This message was deleted.
# k3d
a
This message was deleted.
r
My understanding of the servers & agents flags is that it corresponds to the server/agent options you'd have in a normal k3s install. In a normal k3s install a server has the control plane (and etcd if you are using embedded etcd) but is not tainted to prevent other workloads from running on it by default. Agent nodes will be joined to the cluster and not have any control plane or etcd on them nor will they be tainted to prevent workloads running.
f
Yes i’m running with embedded etcd, so if I created 3 servers there should be 3 etcd run on each server respectively?
That seems to be the case. I have 3 server nodes and each runs an etcd:
Copy code
# kubectl get node
NAME                   STATUS   ROLES                       AGE     VERSION
k3d-scaling-server-0   Ready    control-plane,etcd,master   4m7s    v1.25.7+k3s1
k3d-scaling-server-1   Ready    control-plane,etcd,master   3m49s   v1.25.7+k3s1
k3d-scaling-server-2   Ready    control-plane,etcd,master   3m32s   v1.25.7+k3s1
Copy code
# etcdctl endpoint status -w table
+-------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
|        ENDPOINT         |        ID        | VERSION | DB SIZE | IS LEADER | IS LEARNER | RAFT TERM | RAFT INDEX | RAFT APPLIED INDEX | ERRORS |
+-------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
| <https://172.19.0.2:2379> | d7380397c3ec4b90 |   3.5.3 |  2.8 MB |      true |      false |         2 |       1685 |               1685 |        |
| <https://172.19.0.3:2379> | 94ceaa2d6f7a3716 |   3.5.3 |  2.8 MB |     false |      false |         2 |       1685 |               1685 |        |
| <https://172.19.0.4:2379> | d78ddaef0ac61725 |   3.5.3 |  2.8 MB |     false |      false |         2 |       1685 |               1685 |        |
+-------------------------+------------------+---------+---------+-----------+------------+-----------+------------+--------------------+--------+
r
If you're running embedded etcd then you should have an etcd pod on each of the three servers for a single distributed etcd datastore.
f
Thanks! And I have configured etcd storage to 8GB via
quota-backend-bytes
, so I should have 8GB storage in total, is that correct?
Copy code
/ # ps -ef | grep etcd
    1 0        /sbin/docker-init -- /bin/k3s server --etcd-arg=quota-backend-bytes=8589934592 --disable=metrics-server --cluster-init --tls-san 0.0.0.0 --tls-san k3d-scaling-serverlb
r
No clue on that
f
ok.. thanks for your explaination!