GEM/Standalone Dashboard - Bad Packet Not Numeric Error

[ben]

So I have been keeping an eye on this issue.
The packet errors disappear after rebooting the GEM or software rebooting the module via the 'device management' tab. But after a day or so they return.
The network interface is running: 4.02.10.C1224.01
I am using direct the ethernet connection and have disabled the Wifi AP via the GEM (Adv Tab).

Is it possible that the ethernet interface is faulting? How would I check this?

Marty.

In the GEM Setup Webpage, can you post the "Status" section?

[mminehan]

Here's a cut and paste from the status page. I Have hidden the serial number though. Let me know if you need screenshot.

Current Settings
Serial#: XXXXXXXX
COM firmware ver: 4.33
ENG firmware ver: 1.49
RealTime Status: ON
Packet Send Interval: 10
Pri. Packet Format: 4
Sec. Packet Format: 0
Sys Status: OK
Wifi/Ethernet Module: Enabled
GEM Ver: Single or Polyphase
PCB Ver: 3
BootLoader Ver: 2
COM1 Flow: OFF
COM2 Flow: OFF
Current Constant: 221
Sys Info: 00100001 OK
Sys Flags: 00000000 OK
Keep Alive String:""
COM1 Baud: 115200
COM2 Baud: 19200
Phase Setting: 3 Phase
Freq: 50Hz
Hardware Configuration: 166
WiFi and Ethernet V2

[Teken]

Marty,

Are you really using 3 phase @50Hz??

[ben]

Everything looks OK there.

The issue is it could software or hardware related. We did some testing awhile ago and didn't find any issues with the checksum.

Does it affect the operation of UltraECM at all (missing data?) or is it just filling up log files?

[mminehan]

Thanks for the responses Teken and Ben.
It hasn't faulted for a while, no checksum errors. But they have returned today. A reboot of the GEM Ethernet module via the 'device management' tab seems to fix it for several days.

@Teken Yes we are really running 50Hz 3 phase. In New Zealand our grid supply is 50Hz. But we only have 2 of the 3 phases connected to our house. That came about because we have an induction stove top that can use up to 11kW if all 5 elements are on full. And a 7kW EV charger. This would have exceeded the safe supply limit of a single phase. It's actually a 3 phase cable, but only 2 are connected. This caused some problems in that we had to install 'phase couplers' for our Enphase solar system that uses power line signaling between the gateway and microinverters. The solar array is split between the two phases. Also had to make sure every CT on the GEM was set to the correct phase.

@Ben It's not having a major effect on HS. The entries apper in the log (generated by the plugin). But the plugin seems to be functioning and the devices get updated. But the update interval (packet send) is every 10 seconds. It's hard to tell but I suspect that when the checksum error occurs that particular update in missed. When the erors start they occur 3 or 4 times an hour, so its hard to trap the exect error. So yes.....more annoying that it shows errors in the HS log than any significnt loss of functionality.

[shortwh]

So, I started having this issue with an extraneous colon when my watt-seconds numbers became REALLY large (several years of running). Below I will show you packet dumps from the packet before the problem starts and first packet after then I will give the packet before the problem disappears and the next good packet. Watch channel 18 as the second field grows then when it should rollover to a higher number it suddenly has a colon in the field. I remember reading somewhere that you are using 48bit storage values for the watt-second fields. Could this be a math problem building the 48bit value from smaller integer values?

COM: 5.13
ENG: 1.49

Last Good packet before failure:
192.168.1.110 - - [05/Feb/2021:11:42:53 -0600] "GET /index.php/ecmEngine/greeneye?SN=xxxxxxxx&SC=16769557&V=1248&c1=385684164626,7865822,11.90&c2=3480128051,11,0&c3=49049783402,0,0&c4=15160947604,0,0.94&c5=18712509026,0,0&c6=10203044860,0,0&c7=38758462040,0,0&c8=3160104989,0,0&c9=12691466461,1474265,0.10&c10=8427536174,585452804,0&c11=25754479511,0,4.42&c12=15673174759,0,0.12&c13=3517430908,0,0.08&c14=49710240976,0,0.12&c15=6339791366,0
,0&c16=400096337765,393613440669,160.12&c17=175946168911,175130984827,19.54&c18=100670835479,99999954513,10.80&c19=98213466359,97578306670,10.70&c20=3833950108,0,0.54&c21=2907191842,126861536,0.34&c22=41043426235,39949850062,39.66&c23=41390234638,0,0&c24=2104320550,0,0&c25=43472821421,0,3.26&c26=7372804286,0,0.18&c27=9866284176,0,2.72&c28=6894352207,0,0.68&c29=754666358,0,0.62&c30=703012798,0,0&c31=1264471997,0,0.32&c32=1057388076,0,0.3
2&PL=1,1,1,1&T=48.0,58.0,50.5,,,,,&key=key1_here&Resp= HTTP/1.1" 200 - "-" "-"

First Bad packet after failure:
192.168.1.110 - - [05/Feb/2021:11:43:13 -0600] "GET /index.php/ecmEngine/greeneye?SN=xxxxxxxx&SC=16769577&V=1248&c1=385684190426,7865822,12.24&c2=3480128051,11,0&c3=49049783409,0,0&c4=15160949841,0,0.94&c5=18712509026,0,0&c6=10203044860,0,0&c7=38758462040,0,0&c8=3160104989,0,0&c9=12691466558,1474265,0.12&c10=8427536174,585452804,0&c11=25754487435,0,4.42&c12=15673174858,0,0.12&c13=3517430968,0,0.08&c14=49710241741,0,0.12&c15=6339791366,0
,0&c16=400096690396,393613793300,166.12&c17=175946253585,175131069501,20.30&c18=100670882829,:0000001863,11.52&c19=98213513339,97578353650,11.28&c20=3833950799,0,0.54&c21=2907192242,126861536,0.34&c22=41043602896,39950026723,40.82&c23=41390234639,0,0&c24=2104320552,0,0&c25=43472827520,0,3.34&c26=7372804469,0,0.16&c27=9866289167,0,2.96&c28=6894353173,0,0.68&c29=754667713,0,0.60&c30=703012798,0,0&c31=1264472231,0,0.32&c32=1057388290,0,0.2
8&PL=1,1,1,1&T=48.0,58.0,50.5,,,,,&key=key1_here&Resp= HTTP/1.1" 200 - "-" "-"

Last Bad packet before GEM corrects:
192.168.1.110 - - [05/Feb/2021:14:41:12 -0600] "GET /index.php/ecmEngine/greeneye?SN=xxxxxxxx&SC=3041&V=1248&c1=385707336844,7865822,12.52&c2=3480128051,11,0&c3=49049787600,0,0&c4=15161457324,0,1.08&c5=18712509026,0,0&c6=10208736296,0,0&c7=38758462085,0,0&c8=3165111545,0,0&c9=12691645525,1474265,0.12&c10=8427536260,585452804,0&c11=25759018551,0,6.66&c12=15673374146,0,0.12&c13=3517627264,0,0.08&c14=49710646679,0,0.12&c15=6339791368,0,0&c16=400222617928,393739720832,134.96&c17=175976585981,175161401897,14.92&c18=100687648908,:0016767942,8.64&c19=98230144527,97594984838,8.58&c20=3834323941,0,0.52&c21=2907402424,126861536,0.34&c22=41106717809,40013141636,35.84&c23=41390237806,0,0&c24=2104322571,0,0&c25=43476186116,0,3.30&c26=7373735659,0,0.16&c27=9867276038,0,1.36&c28=6894590839,0,0.36&c29=755003288,0,0.16&c30=703128455,0,0&c31=1264593707,0,0.32&c32=1057501263,0,0.28&PL=1,1,1,1&T=55.0,61.5,59.0,,,,,&key=key1_here&Resp= HTTP/1.1" 200 - "-" "-"

First Good packet after GEM corrects:
192.168.1.110 - - [05/Feb/2021:14:41:32 -0600] "GET /index.php/ecmEngine/greeneye?SN=xxxxxxxx&SC=3061&V=1248&c1=385707363583,7865822,12.36&c2=3480128051,11,0&c3=49049787602,0,0&c4=15161459911,0,1.08&c5=18712509026,0,0&c6=10208736296,0,0&c7=38758462085,0,0&c8=3165111545,0,0&c9=12691645637,1474265,0.10&c10=8427536260,585452804,0&c11=25759030493,0,6.38&c12=15673374247,0,0.12&c13=3517627324,0,0.08&c14=49710647432,0,0.12&c15=6339791368,0,0&c16=400222855858,393739958762,127.60&c17=175976641655,175161457571,15.50&c18=100687680397,100000022215,8.92&c19=98230175360,97595015671,8.38&c20=3834324639,0,0.52&c21=2907402821,126861536,0.34&c22=41106839481,40013263308,31.76&c23=41390237807,0,0&c24=2104322575,0,0&c25=43476192470,0,3.26&c26=7373735841,0,0.16&c27=9867278838,0,1.36&c28=6894591299,0,0.36&c29=755003688,0,0.16&c30=703128455,0,0&c31=1264593924,0,0.28&c32=1057501477,0,0.26&PL=1,1,1,1&T=55.0,61.5,59.0,,,,,&key=key1_here&Resp= HTTP/1.1" 200 - "-" "-"

[ben]

Pauls looking into it, one thing you guys have in common is its shortly after it rolled over from 9999.... to 10000...