Buffer filling to 100% after about 36h

[ConligWX]

I've noticed memory increase too even from day one with the Nano, don't think its anything to worry about.

as for Unifi, yep I have some UniFi too: USG Pro4, AC-LITE, AC Pro, USW 8-60W, and USW-16-150W, UCKG2+ hoping to get a couple of G3 Flex Cameras soon as well.

[staze]

I've noticed memory increase too even from day one with the Nano, don't think its anything to worry about.

as for Unifi, yep I have some UniFi too: USG Pro4, AC-LITE, AC Pro, USW 8-60W, and USW-16-150W, UCKG2+ hoping to get a couple of G3 Flex Cameras soon as well.

Sounds nearly the same as my setup. =)

Good info on memory piece. I did a DB backup and yeah, memory jumped way up. So I assume it's just caching DB in memory (which is perfectly logical)

[staze]

So I think this issue might have been related to a query I had running from my Hubitat.

That said, I just rebooted my wifi and thought this was interesting. It seems like the wifi going down caused the unit to basically soft-reboot?

Code: Select all

system (27.03.2019 19:16:38): Disconnected from WLAN or Internet, reconnect issued.
system (27.03.2019 19:16:45): Access point operation temporarily enabled.
logger (27.03.2019 19:19:59): 10.0.2.16's 2nd request on port 22222 declined.
logger (27.03.2019 19:21:09): message above repeats 3 times.
system (27.03.2019 19:21:18): Access point operation disabled 127878 secs after boot.
logger (27.03.2019 19:21:32): 10.0.2.16's 2nd request on port 22222 declined.
logger (27.03.2019 19:25:36): message above repeats 10 times.
logger (27.03.2019 19:25:54): child process received timeout signal (14).
logger (27.03.2019 19:25:54): disconnect station 0 (Vantage): exit
logger (27.03.2019 19:25:54): disconnect station 9 (System Data): exit
logger (27.03.2019 19:25:55): data logger stopped.
logger (27.03.2019 19:25:55): databases disconnected.
logger (27.03.2019 19:25:55): databases connected. Update interval 3600 seconds.
logger (27.03.2019 19:25:55): data logger (version 5.1a, build 2019) started.
logger (27.03.2019 19:25:55): connect station 0 (Vantage via USB serial).
logger (27.03.2019 19:25:55): connect station 9 (System Data via Plug-in).
logger (27.03.2019 19:25:55): wakeup sent.
logger (27.03.2019 19:25:57): no need to update console date/time.
logger (27.03.2019 19:25:57): setup script executed.
logger (27.03.2019 19:25:57): internal logger activated.
logger (27.03.2019 19:25:58): syncing.
logger (27.03.2019 19:25:58): waiting for sensors (wind0, rain0, th0, thb0) to provide data.
logger (27.03.2019 19:25:58): services paused.
logger (27.03.2019 19:26:08): gettime request sent.
logger (27.03.2019 19:26:18): message above repeats 1 time.
logger (27.03.2019 19:26:18): no Vantage time offset.
logger (27.03.2019 19:26:18): eeprom request sent.
logger (27.03.2019 19:26:18): rain collector size 0.01 inch per tick.
logger (27.03.2019 19:26:18): transmitter 1 active as station type #0 "ISS" (via repeater A).
logger (27.03.2019 19:26:18): firmware version request sent.
logger (27.03.2019 19:26:18): firmware version 3.15 (supports LOOP2)
logger (27.03.2019 19:26:18): no reading of Vantage's internal logger.
logger (27.03.2019 19:26:18): 10.0.2.16's request on port 22222 accepted.
logger (27.03.2019 19:26:19): connection on port 22222 closed.
logger (27.03.2019 19:26:21): 10.0.2.16's request on port 22222 accepted.
logger (27.03.2019 19:26:24): connection on port 22222 closed.
logger (27.03.2019 19:26:24): sensors are now providing data.
logger (27.03.2019 19:26:25): services started.
logger (27.03.2019 19:26:26): 10.0.2.16's request on port 22222 accepted.
logger (27.03.2019 19:27:33): connection on port 22222 closed.
logger (27.03.2019 19:27:35): 10.0.2.16's request on port 22222 accepted.

[staze]

This seems to have been resolved

[staze]

nope, it's something else. My Nano is now failing within about 18 hours. Something causes uploads to halt, even though live data thinks everything is fine. This then causes the buffer to fill.

At wits end here...

[ConligWX]

you are going to have to get Boris to remote into your Nano I guess and see what he can find.

I take it you are not hit with the other bug that runs out of disk space after the clocks changed recently?

[staze]

Not that I’m aware of. The space available for the SD card looks correct. Unless some other partition is filling.

Where’s that bug mentioned?

[ConligWX]

Not that I’m aware of. The space available for the SD card looks correct. Unless some other partition is filling.

Where’s that bug mentioned?

viewtopic.php?f=56&t=13747

its not affecting me (yet) but it does seems an issue for some.

[staze]

emailed info. Let's see if Boris can help.

Problem has definitely gotten worse recently and not better.

[staze]

what's normal turnaround time for getting contact back? I'm curious if I should bother leaving my Nano SD in it's "broken" state?

[ConligWX]

Usually a few days I guess, I'm not sure if Boris does this full time or like the rest of us, has another job, and Meteobridge is done in his spare time.

[staze]

kk. have rebooted the Nano since the outage was 24h at this point.

Here's hoping I can get it to fail again when Boris is ready.

[staze]

Boris was able to help. So far so good after 2 days.

Seems like I might have had two issues. First was the script polling template.cgi. But in an effort to fix that I'd tweaked my "retry" counts for several services up to the max and that was causing the buffer filling.

Now I just need to figure out why Meteotemplate fails so frequently. =(