Replies: 1 comment
-
I’m in the same position where I have a cluster board with SBCs that are (currently at least) quite expensive. I also have a spare RPi4 that I am no longer using, so I was thinking to repurpose it in an identical manner but wary of performance degradation. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi, I am currently running k3s with a single master node and SQLite DB. I would like to move to a higher availability configuration using etcd. The master node has a fast NVMe drive. I have a second, similarly fast node. I also have a spare RPi 4B with 4GB RAM but am aware that etcd needs fast storage (i.e. not the SD card).
My thinking is that since my state DB is small (currently ~50MB) I could create a 500MB or 1GB RAM disk on the RPi and mount it on
/var/lib/rancher/k3s/server/db/etcd
, making it fast enough to operate as an etcd master node. This avoids the cost of adding a third node with fast NVMe, which I don’t really need for any other reason at present (my two nodes have plenty of capacity).Is there any reason not to do this? Will etcd automatically repopulate the DB state after a restart on the RPi?
Beta Was this translation helpful? Give feedback.
All reactions