https://rancher.com/ logo
Join Slack
Powered by
Hey guys - is there an easy way to remove the ranc...
# harvester
m
Hey guys - is there an easy way to remove the rancher-vcluster add-on? I need to see if I can get it to the latest version of Rancher as opposed to 9.3, and deleting the PVC and setting the version in the config isn't forcing it.
b
In Harvester the rancher version is tied/managed to the harvester version
m
that's the built in version, not the rancher-vcluster add-on though
b
Looks like per the docs you just disable it in the UI then delete the PVC in the 
rancher-vcluster
namespace?
m
I did that, heh
b
Did you try deleting the entire namespace out of Harvester?
That's probably what I would try next
m
ok, I'll give that a shot
Thanks for the help!
b
worked?
m
trying it now
it is just sitting on the terminating status for the namespace
b
Anything in the events?
There some stale VM or something?
m
No events, and no VMs at all since this runs the vcluster natively in harvester
b
There's likely some object that it's failing to delete so the finializer isn't finishing.
You can force it by removing the finalizer section, but it's not always recommended.
Either by editing the yaml from the UI, or with 
kubectl edit
or 
kubectl patch
m
huh, oh interesting ... hopping on to the hosts and checking something, it's probably because of this
E0423 15:57:36.996247   86138 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp 127.0.0.1:8080: connect: connection refused
E0423 15:57:36.997933   86138 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp 127.0.0.1:8080: connect: connection refused
E0423 15:57:36.999964   86138 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp 127.0.0.1:8080: connect: connection refused
E0423 15:57:37.001367   86138 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp 127.0.0.1:8080: connect: connection refused
E0423 15:57:37.002319   86138 memcache.go:265] couldn't get current server API group list: Get "<http://localhost:8080/api?timeout=32s>": dial tcp 127.0.0.1:8080: connect: connection refused
The connection to the server localhost:8080 was refused - did you specify the right host or port?
which I'm not entirely sure how to resolve with harvester
b
8080? Isn't that the longhorn port?
m
basically my rke2.yaml file is ... missing
b
For Harvester in general?
m
only on that one host
b
There's a lot here that could be going wrong
If it were me, I'd contact Suse support and open a ticket at this point.
m
Yeah ... I had someone tell me on here how to do it but that thread is gone
b
Must have been a long time ago.
m
Maybe 5-6 months
b
Um. You could put the node in maint, copy the missing yaml from another node and reboot and see if things get better. Since the plugin you're working with is experimental it may just not be working. Is this cluster dev or prod?
m
dev
b
I'd probably just blow it away and start fresh.
m
the plugin was fine until I tried to change the version of rancher 😄
b
¯\_(ツ)_/¯
m
aight, just copied over and now I can use kubectl again
5 Views
Open in Slack
PreviousNext
Powered by