>> That of course, is also a possibility. In which case, there is probably >> nothing we can *ever* do about this. If everything looks normal before >> the >> tests, but the UPS fails during the test, we're toast. There can be a >> number of failures for this, not limited to the battery alone (a relay >> failing for instance, which is a nasty failure mode for line interactive >> UPS'es or bypass switches). > I tried newmge-shut which has a lot more information available.
The latest version from the trunk might even make more information available. :-) > I ran a quick battery test with report no error but made the UPS start > beeping and the a deep battery test which really reported an error and > set the RB status and caused upsmon to report a warning message on > console. If the UPS started beeping after the quick test, chances are that it also set an alarm. Since there haven been quite some additions to the mge-hid subdriver recently (by yours truly), there is an odd chance that the latest development version could pick up the error. > This seems to me a hardware bug as the quick test did not report > anything wrong. Not all problems can be revealed by a quick test, which is the one that is scheduled automatically by the UPS itself. The capacity of the batteries can only be accurately determined through a deep discharge test. Usually, one would schedule this test at a time when it is not a big problem for the system to go down, as you won't have a lot of runtime left when the test is (almost) completed. > I ordered two sets of batteries as my older MGE Ellipse model is also in > bad shape. > > This is a good reminder that a test from time to time is good pratice :) Indeed. Don't overdo this though, since each time you run a deep discharge test, you'll reduce the capacity of the batteries. Once every three to six months should sufficient. In contrast to complete battery failure (which may happen at any time), wearout is quite predictable. When plotted against time, you'll see that the capacity of your batteries will slowly decrease. Best regards, Arjen -- Eindhoven - The Netherlands Key fingerprint - 66 4E 03 2C 9D B5 CB 9B 7A FE 7E C1 EE 88 BC 57 _______________________________________________ Nut-upsuser mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser
