For whatever reason, sometimes linebreaks are only denoted as \n in the
querry-result. This lead to the temperature RegEx matching the dew point. This
caused the temperature to be reported the same as the dew point. These changes
should make the RegEx more specific and work consistently, no matter if
linebreaks are actual linebreaks or just \n.
This adds a plugin for relative humidity, based on the weather_press_ and
weather_temp_ plugins. It's basically a copy of these two plugins with only
minor adjustments.
The read function on an urllib urlopen object returns an object as a response.
Regular expressions using re can't be used on such objects. This causes
the following error:
```
Traceback (most recent call last):
File "/tmp/weather/./weather_press_LOWW", line 43, in <module>
hpa = re_hpa.findall(txt)[0]
TypeError: cannot use a string pattern on a bytes-like object
```
This can be easily fixed, because said object can simply be cast to string.
Which is, what this patch does for both the US NOAA based plugins.
add option verify to enable graphing the number of successful responses.
if verify is enabled (default it is not) then a warning is emitted when
we do not get a response for all queries and critical if no query
succeeded.
add option timeout to be able to limit the time the plugin takes for
unresponsive servers. old max time was 75 seconds times 20 (25minutes).
new default max is 30 seconds timeout times 20 (10 minutes). with
setting a timeout of a few seconds this can now be reduced even more.
added warnings if average and or mean get close to the timeout.
Doesn't work with '--write-mostly' drives like sde1 in :
md94 : active raid1 sde1[1](W)(S) sdd1[2]
I assume to remove everything behind the opening bracket
* lxc_guests2: Enhanced and fixed lxc_guests plugin
- Working with systemd and cgroup version 2 (tested on debian bullseye and debian booksworm)
- Fixed processes (with cgroup version 2)
- Simplified labels
- Simplified memory usage graph
- Added CPU usage in percent (using systemd-cgtop)
- Added Tasks
- No cgrouppath guessing
* Rename lxc_guests2 to lxc_guests to replace old plugin
---------
Co-authored-by: Sebastian L <sl@momou.ch>
* deye: fix production calculation
* solarman: add new plugins
* fix labels
* move api check
* Typo
* deye: quoting
* unfiy labels
* Typo
* Test, set Value to zero if api does not find device
* dont show errors
* Revert "dont show errors"
This reverts commit dc71783f93.
* Revert "Test, set Value to zero if api does not find device"
This reverts commit fa1853955c.
* dependency check
* remove exports
* send 0 instead of U
* dependency check
* remove exports
* check dependency removed
* send U when API isnt reachable
* add AC graphs
* Typo
* Typo
* Get all Panels
* remove var
* Typo
* Typo
* Typo
* packe Logik in Funktionen; Frage Daten im Fehlerfall mehrfach ab
* Zähler umgebaut
* Remove Panel Plugin
* deactivate panel 3+4
* api reachable check
* unknown limits
* warnings
* area
* deye: printf http://munin-monitoring.org/faq#q-why-does-my-users-plugin-report-floating-point-numbers
* deaktiviere graph scaling
* setze base
* schreibe retries
* graph opts
* statefile
* deye statefile
* x
* Info
* add Plugin for Deye Solar Inverters
* better Titles
* fill Area
* Typo
* check if Inverter is reachable
* even better Titles
* changed Header to perl-Style
* Typo
* Typos
* Fix Values
* Set yield to unkown when offline
* Escaping
Default for Python 3 is to return byte-objects, not string. This will cause the plugin to fail with an error: "TypeError: a bytes-like object is required, not 'str'".
Fixed by decoding for UTF8.
This actually silences InsecureRequestWarning.
The Arris SB8200 has 32 downstream channels, but in my experience
report more than that, but the "extra" channels report modulation as
"Other" instead of QAM256. This filters those channels, since they seem
to drive the error count up, but don't provide data.
Signed-off-by: Nathaniel Clark <Nathaniel.Clark@misrule.us>
It doesn't work (looks like calling outdated python lib) and had the same
funcionality than the newer transmission_ plugin. It is quite confusing now
they are in the same directory, but the documentation (README) mentions
only the other one.
