Core: Point: API rate limited

Created on 18 Dec 2018  Â·  12Comments  Â·  Source: home-assistant/core

Home Assistant release with the issue:

Home Assistant: 0.84.2

Last working Home Assistant release (if known):
Not sure, but at least the previous.

Operating environment (Hass.io/Docker/Windows/etc.):

Latest Docker build of homeassistant/home-assistant on a debian host.

Component/platform:

https://www.home-assistant.io/components/point/

Description of problem:
Stopped getting sensors update, seems to be hitting some type of rate limit in their api. I was hoping there was something like update_interval: '00:15:00', but that wasn't supported in the yaml.

Problem-relevant configuration.yaml entries and (fill out even if it seems unimportant):

point:
  client_id: <foo>
  client_secret: <bar>

Traceback (if applicable):

2018-12-18 20:44:41 WARNING (MainThread) [pypoint] Failed request: HTTPSConnectionPool(host='api.minut.com', port=443): Max retries exceeded with url: /v1/devices (Caused by ConnectTimeoutError(<urllib3.connection.VerifiedHTTPSConnection object at 0x7f7e209c2668>, 'Connection to api.minut.com timed out. (connect timeout=10)'))

Additional information:

point

Most helpful comment

This isn't rate limiting. We're seeing long response times on some queries for sensor data though, and are working on a fix for that. We won't deploy that before Christmas for obvious reasons, but rather in the first half of January.
Regarding a "new data" hook, that's on our roadmap and will also be deployed in the beginning of next year. I'll get back to you when we have that in place :)

All 12 comments

Try scan_interval: XYZ (where XYZ in seconds)

Good guess, but that wasn't supported either:

invalid config for [point]: [scan_interval] is an invalid option for [point]. Check: point->point->scan_interval. (See /config/configuration.yaml, line 143). Please check the docs at https://home-assistant.io/components/point/

I put it under the point: yaml, and that where I got the error.

CC @fredrike

Strange, Point polls every minute according to SCAN_INTERVAL = timedelta(minutes=1) and I've been running that since june without throttling logs.

@joltcan Are you running with "your own" keys or the example keys?

Hey!

I didn't know I could use example keys? I'm running with api dev keys that I requested.

On 19 Dec 2018, at 10:44, Fredrik Erlandsson notifications@github.com wrote:

Strange, Point polls every minute according to SCAN_INTERVAL = timedelta(minutes=1) and I've been running that since june without throttling logs.

@joltcan https://github.com/joltcan Are you running with "your own" keys or the example keys?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub https://github.com/home-assistant/home-assistant/issues/19437#issuecomment-448532728, or mute the thread https://github.com/notifications/unsubscribe-auth/AAgyRQTd-kT8CvzlnbZLRAPj2XGRo5qJks5u6gpwgaJpZM4ZY-8Y.

How often do you get the error?

Every other hour or so. Should I remove and re-add the integration and see if it's just a fluke?

@archie @themadinventor, is the Point API rate limited and if so, what would be a reasonable polling time?
Would it be possible to extend the webhooks with a "new data" hook so we can limit the polling even harder?

This isn't rate limiting. We're seeing long response times on some queries for sensor data though, and are working on a fix for that. We won't deploy that before Christmas for obvious reasons, but rather in the first half of January.
Regarding a "new data" hook, that's on our roadmap and will also be deployed in the beginning of next year. I'll get back to you when we have that in place :)

@themadinventor it would be good to support "last value too", I got a reply from @archie that adding ?limit=1 would return last value but apparently that is not the case.

@joltcan, Is this still active or can we hope that the Minut guys have applied the necessary patches?

Looks good to me now!

Was this page helpful?
0 / 5 - 0 ratings