Blink Integration not Arming and Disarming via Automation/Service

pete101 · November 4, 2020, 5:20pm

Hi All,

I have a problem that my Blink system will not always Arm and Disarm via automations, in fact it rarely works

The integration seems to be set up correctly (long 30 min scan interval) as sensors etc show and when using the alarm card in lovelace to arm/disarm it always seems to work. However when running via automations it is very inconsistent, sometimes one of the 2 Sync modules will arm/disarm but never both and often neither respond. The result is the same if using the webhook to trigger it or manually executing the automation in the interface.

Error:

Connection error. Endpoint https://rest-prde.immedia-semi.com/api/v1/accounts/xxxx/networks/yyyy/state/disarm possibly down or throttled.
Connection error. Endpoint https://rest-prde.immedia-semi.com/api/v1/accounts/xxxx/networks/zzzz/state/disarm possibly down or throttled.

Sample automation

alias: Blink - Disarm from Webhook
  description: Disarm blink from Webhook
  trigger:
  - platform: webhook
    webhook_id: disarm_blink
  condition: []
  action:
  - service: alarm_control_panel.alarm_disarm
    data: {}
    entity_id: alarm_control_panel.blink_outside
  - delay: 00:01:05
  - service: alarm_control_panel.alarm_disarm
    data: {}
    entity_id: alarm_control_panel.blink_inside

I have tried various delays between the service calls but it makes no difference and in fact it is often the first call that fails and then the 2nd will works (although not always)

Running 0.116.4 on Docker on a Pi

Any help/suggestions would be appreciated as this has been driving me nuts since I set it up and I’m trying to move away from IFTTT

Thanks Pete

fronzbot · November 4, 2020, 3:47pm

how are you verifying that the sync modules are armed/disarmed? Do you look in the Blink app or are you relying on the HA frontend?

pete101 · November 4, 2020, 4:30pm

I’m checking from the Blink App as I know I have a long refresh set on the integration. At first I thought I was getting a better hit rate by calling the service for each module twice with a 5 sec gap between each call but still sometimes 1 of them doesn’t switch

Not sure if it is related but I have also noticed that HA is logging the below errors every 30 mins (which is the integration scan interval I have set). Interestingly it is only the one camera sensors that error (I have 3 cameras across the 2 modules) .

2020-11-04 16:04:49 WARNING (MainThread) [homeassistant.helpers.entity] Update of binary_sensor.blink_garden_battery is taking over 10 seconds
2020-11-04 16:04:49 WARNING (MainThread) [homeassistant.helpers.entity] Update of sensor.blink_garden_temperature is taking over 10 seconds
2020-11-04 16:04:49 ERROR (SyncWorker_4) [blinkpy.auth] Connection error. Endpoint https://rest-prde.immedia-semi.com/network/xxxx/update possibly down or throttled.
2020-11-04 16:04:49 ERROR (SyncWorker_25) [blinkpy.auth] Connection error. Endpoint https://rest-prde.immedia-semi.com/network/xxxx/update possibly down or throttled.
2020-11-04 16:04:51 WARNING (MainThread) [homeassistant.helpers.entity] Update of alarm_control_panel.blink_outside is taking over 10 seconds
2020-11-04 16:04:51 ERROR (SyncWorker_19) [blinkpy.auth] Connection error. Endpoint https://rest-prde.immedia-semi.com/network/xxxx/update possibly down or throttled.

Thanks

fronzbot · November 4, 2020, 4:35pm

Hmm… that’s odd. Everything else is updating properly though? For example, motion detection sensor triggering, temperature changing every 30mins, etc?

pete101 · November 4, 2020, 5:13pm

Yes the temps are updating but now you come to mention it, they just went 2 hours before updating (not sure if that was no update or just no temperature change on any of them)

Unfortunately I cant see before a few hours ago because as I am running HA on Pi, I have recorder set to use RAM (to save eating the SD card) but that does mean I don’t currently have much history as I restarted HA 2-3 hours ago. I will leaving it running for a bit to get a better history picture

I never had much luck with motion detection I assume because I have such a long scan interval, although I do intend to reduce this if I can get the arming/disarming working consistently .

pete101 · November 5, 2020, 8:41am

Ok so temp updates are not reliable for me either, the garden update that I posted in the screen shot yesterday was the last time it and the other outdoor camera updated in HA. However the indoor camera (on a different sync module) has updated a number of times over night. I have checked the Garden camera temp in the Blink App and it is shows correctly there.

pete101 · November 8, 2020, 3:01pm

Hi @fronzbot, I am still having problems getting this to work reliably. Is there any additional logs or anything that I can check or do you have any other trouble shooting suggestion please? I can’t seem to find a common issue/thread as all cameras and both modules seem to work some times and not other.

Error mostly is this one…

Logger: blinkpy.auth
Source: /usr/local/lib/python3.8/site-packages/blinkpy/auth.py:187
First occurred: 14:35:00 (5 occurrences)
Last logged: 14:49:23

Connection error. Endpoint https://rest-prde.immedia-semi.com/network/7666/update possibly down or throttled.
Connection error. Endpoint https://rest-prde.immedia-semi.com/api/v1/accounts/6145/networks/116281/state/arm possibly down or throttled.
Connection error. Endpoint https://rest-prde.immedia-semi.com/api/v1/accounts/6145/networks/7666/state/arm possibly down or throttled.

Many thanks

pete101 · November 8, 2020, 9:26pm

Have tried deleting and reinstalling the integration and will see if that make a difference.

fronzbot · November 9, 2020, 1:44am

Hmm is it possible the wifi connection to your outdoor camera/sync module is spotty resulting in dropped calls? If that is not a possibility, it’s a bit of a head scratcher as to why one of your cameras works and the other doesn’t so I’d have to give it more thought

pete101 · November 9, 2020, 4:28pm

Thanks for getting back to me and continuing to think about it. I will bring the Garden camera inside closer to wifi and see it that helps. The signal shows OK in the Blink app but you never know.

Some more info on my setup for background:

Reinstalling the integration didn’t make any difference
I have 2 sync modules:

“Inside” - XT black one with firmware 2.13.26 with one XT camera “Living Room” . network_id: 116281
“Outside” - older white internal one with firmware 2.12.28 (not sure if being lower is an issue but will ask Blink that question) with 2 XT cameras “Road” and “Garden”. network_id: 7666

All 3 cameras are on firmware 2.151
Integration set to 30 min scan interval
Overnight the only camera to update its temperature in HA was Living Room, however the others have previously reported correctly, but not consistently
The biggest usability issue for me is that neither sync module will arm or disarm consistently using the services: alarm_control_panel.alarm_arm_away and alarm_control_panel.alarm_disarm. Some times it will work repeatedly fine and a bit later one or neither will work.
Have tried all manner of delays and orders in automation and just calling services from Developer Tools but results are never consistent.
At the moment every 30 mins it is only one module and one camera that is erroring as below. On the regular 30min polls it is only ever the Garden Camera that times out however both modules have errored at some point.

2020-11-09 08:54:14 WARNING (MainThread) [homeassistant.helpers.entity] Update of binary_sensor.blink_garden_battery is taking over 10 seconds
2020-11-09 08:54:14 WARNING (MainThread) [homeassistant.helpers.entity] Update of sensor.blink_garden_temperature is taking over 10 seconds
2020-11-09 08:54:14 ERROR (SyncWorker_33) [blinkpy.auth] Connection error. Endpoint https://rest-prde.immedia-semi.com/network/7666/update possibly down or throttled.
2020-11-09 08:54:15 ERROR (SyncWorker_7) [blinkpy.auth] Connection error. Endpoint https://rest-prde.immedia-semi.com/network/7666/update possibly down or throttled.
2020-11-09 08:54:16 WARNING (MainThread) [homeassistant.helpers.entity] Update of alarm_control_panel.blink_outside is taking over 10 seconds
2020-11-09 08:54:16 ERROR (SyncWorker_38) [blinkpy.auth] Connection error. Endpoint https://rest-prde.immedia-semi.com/network/7666/update possibly down or throttled.

Not sure if this helps or not but would really appreciate any thoughts you do have please.
Thank again.

pete101 · November 9, 2020, 9:57am

Sorry originally got the network IDs the wrong way round in my last post. Edited it now so they are correct

TazUk · November 9, 2020, 4:24pm

I’m also seeing some dodgy disarm behaviour on one of two sync modules together with a periodic “possibly down or throttled” message. It’s been playing up for some time but I’m looking to gradually get rid of my Blink setup so hadn’t bothered to raise it.

If it helps diagnose, I’m intending to pull my outside module and camera next weekend, which will leave the misbehaving indoor module and camera as the only module on my Blink account. It might provide some indication of whether this is a “two modules, one account” issue or related to specific modules / endpoints.
I’ll feedback here when I’ve removed the outdoor camera and cleaned my account and integration up.

pete101 · November 9, 2020, 4:32pm

Thanks, that would be useful. As suggested above, I also pulled my “Garden” camera inside and placed it along with the sync module, within 2m of a wifi AP and it made no difference so I don’t think my issue is down to a sketchy wifi connection.

wizardels · November 11, 2020, 11:26am

Also ran into the same issue this morning. Automation triggers and error occurs, however when I trigger it manually it’s okay and pretty consistent. @fronzbot any ideas?

fronzbot · November 11, 2020, 12:39pm

I don’t think it’s a problem with TWO modules just because it’s the first one that fails not the second. I have an old white sync module I can hook up to debug (might take me some time though). Maybe they changed an endpoint for the older modules

@wizardels - that is weird- calling it manually does nothing different than calling it from the automation. The only problem you might run into with automations is not allowing enough time to pass between sequential API calls (ie. No delay or a delay that’s too small). Can you confirm that you have appropriate delays between two or more service calls?

pete101 · November 12, 2020, 7:57am

@fronzbot would turning debug logging on provide any possible further insight do you think, and if so which components shoudl I add to the logger config please?

wizardels · November 16, 2020, 10:26am

Hey! I think mine was related to one of the cameras having wifi issues. Have fixed that now and all seems to be working. Thanks!!

pete101 · November 17, 2020, 4:43pm

Ok so on the back of @wizardels point above, I moved both my external cameras inside close to the sync module and WiFi and it did seem to be better. What is strange is that connectively on one module will stop the other one from arming/disarming. I put the cameras back out and moved one of my APs to be near one of the cameras and both now show full signal on wifi, One still has a lower signal on the module strength but as one camera is out the front and one is out the back of the house there is a limit to how close I can get the module to both cameras. I also re did my automations to trigger both modules twice (so 4 call in total) with 10 seconds between each one and while the odd call is failing, so far both have not failed for the same module. I still get one of the module (the older one) erroring in the logs each time HA pols and I don’t think the temps etc always update but I can live with that as is the arm/disarm in automations that I need to work . Will see how it goes. Worth noting that I still don’t get any issue arming and disarming from the blink app.