You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Aug 29, 2024. It is now read-only.
During heavy traffic and when a lot of permits are generated, we hit the RPC endpoints quite often and sometimes reach the call limit and get denied. Example of faulty run here.
Problem is that in the end no permit in generated thus no reward gets dispatched to the users, like here. We should instead retry, or switch endpoints until the generation succeeds. As the GitHub Actions can run for hours, it would be okay to slowly hit the endpoint until it works again. Ideally, switching to another available endpoint is the preferred solution.
The text was updated successfully, but these errors were encountered:
@rndquu Correct. This issue is related to RPC endpoints not the bot itself sorry if it wasn't clear. Basically, the more we hit RPC endpoints the more chances we have to be denied. Maybe our RPC handler solves that issue.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
I am creating this ticket here because it is the module that failed, but this would very likely also happen within https://github.com/ubiquibot/conversation-rewards which itself calls https://github.com/ubiquibot/permit-generation where this take should probably be carried out.
During heavy traffic and when a lot of permits are generated, we hit the RPC endpoints quite often and sometimes reach the call limit and get denied. Example of faulty run here.
Problem is that in the end no permit in generated thus no reward gets dispatched to the users, like here. We should instead retry, or switch endpoints until the generation succeeds. As the GitHub Actions can run for hours, it would be okay to slowly hit the endpoint until it works again. Ideally, switching to another available endpoint is the preferred solution.
The text was updated successfully, but these errors were encountered: