to put things back and let it rest, I 
think.



On Wednesday, March 23, 2022 at 3:17:09 PM UTC-8 Tom Hogland wrote:

> I'm using 4.7 installed on Ubuntu server - I downloaded the github version 
> of weewx-multi since the Ubuntu package doesn't include it. There's no log 
> info - it just doesn't work correctly when trying to stop or report status, 
> while it *does* work correctly when starting or restarting. Just walked 
> back through the script, thinking maybe there was a typo in PIDFILE or 
> related, but nothing looks amiss. So, I re-grabbed a new copy from github 
> and saved it. Didn't touch anything in it - set all the variables in 
> /etc/default/weewx-multi and tried again - same result. Restart worked, but 
> status says it isn't running. Checked /run/tempest.pid and it correlates to 
> the actual pid of the running tempest instance. Killed all the running 
> weewx instances, deleted the pid files, started them back up with weewx 
> start vp2 and weewx start tempest, both are running, both pids are the same 
> as the pid files created, and yet it still says neither is running when 
> querying status...
>
> This is truly weird, although since it's running and doing all it's 
> weather goodness it's not an emergency or anything. I might let it stew for 
> a few days and look at it again this weekend.
>
>
>
> On Wed, Mar 23, 2022 at 9:44 AM matthew wall <[email protected]> wrote:
>
>> tom,
>>
>> which operating system are you using, and which release?
>>
>> which version of weewx are you using?  more importantly, which 
>> weewx-multi script are you using?
>>
>> if you are not using weewx-multi from the latest weewx release (4.7.0), 
>> could you grab weewx-multi from github and see if that behaves any 
>> differently from the weewx-multi currently on your system?
>>
>> m
>>
>>
>>

-- 
You received this message because you are subscribed to the Google Groups 
"weewx-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/weewx-user/a01c2ca2-1aa8-416f-b4b3-ed8a77bb1246n%40googlegroups.com.

Reply via email to