🥇Overvågning af UPS. Del to - automatisering af analysen

For some time, I have been working on a system to assess the viability of office UPS units. The assessment is based on long-term monitoring. As I used the system, I refined it and learned many interesting things, which I will share – welcome under the cut.

Første del

Overall, the concept turned out to be correct. From a one-time request to the UPS, one can only find out that life is painful. Some parameters only relate to reality when not connected to 220 V; some turn out to be outright nonsense based on analysis, and some need to be recalculated manually, cross-referencing with reality.

To get ahead of myself, I tried to incorporate these nuances into the system. After all, we are automators, not just doing things manually.

Here, for example, is the parameter "battery charge percentage". As a single value, it conveys nothing and is usually equal to 100. What really matters is how quickly the battery discharges, how quickly it charges, and how many times it has been depleted to critical levels. Surprisingly, part of this work is done by the UPS itself, though through some very strange formulas; more on this below.

The parameter "load on the UPS" is quite good and useful. However, looking at it dynamically, it turns out that sometimes it's nonsense, while other times it provides interesting information about the connected equipment.

«Battery voltage". Practically the Holy Grail, if it weren't for one thing: the vast majority of the time, the battery is in charge, and the parameter shows the charge voltage, not the battery voltage. Wait, shouldn't this be handled by the self-test procedure?..

«Self-test". It should, but its results are not output anywhere. In the event of an unsuccessful self-test, the UPS will turn off and scream like it's being cut, and that's the only available result. Plus, not all UPS units report even the fact that a self-test has taken place.

And "nice try vendor" – the most interesting of the available parameters is "battery runtime". It is supposed to predict how long the battery will last under the current load. It also drives the internal logic of UPS behavior. In reality, it shows rosy dreams, especially when fully charged.

There are also some nuances of an organizational nature.

For example, all the UPS units I encountered have data about the battery date (two fields, in fact). However, I could only record this information (after replacing the battery, of course) with APC products, and that took a lot of effort. With Powercom, this information simply cannot be included, at least not under Windows.
The same Powercom had identical values in the 'serial number' field. It cannot be recorded either.

The calculation of 'battery runtime', apparently, includes values from periods when the UPS was connected to 220 V, and therefore, the battery data is blatantly incorrect. In fact, the battery runtime can confidently be divided by 2, or even by 3. And still, it remains purely a synthetic value. Furthermore, it is based on 'battery load', which has also shown some inconsistencies: on some units it does not reset quickly after a high load, while on others, it tends towards zero.

Despite such a mishmash, it can be seen that all parameters are still subject to some algorithmic processing. This means that one can not only look at the data (and certainly not manually go through all available records), but can immediately input the entire dataset into an analyzer and build recommendations based on them. This has been implemented in the new version of the software.

The UPS information page will issue warnings and recommendations:

at least one self-test failure has been recorded (if the UPS provides such functionality)
the need to replace the battery
unusual load values on the UPS
absence of battery data
unusual input voltage values
recommendations for data usage and UPS maintenance

(all possible options can be found in ups_additional.php)
A necessary condition for accurate analytics is, of course, the maximum possible data collection.

On the main page, you can immediately see the maximum and critical values along with the corrected prediction of runtime.

And also:

now the maximum power loss time is calculated correctly
current information from the UPS is marked in green, outdated information in gray, and critical in red and orange
a database optimization procedure has been added (it runs manually, automatically creating a backup)
unnecessary information has been removed from the main screen and useful information has been added 🙂

Ansvarsfraskrivelse:
Selvfølgelig, dette er slet ikke enterprise. Næsten hele installationen er manuel. Der har været for få tests, og der opstår fejl hist og her. Alligevel bruger jeg det nyttigt, og jeg ønsker det samme for jer.
github.com/automatize-it/NUT_UPS_monitoring_webserver_for_Windows