Running two APC UPS devices [Closed]

[rodgoslin]

Some years ago, due to frequent power cuts, I installed an APC Back-UPS ES700 unit. This has worked well, but with my system getting somewhat more elaborate, with more storage units, i decided to buy another UPs. One to run the PC and ancilliary equipment, and another, largely to cope with three Drobo NAS units, and other utilities. The software in the Mageia repositories provides for a command line interface, which will display information as to the running of the UPS. It was quite some time ago that the software was installed, and information is a bit on the short side. Both units are connected to the PC by USB, but only the details of the original UPC are available. Scratching about, I've been unable to find any information regarding having two UPS units running.
Any ideas, anyone?

On a slightly related subject, I'm a bit curious in regard to the somewhat brief duration of the battery backup. To all intents and purposes, the sealed lead-acid battery is nothing outstanding, and of a somewhat limited capacity. A lead-acid battery is a lead-acid battery, is a lead-acid battery. There are variations for heavy current drains and for deep discharges, which would appear suitable as an external power source. Looking for information on this I get conflicting statements. Yes, you can use external batteries. No, you cannot use external batteries. The only reason for the limited size of the fitted battery that I can come up with is that the heat generated in the UPS over a longer period of time might damage the system, but then cooling fans are cheap, and quiet. For something like twice the price of a replacement 10AH battery, I can buy a 130AH leisure battery, which will keep the system up for hours.

[jiml8]

I use gapcmon, which has worked quite well for me for many years now. I only control 1 UPS on my workstation, but the gapcmon GUI shows a capability to configure/control multiple devices. I have never tested that capability.

As for the batteries, sure you could use an external battery, if you wanted to wire the device up to handle it. I don't know how well the charger would like a much larger battery than it is designed for; seems to me the internal resistance of such a battery might be low enough that the charger would over-current its outputs (unless they are explicitly limited...I don't know if the design of the APC UPS does that or not.

As for the duration of the built-in battery, basically you buy the amount of time you are willing to pay for. Larger batteries cost more, both initially and when they have to be replaced. The only criterion here is cost; are you going to hold your system up during a 4 hour blackout because a car hit the transformer that services your neighborhood? Or do you just want to hold it long enough for it to undergo a safe shutdown?

[rodgoslin]

I'd rather the system stay up longer than the time required to take it down properly. In the last, and recent power failure, it was the middle of the night. By the time I'd come to enough to recognise the beeping, the system was down. On the other hand, Linux seems robust enough to ignore unscheduled shutdowns and the UPS does not have the capability of handling the NAS units shutdown, anyway.
The cost of a larger battery, as I pointed out, is not really a factor. UPS batteries are very expensive for what they are. One with thirteen times the capacity, costs significantly less than twice the cost of the official replacement. They're both high current, deep discharge, sealed batteries. Or at least sealed in the respect that neither has the facility, or the need for electrolyte top-ups.
I live in a rural area, in a house without organised central heating, open fires, and it would be nice if I had at least one access to entertainment when the lights go out. The UPS also provides telephonic services, since there's no mobile signal in the area, through a femto cell unit, since I junked the landline telephone when they brought in fttp.
The UPS software does provide for multiple UPS units, The following snip from the man pages does so indicate.

-P file --pid-file file
Create the specified process ID file. The default is /var/run/apcupsd.pid. It must be changed when run‐
ning more than one copy of apcupsd on the same computer to control multiple UPSes.

Unfortunately, that is all it does say, I've been unable to find out how the apcupsd.pid file should be structured, or how it should be used to facilitate a second UPS. I've raised a query in the official APC forum, but, so far, there's been no replies to it.
Charging a larger battery should present no problems. An internal charger for the standard 10AH battery will be unlikely to be able to output more than one Amp anyway, no matter what the capacity of the external battery is. This can be independently charged, anyway.

[doktor5000]

Usually the only thing a .pid file contains is the PID (process ID) of a running process, in this case of apcupsd. So your question should probably about how to configure apcupsd for multiple devices.
Maybe check e.g. https://wiki.debian.org/apcupsd#Configu ... Devices.29

[rodgoslin]

Thank you Doktor500. That is most helpful. However, the article in the URL, does say that the setup described is for a Sys-V setup and that there are problems with it's use on a Systemd setup. In addition, the whole thing is extremely complicated. The software has limited effect on a NAS system as far as I'm able to judge, so would probably not help in a tidy shutdown. The NAS units have never exhibited any problems on re-booting from an unplanned shutdown, so I think that I'll forget about the whole thing. If I ever decide to monitor the second UPS, it's a simple task to simply unplug the USB lead from the one UPS and plug it into the other. From the config file, there doesn't seem to be any factors that will prevent the display of the data. Oddly, for some reason best known to them selves, APC have used a RJ45 connector on the UPS, and a standard USB2.0-A connector at the other end of the cable
I'll set this thread as closed.