Changes maybe req'd for xAP?
Hi Derek,
As you know I'm busy building my xAP remote control system for my Joggler.
I have now finally connected a CC monitor to my HAH, no problem.
When trying to get the level details to display on my Joggler I have come accross some issue with what is expected in the xAP message. I don't fully understand all this but as far as I can tell and have been reliably informed, our xAP messages from the CC monitor aren't well formed.
Aparently the xAP spec says levels should be formatted so: level=value/max eg level=944/10000
This allows for percentile calculations to made based on the value.
I've also noticed another issue, which has been highlighted in my endeavours. Sometimes my controller seems to loose any idea of what status the hah thinks the rf devices are in. Now I know that the devices don't actually respond but I would assume that correct or not, the HAH would also have an 'opinion' as to what the status should be. It seems that it doesn't always want to reply to requests for this opinion and so the status of my controller disappears.
I hope I've explained these well enough, I would be happy to do some more diagnostics, you'll just need to let me know what to do?
Thoughts?
Gary
Hi Gary,
The Joggler piece is looking very good - we hope to put some HAH specific deatil on using the Joggler on the Wiki before too long.
As Brett points out, there is a known issue with the CurrentCost adaptor. I'm actually beta testing a HAH firmware version that fixes this - Brett will make the call as to when this release goes public.
Re your other point about the RF device status. The HAH keeps internal tracker variables to hold the state of each RF channel. So long as you don't reboot the HAH, these should report correctly. Could you possibly use xFX viewer to catch an example of where the HAH fails to report properly? Also, does the HAH web UI reflect this 'unknown' RF status (like when you press the 'Refresh' button on the UI)?
Cheers,
Derek
Sorry for the terse response. I just started typing when SWMBO said it was time for dinner. build 252 will have a few things fixed in it:
- Support for state=toggle used by the joggler interface.
- A new action item "toggle" on the web interface.
- Twitter as a service deletes messges after its sent them. (fixed)
- 1wire bus device timeout check. A 1wire device reports when the temperature changes however if it gets disconnected the HAH won't know this so will continue to report its previous value. This timeout check allow the HAH to reset the value to ? if no change is detected withing a certain time frame. The timeout can be setup on the 1wire configuration web page. 0 will disable checking.
- The current cost issue that you and others have pointed out
Currently Derek is beta testing these changes to make sure things are ok. I've not got around to fixing the currentcost issue yet, its next on my list.
If you want to grab the BETA. Note: this will always overwrite your existing firmware - no version check is made, your /etc directory will be preserved but as with any update backup things you won't want to loose first just in case.
/etc_ro_fs/update-dev hah-beta.dbzoo.com
As I've just got settled in the US I've been a little behind in firmware fixes, hopefully once I get more establish I'll get some time for other improvements.
Brett
On IE8 and Chrome, with Build 251.8
If you have a large amount of data and/or pachube is slow it will take a few seconds for all the data to transfer to your browser before the Javascript/Flash can turn it into a graph. When I get time I'll check this out althought I've never had and/or seen the issues are describing. Not to say they aren't happening to you of course. :)
When you say you downloaded the beta I'm now assuming you're on 251.8 ?
See BUG http://code.google.com/p/livebox-hah/issues/detail?id=7