Hi,
I am running Home Assistant core 2021.12.10 in a Docker on a Synology NAS. I have the Nest integration up and running with my Nest thermostat v3, Nest Cam IQ indoor and Nest Doorbell v1. Thermostat works fine and for both camera’s i can view the streams and event snapshots (of most events) in the Media Browser.
However, in the Home Assistant log, I have the repeating warning below. I have enabled the debug logging as mentioned on the Nest integration page. In the debug log I can also see that for some camera events there are no messages related to getting the camera snapshot (request[post], request[post json] and request[get]).
Hope you can help.
Logger: google_nest_sdm.event_media
Source: /usr/local/lib/python3.9/site-packages/google_nest_sdm/event_media.py:464
First occurred: 19:15:32 (8 occurrences)
Last logged: 21:49:05
Failure when pre-fetching event ‘CiUA2vuxr6BaVk3ADmwKSayP_YWCOMXuA380R4m2dbcKiJy6gF3LEr4BAGJvXccwgVBNbstC35FlNR5iHEZJVpcBe3dxSe35jtLJrV4MTz2j1urXfBsHNqIF8J3ag3OV3hH1LXkEX18HkgAnCl4SVmccsCwMUGbMMG64oNavwE5d2TFtNSvi_ZIUrzaO8cpssDCGo5OgOAeonsvj7z4C791wpvvR7kGmw1R1LZpGH5ysyr8mgT5H40eRDDXfpPuzvZcskfAivaamSwqGivmSKe_Mh9CRmPLxOxF0GXZZ3UGtCSF1GA5j5g’: Internal Server Error
Failure when pre-fetching event ‘CiUA2vuxr48OIa9QqAj7vp0XqY8DFe92D_A-9vMt3UCayqocF-cGEsABAGJvXcd1-VyUuuURxUSGbhFcGaFeDLmvHeCIx7aQTZLXvDiRSh1HZkHB03ZBrEbhYYbjVa2tCIZgcJsrFl2lQrspFsVB-06NTD9J5VWMG54MJXgr3wEqGk-3r-hcLJWW0kC6iqIsNnA3jxZ-zGch8DQEOG1nT2nxTx1tFl4kjkRuoN1KheqJiUriD6Qfby2s_v4Xiz5TLIhn4E2E0eTS8aWHs04bSP0f_Poqyw3tl76r9h0M8uJ2KaZ8KcEoWuzd’: Error connecting to API: Cannot connect to host nexusapi-eu1.dropcam.com:443 ssl:default [Try again]
Failure when pre-fetching event ‘CiUA2vuxr7uOcfmsV5Fdms9o9AVgG6owulZdOhwhV1NBW4eRuXqjEr8BAGJvXccG-O2BQNhnGHv21EBxSm15X2qJ7XcUqQ8zOyf9jW680UWRZ9uyFWKPqKH4sfCE0Ok0lnD4BnGwC5jLM7kOXMPUmshyjUrpk2b_tMnfYTJ2GMfE64MtXbel-smnnFPTHZbnprEX47IvPcQHecdeV3_yGm75TmZNyPguvA3NaZANhlRptFiMws9tr1vxKXb12S7AZZ7dyJjYqmwY2LxX2NMP4t6gsKnTqr6_WhTIS7NlZE-L0BV_yUeN-Pg’: Error connecting to API: Server disconnected
Seems like a network issue. Do you have ipv6 enabled by chance? I often see folks report that turning off ipv6 resolves these kinds of network issues sometimes (though not sure why it wouldn’t work). Presumably your network should be working given you received the message notifications, though.
Thank you for the response and sorry for the late reply. I am not using ipv6 as it was giving me other problems as well. I tried playing around with DNS server settings, but it looks like it is not related to that. I still have around 24 warning messages with the Internal Server Error every 24 hours.
Is there anything else i can try?
Now the works with Nest is removed from HA I need to migrate to the Device access Console. But is it still possible to set the home/away status of the nest thermostat? I read somewhere that this is not the case…
if I enable my living room thermostat which controls the whole house forced air furnace it would switch to eco mode while I &/or others were home but not in the living room.
I cannot use my phone app since I am often not at home while others are.
Regardless of my reasons, I don’t think the Google API exposes that feature but @allenporter is the Guru on this…
I use Hoobs (Homebridge) as a bridge to the missing features of the HA integration (such as the Protects and Home/Away status). This then presents to HA via HomeKit. Which works pretty reliably. Not ideal, but works for my use cases.
Because HA uses the new Google SDM api (which in theory is Google’s go forward api), whilst Homebridge uses an old Nest api (I think Works with Nest, but I’m not 100% sure).
It’s using unofficial apis and authentication methods that are entirely unsupported and could theoretically change or get blocked (no idea how likely this is, but so far so good. Home Assistant uses officially blessed and supported SDM APIs following security and privacy best practices. Edit: This is not intended to pass judgment.
@allenporter I redid the last step of configuring the Nest integration within Home Assistant and one thing i noticed is that i do not get the ui.nabu.casa site name at the permissions part, but homeassistant which is the project name i used. Is this ok?
I see, yes, that is expected. I think in the old setup that was there because the redirect url was someones own nabu casa URL. Now we use the method where you copy and paste an access token instead.
Ever since 2022-02, I’ve been seeing a lot of errors related to my cameras like these exhibits. Sometimes the devices show “Unavailable” in the Devices page. And my automations which call the “camera.snapshot” service periodically are failing most of the day over the last couple days.
2022-02-13 15:31:30 ERROR (stream_worker) [homeassistant.components.stream.stream.camera.front_door] Error from stream worker: Error opening stream (HTTP_NOT_FOUND, Server returned 404 Not Found) rtsps://stream-us1-alfa.dropcam.com:443/sdm_live_stream/<long text string here - redacted>?auth=****
2022-02-13 15:38:14 WARNING (MainThread) [haffmpeg.tools] Timeout reading image.
2022-02-13 15:50:29 WARNING (MainThread) [haffmpeg.tools] Timeout reading image.
2022-02-13 15:50:30 ERROR (stream_worker) [homeassistant.components.stream.stream.camera.back_door] Error from stream worker: Error opening stream (INVALIDDATA, Invalid data found when processing input) rtsps://stream-us1-charlie.dropcam.com:443/sdm_live_stream/<long text string here - redacted>?auth=****
Hi, there are a couple things. The error handling was “improved” to actually catch the errors. It use to fail silently or the image would freeze, now it correctly marks the camera as unavailable. There was bug in the error handling, however, fixed in 2022.02.06. The general issue of 404s from streams is being tracked in https://github.com/home-assistant/core/issues/65697 and i’ve reached out to the nest SDM API team about it, though we don’t have resolution yet.