Question - is it possible that the returned Future from a write is completed successfully before the actual write happens? about folsom

[spkrka]

Thanks for the report.

The asynchronous set operation is supposed to return a future that completes once the memcache service responds.

To troubleshoot this further it would help to have more information:

1. Can you share a branch or repo with a minimal test case to reproduce this?
2. What kind of memcache service are you using to talk to? Is it possible that the problem is there?

Some preliminary guessing follows:
Perhaps the folsom code or the memcache service is rounding up/down the time in full seconds when setting the expiration, so it would pass or fail depending on when the test runs. I would personally change the sleep to 3000 milliseconds instead to make it more reliable.

[spkrka]

Closing this since there hasn't been any updates. Please reopen if there's any new data

[flozano]

I am afraid I missed this notification and never answered 🤦

As a reference, current versions of folsom don't reproduce the issue. The memcached service was just the then-current version memcached (hosted by AWS elasticache)

Thanks anyway and sorry again for missing the answer.