decrease keepalive sensitivity to reduce disconnections

[pb66]

See https://community.openenergymonitor.org/t/a-question-over-whether-emoncms-is-compatible-with-mqtt/7180/12?u=pb66

I suspect setting a keepalive of "5" is actually resulting in MORE disconnections than a more relaxed setting of "60". But it may be better to go for "180" or more to accomadate setups with just a couple of emonTH's.

PLEASE DON'T MERGE WITHOUT DISCUSSION AND CONSIDERATION OF THE CLIENT ID PR TOO!

[glynhudson]

This also sounds great. Shall I merge into master to open up the testing pool? All (or the majority) of emonPi/emonBase/emonSD users are running Stable. I can't see it causing any adverse issues

[borpin]

Is working for me with the keepalive removed completely so it falls back to the default. I should say though, I'm using #822 as well to catch any disconnections/reconnections.

[pb66]

It will have an adverse effect on anyone using non-timestamped mqtt which is 99.99% of users (Brian uses timestamped MQTT). Can we keep the discussion in the forum, in one place please? I could easily merge this if it was ready, I only created a PR to try and get some dialog going, that why I said "PLEASE DON'T MERGE WITHOUT DISCUSSION AND CONSIDERATION OF THE CLIENT ID PR TOO!".

[edit] The issue I was originally concerned about turns out to NOT be an issue. the pingreqs etc are working so the 5s keepalive is NOT causing a disconnection/reconnection, it is just causing excessive MQTT control packets, the extra sensitive keepalive setting is probably useful whilst we have connection issues, setting a keepalive of 60s could result in no data for 60s, the (depending on whether the clientid is used) it will post the 1min old data as "now" or the data will have been lost in the dis/re-connection.