Comparison
DaemonSetvsStatefulSet
DaemonSet
the log collector appears on every node automatically, including the one that joined the cluster four minutes ago.
A workload type that runs exactly one copy on every node, or every node matching a selector. It is how node-level infrastructure — log shippers, metrics agents, network plugins, storage drivers — is deployed, and it scales with the cluster rather than with traffic. Its capacity is easy to forget: each one takes CPU and memory from every node, so a fleet of five agents can quietly consume a meaningful fraction of what you are paying for.
Full entry →StatefulSet
the replicas are named zero, one and two rather than randomly, and each one gets its own disk back after a restart.
A workload type giving each replica a stable identity, a stable network name and its own persistent volume, and rolling them in order. It exists for software that cares which member it is — databases, brokers, consensus systems — where interchangeable replicas would be wrong. It is markedly harder to operate than a stateless deployment, which is the usual argument for running data stores as a managed service instead.
Full entry →