ETOOBUSY 🚀 minimal blogging for the impatient
Leader in etcd
TL;DR
Who’s the leader in etcd?
As you surely know, etcd is:
A distributed, reliable key-value store for the most critical data of a distributed system.
I admit: I never used it (directly). Anyway, it happens to be the memory of a Kubernetes cluster, so it’s good to know it exists.
Deploying etcd in high availability mode means that there will be more than one instance of it. As a matter of fact, it just makes sense to have an odd number of instances, and most cluster I’ve seen settle on three.
At any time, the instances agree on a leader. They elect it autonomously, so there’s nothing to worry about. Except that, sometimes, you might want to know who’s the leader at the moment.
Assuming your etcdctl
program is properly configured, this can be done
like this:
# get the identifier of the leader
leader_id="$(etcdctl endpoint status -w json | jq .[0].Status.leader)"
# now get the name and the URL of the leader
etcdctl member list -w json \
| jq -r '.members[]
| select(.ID == '"$leader_id"')
| .name + " (" + .clientURLs[0] + ")"'
Yes, this can be overkill… but good to know!