FLUSHALL and FLUSHDB documentation should clarify async operation.

[milescrawford]

A question has come up several times - what are the concurrency implications of the ASYNC modifier for FLUSHALL/FLUSHDB?

Specifically, are keys inserted after the FLUSH* ASYNC command candidates for removal? Or does the async thread only remove keys that existed at the time the flush command was executed?

A quick sentence or two on this topic on these pages would suffice:
https://redis.io/commands/flushall
https://redis.io/commands/flushdb

[itamarhaber]

Hello @milescrawford

The latter, only keys that existed before the command's execution will be flushed. Please feel free to open a new Pull Request with the suggested clarification that fixes this issue.

[milescrawford]

Thanks, I will later today.

…

On Sat, Dec 15, 2018, 6:23 AM Itamar Haber ***@***.*** wrote: Hello @milescrawford <https://github.com/milescrawford> The latter, only keys that existed before the command's execution will be flushed. Please feel free to open a new Pull Request with the suggested clarification that fixes this issue. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#163 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AKiya88ryQAHwr_k1ttmGwEuecsETwJSks5u5QX0gaJpZM4ZURIY> .

[schnapster]

As a follow-up question, I'm wondering, what the behaviour is for keys that are updated after executing an async flush command. Is there a chance the updated key/value will be flushed?