BTMon dies

Post any 3rd party software here.
mwall
Posts: 140
Joined: Wed Dec 07, 2011 6:25 pm

Re: BTMon dies

Post by mwall » Fri Jan 25, 2013 6:16 pm

it looks ok from the client side of things, and the 'Status: 200 OK' response from the server indicates that SEG received the data.

you might want to ask sam at SEG if there is a limit to the number of channels that can be posted at once.

check with sam to see if the SEG server logs are complaining about your postings.

also, there might be a limit to the length of a string when making an http post - i'll check the specs and standards.

you can test to see whether the length is the issue by doing one of the following:
1) post to SEG every 15 seconds instead of every 60 seconds (seg_upload_period)
2) use a map to upload only 2 or 3 channels instead of all of them

fwiw, i have periodically experienced gaps in data at SEG when uploading from 8 ECM all at once. btmon gets the 'Status: 200 OK' but no data show up at SEG. once it was due to some server maintenance in progress at SEG. two other times no explanation, and things just started working again after awhile.

generally speaking SEG has been very reliable, but the GEMs might be hitting it with more data than it was ready for. most of the online services are targeted toward whole-house monitoring, so 32 channels is a lot to them.
mwall
Posts: 140
Joined: Wed Dec 07, 2011 6:25 pm

Re: BTMon dies

Post by mwall » Fri Jan 25, 2013 6:23 pm

Tirnanah wrote:I am on firmware 1.70 (Engine ver 1.40). Could it be something with that?
not likely. communication with the GEM looks ok.

if you are concerned about losing data, and since you are saving to database already, you might want to run two instances of btmon - one to collect data and save it to database, and a second to upload data from the database to SEG (and any other services). see the section about running multiple instances of btmon in the guide:

http://lancet.mit.edu/mwall/projects/po ... howto.html
mwall
Posts: 140
Joined: Wed Dec 07, 2011 6:25 pm

Re: BTMon dies

Post by mwall » Fri Jan 25, 2013 6:25 pm

also, do not leave your configuration posting to SEG at 15 second intervals! it is ok for testing, but once you sort things out switch back to 60 or 90 second intervals.
Tirnanah
Posts: 60
Joined: Fri Jun 03, 2011 1:48 am

Re: BTMon dies

Post by Tirnanah » Fri Jan 25, 2013 6:37 pm

I did change the interval to 60 seconds. I don't remember changing it, but I must have. I've also send an email to Sam to see if can check his side.

Thanks for the continued support.

Doug
Tirnanah
Posts: 60
Joined: Fri Jun 03, 2011 1:48 am

Re: BTMon dies

Post by Tirnanah » Tue Jan 29, 2013 1:32 pm

I may have figured this out.

mwall: Does btmon make the adjustment for watt hours prior to upload to SEG? One of the things in the SEG/GEM setup documentation is to put a calibration (multiplication) of .000277 on the SEG site to make the energy data display correctly.

I created an entirely new device to start fresh, and noticed that as soon as I put that calibration in, all of my energy values dropped to 0 and I cannot get any data from them now, whereas before, there was still a little bit of data that was building correctly.

Thanks!
Doug
ben
Site Admin
Posts: 4262
Joined: Fri Jun 04, 2010 9:39 am

Re: BTMon dies

Post by ben » Tue Jan 29, 2013 1:45 pm

Tirnanah wrote:mwall: Does btmon make the adjustment for watt hours prior to upload to SEG? One of the things in the SEG/GEM setup documentation is to put a calibration (multiplication) of .000277 on the SEG site to make the energy data display correctly.

I created an entirely new device to start fresh, and noticed that as soon as I put that calibration in, all of my energy values dropped to 0 and I cannot get any data from them now, whereas before, there was still a little bit of data that was building correctly.
These calibration values are unique to our direct GEM posts to SEG to convert wattseconds to watthours (1/3600000) to take some strain off the GEM itself.

btmon.py looks to send delta watt hours, so you don't need the same value as above.

Ben
Ben
Brultech Research Inc.
E: ben(at)brultech.com
Tirnanah
Posts: 60
Joined: Fri Jun 03, 2011 1:48 am

Re: BTMon dies

Post by Tirnanah » Tue Jan 29, 2013 1:49 pm

Thanks Ben. I'll take those calibration values back out and see what happens.

Thanks,
Doug
Tirnanah
Posts: 60
Joined: Fri Jun 03, 2011 1:48 am

Re: BTMon dies

Post by Tirnanah » Tue Jan 29, 2013 3:39 pm

Success. Data starts to flow again.

Thank you everyone for all the help.

Doug
Post Reply