Thanks for the insights Tom!

The thought of corrupted data on the logger has also crossed my mind, but 
weewx builds a new database right past the problematic timestamp.

However, the old database file stops at 1772604000, which is March 4, 2026 
at 12:00:00 AM CST, rather than March 8, 2026 03:00:00 AM CDT.

The new database goes back to February 27, 2026 at 3:20:00 PM CST.

As far as the logs are concerned, setting debug = 1, doesn't print any 
information on startup. I have restarted 5.3.1 several times w/o any output.

The only thing in the /var/log/messages file are comm errors:

Mar  8 23:48:18 mars weewxd[23459]: CRITICAL weewxd: Caught WeeWxIOError: 
LOOP max batch errors (3) exceeded.

Mar  8 23:48:18 mars weewxd[23459]: CRITICAL weewxd:     ****  Waiting 60.0 
seconds then retrying...

Mar  9 00:00:04 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 00:05:03 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 27 instead

Mar  9 00:05:23 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 00:38:47 mars syslogd: last message repeated 1 times

Mar  9 00:39:05 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #2; error: Expected to read 99 chars; got 0 instead

Mar  9 00:56:51 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 01:13:05 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 62 instead

Mar  9 01:29:04 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 01:47:53 mars syslogd: last message repeated 1 times

Mar  9 02:03:13 mars syslogd: last message repeated 1 times

Mar  9 02:03:59 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #2; error: Expected to read 99 chars; got 0 instead

Mar  9 02:37:24 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 02:38:00 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #2; error: Expected to read 99 chars; got 0 instead

Mar  9 03:28:44 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 03:29:38 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #2; error: Expected to read 99 chars; got 0 instead

Mar  9 03:29:48 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #3; error: Expected to read 99 chars; got 0 instead

Mar  9 03:29:48 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP max 
batch errors (3) exceeded.

Mar  9 03:29:48 mars weewxd[23459]: CRITICAL weewxd: Caught WeeWxIOError: 
LOOP max batch errors (3) exceeded.

Mar  9 03:29:48 mars weewxd[23459]: CRITICAL weewxd:     ****  Waiting 60.0 
seconds then retrying...

Mar  9 03:45:44 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 04:02:08 mars syslogd: last message repeated 1 times

Mar  9 04:03:30 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #2; error: Expected to read 99 chars; got 0 instead

Mar  9 04:19:54 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 04:37:04 mars syslogd: last message repeated 1 times

Mar  9 04:54:04 mars syslogd: last message repeated 1 times

Mar  9 04:54:14 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #2; error: Expected to read 99 chars; got 0 instead

Mar  9 04:54:50 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #3; error: Expected to read 99 chars; got 0 instead

Mar  9 04:54:50 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP max 
batch errors (3) exceeded.

Mar  9 04:54:50 mars weewxd[23459]: CRITICAL weewxd: Caught WeeWxIOError: 
LOOP max batch errors (3) exceeded.

Mar  9 04:54:50 mars weewxd[23459]: CRITICAL weewxd:     ****  Waiting 60.0 
seconds then retrying...

Mar  9 05:27:20 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

Mar  9 05:27:38 mars weewxd[23459]: ERROR weewx.drivers.vantage: LOOP batch 
try #2; error: Expected to read 99 chars; got 0 instead

Mar  9 05:45:06 mars weewxd[55633]: ERROR weewx.drivers.vantage: LOOP batch 
try #1; error: Expected to read 99 chars; got 0 instead

I restarted weewxd toward the end of this timeframe. Nothing here.
On Sunday, March 8, 2026 at 8:38:45 PM UTC-5 Tom Keffer wrote:

> Your symptoms smell like corrupted data in your logger, but without the 
> logs, we cannot be sure. Humor me, set debug=1, restart weewxd, let it run 
> through the first reporting cycle, then post the logs.
>
> The database is not your problem. If it's working for you, there is no 
> reason to "upgrade" from the old schema to the new schema. It just offers 
> more types. 
>
> On Sun, Mar 8, 2026 at 2:10 PM Marius Schamschula <[email protected]> 
> wrote:
>
>> I'm not sure what the log would tell you in this case. In the weewx.conf 
>> file for 4.10.1 had debug = 1 set. I have now also set that for the new 
>> install. I see nothing useful in the log file since I did that, i.e. no 
>> errors other than the ones I always see (e.g. ERROR 
>> weewx.drivers.vantage: LOOP batch try #1; error: Expected to read 99 chars; 
>> got 46 instead). It did show that an old cron job that was used to track 
>> the active tty caused multiple instances of weewx to run (FreeBSD changes 
>> the port from /dev/ttyU0 to /dev/ttyU1 or in reverse at random times). I 
>> commented that out in crontab to make sure only one instance is active. I 
>> tried one more time with the old .sdb file. Monitoring the /var/log/message 
>> file, I see no errors in the log. As before, the database doesn't update 
>> and a web page for 3:00 AM is rendered.
>>
>> As the same thing happens with both weewx 4.10.1 and 5.3.1, but a clean 
>> database file works correctly with 5.3.1, so there is no problem with the 
>> basic functionality of both FreeBSD packages and weewx.
>>
>> P.S. I'm leaving the root based install in place, as that is no different 
>> than what I was using for 4.10.1. No warnings of the sort:
>>
>> Installing collected packages: pyserial, ephem, pyusb, PyMySQL, CT3, weewx
>>
>>   WARNING: The scripts pyserial-miniterm and pyserial-ports are installed 
>> in '/home/marius/.local/bin' which is not on PATH.
>>
>>   Consider adding this directory to PATH or, if you prefer to suppress 
>> this warning, use --no-warn-script-location.
>>
>>   WARNING: The scripts weectl and weewxd are installed in 
>> '/home/marius/.local/bin' which is not on PATH.
>>
>>   Consider adding this directory to PATH or, if you prefer to suppress 
>> this warning, use --no-warn-script-location.
>>
>> when I installed as root. Of course, I've moved everything onto the ZFS 
>> array, rather than leaving it on the boot drive.
>>
>> BTW: What happened to $HOME/bin? That's already in the $PATH.
>>
>> In my experience some of the python version requirements are out of date 
>> or too strict, i.e. often you can ignore them and everything works 
>> correctly.
>>
>> And yes, the venv install installed a newer version of py311-pillow than 
>> the current one provided by FreeBSD ports. So much for the install 
>> instructions for FreeBSD!
>>
>> I now have to redo all the customization of the skin for both the private 
>> and public versions of the web site. It's been a while since I first did 
>> that so I have to rediscover what all needs to be edited.
>>
>> I still need to get my all old data back!
>>
>> I see that 5.3.1 uses an extended schema by default, while 4.10.1 was 
>> still using the old wview schema (yes, I did start out with wview). Could 
>> there be a problem with the database using the old schema?
>> On Sunday, March 8, 2026 at 2:56:06 PM UTC-5 Vince Skahan wrote:
>>
>>> Sure.  The other os have python libs they install at the system level 
>>> too.  Run 'pip3 list --verbose' to see what's where.
>>>
>>> I'm personally ok with things installing as root.  I just like not 
>>> 'running' as root whenever it can be avoided.
>>>
>>> FWIW - I have run into the problems venv solve quite a lot.  Things 
>>> needing certain versions of a library (or minimum versions thereof) and the 
>>> os freezing to something lesser.  Any RHEL-like system or LTS debian is a 
>>> good example.  They go for many year stability and freeze to old versions.  
>>> The venv thing is a good way to not be limited by the os vendor's choices.
>>>
>>> On Sunday, March 8, 2026 at 12:45:33 PM UTC-7 Greg Troxel wrote:
>>>
>>>> Vince Skahan <[email protected]> writes: 
>>>>
>>>> > You generally can't avoid venvs on a modern python on a current os. 
>>>> Nobody 
>>>> > here did that. The python project forced that on everybody. 
>>>>
>>>> That's not strictly true. pkgsrc has packages for a vast number of 
>>>> py-foo all installed in the system site-packages, and it works fine. I 
>>>> am actually running weewx that way, with the weewx code in 
>>>> /usr/pkg/lib/python3.13/site-packages/weewx and so on. 
>>>>
>>>> In my case, the weewx progarm files are owned by root and live in the 
>>>> system. I am running it in a data directory (with config file and 
>>>> database) that is owned by a non-root user. Stepping back from weewx 
>>>> and python, this is totally normal, to use installed programs with your 
>>>> own data. 
>>>>
>>>> <rant> 
>>>>
>>>> I find that venvs are required because python culture says it is ok to 
>>>> have requirements as foo==x.y.z, rather than foo>=x.y. Thus, there is 
>>>> no way to have everything needed installed, and python packages with 
>>>> unreasonably specific dependencies (Home Assistant) have to be in a 
>>>> venv 
>>>> for isolation. 
>>>>
>>>> The root cause of pinned deps, besides people thinking it is ok, is API 
>>>> instability within modules. 
>>>>
>>>> </rant> 
>>>>
>>> -- 
>> 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 visit 
>> https://groups.google.com/d/msgid/weewx-user/1e3e0173-f581-45cd-a954-d8ec514b558dn%40googlegroups.com
>>  
>> <https://groups.google.com/d/msgid/weewx-user/1e3e0173-f581-45cd-a954-d8ec514b558dn%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>

-- 
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 visit 
https://groups.google.com/d/msgid/weewx-user/0e70d3d6-062b-49ed-a8cf-580f9bdd5b64n%40googlegroups.com.

Reply via email to