On 16/08/2018 10:32, DrTeeth wrote:
> On 15 Aug 2018 13:06:32 GMT, just as I was about to take a herb, Tim+
> <
tim.d...@gmail.com> disturbed my reverie and wrote:
>
>> Well if I hadn’t been sure about the cell I was in it would have been a
>> good idea. As it was, the problem was very clear, very localised and very
>> consistent. I’m also not too sure of the exact mast position so I would
>> have had to hunt it down. Anyhow, it’s fixed now and I’ll certainly look
>> next time I want to report a data-dead cell.
>
> Shouldn't the company be monitoring the masts themselves? It seems
> strange that they seem to need to rely on customer reports.
>
That kind of failure is detected using statistics, not on/off alarm
indications.
The failure will only be flagged when the report is run, and only if the
data indicates a problem. Sometimes "abnormal" for one site is "normal"
for another, so the periodic report simply won't flag it up.
Even then, the statistics driven reports will only be used to trigger
proactive/"fill in" tasks - jobs tasked to field engineers when they
haven't got an emergency outage to deal with (e.g. dead site).