Understanding Redis cluster setting cluster-require-full-coverage

[Wei Si]

Hi,

I'm new to Redis cluster and trying to understand the cluster-require-full-coverage flag:

If this is set to yes, as it is by default, the cluster stops accepting writes if some percentage of the key space is not covered by any node. If the option is set to no, the cluster will still serve queries even if only requests about a subset of keys can be processed. (from tutorial)

It looks to me that it's more desirable to set cluster-require-full-coverage to no so that clients can continue to write to available shards in case of node failures that cause slots not being covered. Are there any scenarios that benefit from cluster-require-full-coverage=yes?

Thanks!

Wei Si

[Tuco]

setting it to 'yes' gives you an indication that something is wrong with your cluster, and needs to be fixed immediately.  If you set it to 'no', it's possible that only some clients are affected, no alert is generated and the issue will go undetected for longer time. In system design, such a philosophy is called 'fail fast philosophy', see here.

[Wei Si]

That makes sense, thanks a lot for the response!