You’ve probably seen it, lots of outages lots of delay and well not a lot from me.
life has been incredibly busy as I thought I was out of the IT rat race, but things happened, and I’ve ended up staying in.
so yes, once more again “docker” and k8 make everything easy to deploy but maintain and work, well that is another story. It turns out that Rancher OS k8’s will shut down once the disk hits 85% capacity.
ok that sounds like a lot but on a 1tb volume that is 125GB?!!! No warning no, no nothing just ‘disk pressure’ on the console and that is it.
root@steady35:/opt/local-path-provisioner# df -h .
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 873G 712G 115G 87% /opthonestly WTF!
I like the ability to publish secrets and import and export settings between containers, but the incredible lack of logging is unreal. The k8 manifest for MySQL kept on failing on my server. I have no idea why.
zero logging info
I deploy the docker version directly, and if you can read this it’s working.
clearly I’m the old man out of touch.
but for a single node, UML is the way to go.
plus, you can use your entire disk.
so yeah, trying something new although I don’t think it’ll last all that long.