Rebirth failed requests
No credit card required
Rebirth failed requests
No credit card required
Rebirth failed requests of past runs into a pristine state with no retries so you can rescrape them by resurrecting the run.
There isn't an easy way on Apify to retry fully failed (or just handled) requests. This actor allows you to set those requests to pristine unhandled state with 0 retries so you can resurrect the run and process them again.
How it works
This actor scans all requests in a queue of a run and recognizes failed requests by their retryCount and errorMessages properties. If your actor deliberately changes these 2 properties (outside of the default Crawler behavior), the rebirth will not work properly.
Requirements
- Runs must use request queue. (Request list support might be added in the future)
- The run should be able to be resumed with a proper state management (imagine actor migration)
How to run
Detailed input description is available on actor's page.
- You can provide either:
- run IDs to scan for requests to be rebirth
- actor or task ID with combination of dates to find all runs in that timespan (to scan for requests to be rebirth)
- After requests are rebirth, you will see unhandled requests in the run's queue and you can resurrect the runs to get them processed again
- You can check automatic resurrecting of runs with specified concurrency (to work with your max memory limit)
- You can override a build. Normally, the actor is resurrected with the same build it had but often you might want to run newest version like
latest
Lukáš KřivkaActor Metrics
3 monthly users
-
2 stars
>99% runs succeeded
Created in Sep 2021
Modified 8 months ago
