S330: As an administrator, I want to be able to know whether my cluster can tolerate a failure of a node or locality without running out of disk space.

cockroach-staging / hello-world

0 stars 0 forks source link

S330: As an administrator, I want to be able to know whether my cluster can tolerate a failure of a node or locality without running out of disk space. #588

Open exalate-issue-sync[bot] opened 4 years ago

exalate-issue-sync[bot] commented 4 years ago

As an administrator, I want to be able to know whether my cluster can tolerate a failure of a node or locality without running out of disk space.

exalate-issue-sync[bot] commented 4 years ago

Rachel Casali commented: • The addition of zone configurations complicates this: because data can be constrained to certain physical machines, the failure of a node might distribute data very unequally to the remaining nodes in the cluster. Anticipating which nodes need how much available storage is a challenge as pointed out by Andrew C in the below github issue.