weewx.pid and systemd

[gary....@gmail.com]

I had a strange thing happen today. Power failed during the tropical storm.

I shut down the server as the UPS will only hold for 30 minutes.

However, after 25 minutes, the power returned and I started the server.

After a couple of hours, I checked my weather page and found it hadn't updated since right before the shutdown.

So, I checked logs and found little past this:

Aug  4 20:35:50 srvr systemd[1]: /etc/systemd/system/weewx.service:13: PIDFile= references a path below legacy directory /var/run/, updating /var/run/weewx.pid → /run/weewx.pid; please update the unit file accordingly.

Surely that wouldn't stop weewx from running, so I restarted the server. Same behavior.

I also noticed that mosquitto didn't seem to run properly.:

FIFO /tmp/dlt cannot be opened. Retrying later...

Eventually after unproductively chasing the mosquitto/dlt error, I stopped weewxd then started from the command line. It ran and produced loop data. CTL-C out and started the service.

Runs fine.

Beyond my level for sure. Any ideas/requests for logs, etc would be appreciated.

[Greg from Oz]

Edit /etc/systemd/system/weewx.service

If the above file doesn't exist then edit /etc/init.d/weewx 

Wherever /var/run is change to /run/

Restart and see what happens.

[Graham Eddy]

/run and /tmp are both (usually) tmpfs file systems - maybe something screwed up tmpfs then the files/dirs/symlinks meant to be auto-created by tmpfiles.d didn’t happen - would be interested to see `mount | grep tmpfs` and also, if /run and /tmp mounted correctly, what is under them pre-mount (which is hard for these particular mount points because of running system dependencies)

--
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 weewx-user+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/51c4a7ba-855d-47c0-b428-aa21d2a7003bn%40googlegroups.com.

[Graham Eddy]

> On 5 Aug 2020, at 12:03 pm, gary....@gmail.com <gary....@gmail.com> wrote:

> Aug 4 20:35:50 srvr systemd[1]: /etc/systemd/system/weewx.service:13: PIDFile= references a path below legacy directory /var/run/, updating /var/run/weewx.pid → /run/weewx.pid; please update the unit file accordingly.

normal - can be ignored

> FIFO /tmp/dlt cannot be opened. Retrying later…

not normal, hence my previous comments about /tmp

[gary....@gmail.com]

mount | grep tmpfs

udev on /dev type devtmpfs (rw,nosuid,noexec,relatime,size=8118412k,nr_inodes=2029603,mode=755)

tmpfs on /run type tmpfs (rw,nosuid,nodev,noexec,relatime,size=1632724k,mode=755)

tmpfs on /dev/shm type tmpfs (rw,nosuid,nodev)

tmpfs on /run/lock type tmpfs (rw,nosuid,nodev,noexec,relatime,size=5120k)

tmpfs on /sys/fs/cgroup type tmpfs (ro,nosuid,nodev,noexec,mode=755)

tmpfs on /run/user/1000 type tmpfs (rw,nosuid,nodev,relatime,size=1632720k,mode=700,uid=1000,gid=1000)

tmpfs on /run/user/0 type tmpfs (rw,nosuid,nodev,relatime,size=1632720k,mode=700)

[Graham Eddy]

i meant it would have been good to see this when the problem was manifesting… in particular whether /run and /tmp were writeable at the time and whether they were on root volume or tmpfs volumes

well, /tmp is not on tmpfs now so that should be fixed regardless (but system will function without it) - if it is already configured to automount /tmp on tmpfs then this is smoking gun that something broke (and is still broken), otherwise mounting it is just an improvement

To view this discussion on the web visit https://groups.google.com/d/msgid/weewx-user/4c82b777-be7b-4255-9207-b9c4f4c8098dn%40googlegroups.com.