5.0.15 crashes without error files, after update to 5.0.21 no data measurement #234
Comments
|
I do not know how it happened but while I was writing the problem, I go back to check out and all be normal again :/ |
This is returned when the web UI (frontned) cannot communicate with the daemon (backend). If this is happening, the daemon is most likely not running and "Mycodo" in the top-left of the web UI will be red, because this is the "check" that's occurring and failing.
The upgrade failed because the mycodo daemon could not be restarted, and the install was reverted back. An upgrade cannot occur if the mycodo daemon isn't functioning properly, so always make sure it's running and responding before performing an upgrade.
This error occurs when influxdb (measurement database) is not responding, and typically happens at startup because it can sometimes take a long time to start. Slower systems (Pi Zero, Pi1) have more of an issue with this than faster (Pi3). Measurements cannot be retrieved or written while influxdb is starting, and until it fully starts, this is the error it returns. Once started, these errors go away and everything works fine (reading/writing to the measurement database). |
|
Excellent, I already understood that it turns red when the deamon is not running, but try to find the error that had caused it to stop, about 2 hours ago, I realize that I was not running, when I check the logfiles , I found nothing more than the following, and that's where I started what I said before |
|
There's nothing abnormal in that log, and only web UI (frontend) errors will appear there, not the daemon. You will need to refer to the daemon log for those errors (through web UI or /var/log/mycodo/mycodo.log). |
|
The problem is that there was nothing strange in that logfile: It is as if he had not registered the error that caused him to stop executing |
|
The worst kind of bugs. How long had it been running before the issue? |
|
Had been running since Friday, March 31, and from what I saw in graphics, the last measurement was recorded on April 2 at 2:45 pm, there I think that the daemon stopped and did not register anymore |
|
I was meaning, that you have been running Mycodo for weeks and this is the first time the daemon stopped? The code hasn't changed that much in the last few updates that I think would start to cause daemon issues, so I think this may be operating system-related and not Mycodo. I would test some more and if it continually happens, then you should try to narrow down what part of Mycodo is causing it, be selectively turning off (or on) features until you can isolate the problem area. |
|
No no sorry, I was having problems with the daemon for some time ago, but I did not find out what the cause was, I rebooted the raspberry and everything was fixed, I always look for updates to keep the latest version of mycodo. Now I will be more attentive, it usually happened every 3-4 days that worked perfect and then stopped, I suspected that it was DHT-22 sensor problems (I have the AM2302, but I understand they use the same library), but I have not Confirm it. Apparently my problem are similar to #198 |
|
I use both the DHT22 and AM2315 and those systems haven't had issues, even running for weeks between updates. So, I would be hard-pressed to believe it's Mycodo. Power supplies can be an issue, specifically those that don't filter out spikes caused by things such as refrigerators, generators, motors, etc. I would change variables (connected sensors, controllers activated, power supplies, locations power supply is plugged in at your location, etc.) and see if you can isolate the issue. |
|
I'm also having this intermittent issue with 5.0.21. When using ● mycodo.service - Mycodo server Apr 03 22:51:46 raspberrypi systemd[1]: mycodo.service: main process exited, code=killed, status=6/ABRT I'll see if I can start the Daemon in debug mode, change power supplies and see how it goes. |
|
Not sure about those errors. Perhaps a fresh install could fix it. |
well, I having this problem some days ago, specifically when I noticed that haven't record after April 2, at 14:55 (this was the last record), when i looking in .../apache2/error.log:
i follow the instructions in this site.
and next I proceed with an upgrade for 5.0.21 with error:
then i decide reboot the rpi, the deamon its ready but no data are recorded with error in apache2:
The text was updated successfully, but these errors were encountered: