node-red-contrib-ccu 4.5.1

v4.5.1

on Node.js NPM

Fixed

• A burst of getParamsetDescription Error: XML-RPC fault: Invalid device
  at the first start after an update (B-36). A device deleted from the CCU
  while Node-RED was not running stayed in the cached device table
  (ccu_<host>.json) and in the values cache (ccu_values_<host>.json).
  At the start its paramset descriptions were queued from the cached table;
  hmipserver answered -2 Invalid device, a failed description was never
  remembered, so every lookup queued it again (one item per channel and
  paramset, duplicates included), each fault was logged twice without the
  address, and each one threw away and rebuilt the interface's RPC client.
  Seen on a CCU3 after a remote had been replaced: 150 failed calls in 40 s.
  Now a description the interface refuses with a fault is asked for once
  per address and paramset in a run, with one warning that names the
  address, the paramset and the fault (a device paired anew is asked
  again); the queue holds each description key once; deleteDevices drops
  the device's queued fetches and its address from the type lists, and the
  values cache is pruned against the device table at the start and a few
  seconds after a deleteDevices or the ReGa's value read (the ReGa keeps
  reporting the datapoints of a deleted device). An XML-RPC fault no longer
  closes and recreates the RPC client: only a transport error does; the
  fault is logged once, with the arguments of the call.

Commits since v4.5.0

• 4.5.1 - a stale cached device costs one refused paramset description, not a burst of them (8d0acf7)
• A stale cached device costs one refused paramset description, not a burst of them (B-36) (3de76b8)
• Specs on hm-simulator 1.3.1 (task 15, step 1) (96b8c69)