Open matiasAS opened 4 months ago
I have the same problem and didn't found the solution yet.
@tadcom0 For me, at times it works and at other times it is like this continuously.
@ab77 help me please with this issue, long time waiting for a solution
Same issue. Deployment with Docker Desktop and WSL2
@tadcom0 hello, Have you found the problem? It takes a long time for developers to respond to issues, that's the bad thing.
@tadcom0 hello, Have you found the problem? It takes a long time for developers to respond to issues, that's the bad thing.
Hi,
Does this block any functionality on your side? Because these are only supposed to be warnings.
One way to to silence these warning could be to use the CONTRACT_ALLOWLIST
env var for the API service on your docker-compose, and list there only the contracts (device types, cpu architectures etc) that you care to import on your instance.
Let me point you to its definition & description:
https://github.com/balena-io/open-balena-api/blob/master/src/lib/config.ts#L96-L110
Hi, I would like to let you know open-balena-api v31.1.0 should no longer log these warnings. Can you give it a try and let us know?
Kind regards, Thodoris
@thgreasi Is that version of the api compatible with openbalena v3? I also have v4, I will test how it behaves
@thgreasi
The errors are not shown in version 31.1.0, but if I make a request to the url:
https://api.
Best regards
Hi @matiasAS,
There were no behavioral changes in 31.1.0, other than silencing the warnings, since in unauthenticated open-balena they are expected, thus are just noise.
What version of open-balena-api were you running before? The only think that I can imagine being related to what you describe is when open-balena-api v0.206.4 changed the way that its /device-types/v1
cache was working to a lazy/on-demand population method.
You mention that it coincides with when these errors warnings where seen before, which sounds like the /device-types/v1
cache at that point had expired, either b/c it became stale or b/c the Redis cached item was evicted (which doesn't happen on balenaCloud, either because we continuously get such requests & we don't have space based evictions from Redis). On the other hand populating the cache shouldn't take that much time (should be ready in less that 5s when I was testing it a few years ago), but that of course depends on the networking bandwidth that your instance.
One way to make things faster is trying to use the CONTRACT_ALLOWLIST
env var to have your instance only fetch the device types that you actually use.
Hi @thgreasi , Are you from the openbalena development team or a user? Do you know if the is_online attribute can be made reactive and in real time? And another question, for version 3.x and 4.x, what should I do to change the domain?
Description I'm experiencing repeated Access Denied errors in the API service of OpenBalena. Below is a screenshot of the errors:
The errors look like this:
Questions What is causing these Access Denied errors? How can I resolve these errors to ensure the API service functions correctly? Any guidance or suggestions to resolve this issue would be greatly appreciated.
Thank you!
Best regards, Matias Alvarez Sabate