@Chris: Version 3 weewx with my wmII was rock solid. I didn't run the wmII more than a day under version 4, so I can't speak to that. I had no issues restarting weewx - didn't need a reboot.
NOTE: systemd behavior, being what it is, if a service crashes, it still is not STOPPED, and it won't start a service that it thinks is still running. Instead, you have to stop it (service weewx stop) before you can start it (or, you could do a "service weewx restart" which does a stop first.) JRJ On Wednesday, June 9, 2021 at 12:09:08 PM UTC-5 [email protected] wrote: > JRJ: > > Are you saying that weewx stopped generating reports? My weewx stopped > generating reports, with no error messages, at 4:00 AM. Restarting service > weewx did not fix it. Rebooting my raspberry pi DID fix it. So, I added a > crontab entry to reboot my raspberry pi each morning: > 5 4 * * * /sbin/shutdown -r now > > So far, it has not happened again. I'm using a WMII, with serial port > connection, using Python3 and the latest version of the driver from > JayJaeger > > Chris Shaker > > On Wednesday, June 9, 2021 at 5:46:52 AM UTC-7 [email protected] wrote: > >> @michael: It should not be related to the SCP upload, which continues >> even after weewx has gone "night night". It is running from a cron, not >> under weewx. It merely copies the generated HTML/graphics up to another >> machine. It runs every 17 minutes. If it were, say, locking up a file >> and causing a report to fail there ought to be a message from Python about >> that, and there are no such messages. This same SCP was running with weewx >> version 3 for many months without issues - originally every 15 minutes, I >> changed it from 15 to 17 yesterday to put it out of sync from the report >> generation a bit it after the second failure, just in case. >> >> [It did occur to me to add a little mod akin to the RSYNC present in >> weewx itself, or to cut over to using then RSYNC under weewx after the >> rest of this gets straightened out.] >> >> I also verified that all of the files and directories are owned by and >> have the same group as the weewx process, and they are, so it should not be >> an issue of file permissions during report generation, either. >> >> Regarding the dev release driver, sure, I'll give that a try, and turn on >> more logging later today. >> >> JRJ >> >> On Wednesday, June 9, 2021 at 1:40:52 AM UTC-5 [email protected] >> wrote: >> >>> Nah, the logging level change doesn't matter, I've published a dev >>> release a few days ago making exactly that change. >>> >>> I'd suggest you to try the new driver version anyway since I also made >>> some changes regarding the HTTP interaction with the WLL. You should also >>> enable debug logging so we get a more exhaustive look into what's going on. >>> >>> My suspicion is that something's wrong with the report generation or the >>> SCP upload. Was the last iteration of the report actually uploaded >>> correctly? >>> >>> >>> >>> vince schrieb am Mittwoch, 9. Juni 2021 um 06:10:01 UTC+2: >>> >>>> On Tuesday, June 8, 2021 at 12:43:44 PM UTC-7 [email protected] wrote: >>>> >>>>> (Note: I did "tweaked" the WLL driver code to suppress the "Emitting >>>>> poll/(broadcast) messages by changing those two log calls to >>>>> "log.debug"). >>>> >>>> >>>> I'd suggest running the unaltered driver and especially in this case so >>>> you get the maximum logging so the author can help you. >>>> >>>> If you start changing his code, eventually it'll be "you changed it, >>>> you own the results good or bad" when he can't recreate the issue because >>>> your code has diverged from the authoritative version. >>>> >>>> -- 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/868ff588-eed6-4cee-b94b-61143e0cfa85n%40googlegroups.com.
