rancher-users #harvester

harvester

sticky-summer-13450

04/17/2022, 12:11 PM

I have finally added a third node to my Harvester v1.0.0 cluster. (I started with node 003, then added 002, and finally added 001 today). I thought that after adding the third node the cluster would turn into an HA cluster where all three nodes are the control plane with etcd, but that has not happened. Have I done something wrong?

$ kubectl get nodes -o wide --context harvester003
NAME           STATUS   ROLES                       AGE   VERSION          INTERNAL-IP   EXTERNAL-IP   OS-IMAGE           KERNEL-VERSION         CONTAINER-RUNTIME
harvester001   Ready    <none>                      29m   v1.21.7+rke2r1   10.64.0.16    <none>        Harvester v1.0.0   5.3.18-59.37-default   <containerd://1.4.12-k3s1>
harvester002   Ready    <none>                      63d   v1.21.7+rke2r1   10.64.0.17    <none>        Harvester v1.0.0   5.3.18-59.37-default   <containerd://1.4.12-k3s1>
harvester003   Ready    control-plane,etcd,master   84d   v1.21.7+rke2r1   10.64.0.18    <none>        Harvester v1.0.0   5.3.18-59.37-default   <containerd://1.4.12-k3s1>

adorable-ocean-16595

04/29/2022, 12:03 PM

Hey Can anyone help me troubleshoot this pod?

polite-megabyte-63476

04/29/2022, 2:08 PM

Is it possible to change the management nodes to different hosts?

best-lawyer-73573

04/29/2022, 10:40 PM

Quick question: how do I restart the kubelet in harvester? When I describe the node, I see that kubelet has stopped posting status. Also, where are the kubelet logs?

bumpy-portugal-40754

05/01/2022, 12:41 AM

I noticed that it's possible to rollout various ssh-keys to the Harvester nodes at installation time. To follow the normal lifecycle it is necessary to be able to add or remove keys later at runtime. How can this be achieved?

elegant-ram-99031

05/03/2022, 4:50 PM

Can anyone clarify the best practice when you need both k8s and vms? Rancher helm chart on harvester's k8s? k3s with rancher in vms? Separate rancher for management on bare metal or vms? Just use harvester k8s?

polite-megabyte-63476

05/03/2022, 5:30 PM

Also what causes this

adventurous-engine-93026

05/04/2022, 1:45 AM

Has anyone had reliable success creating a VM from an ISO? I've been trying over and over to get something like Ubuntu desktop (both 22.04 and 20.04). I either get "starting" or on the off chance that it does start and I can go through the setup phase, it can't find the disk after the install and reboot. I've tried using the template and creating the VM from scratch.

big-judge-33880

05/05/2022, 9:22 PM

Is it possible to create dual-stack rke2/k3s clusters using the harvester node driver on rancher?

little-dress-13576

05/06/2022, 7:23 PM

is it true that minimal number of nodes in a HA cluster is 3 nodes? could we deploy new VMs on these 3 nodes, or these 3 nodes have to be k8s master nodes, and they could not be used for anything else (meaning running new user VMs)?

salmon-gigabyte-65836

05/07/2022, 5:36 PM

Trying to upload RHEL 8.5 to Harvester but fails. Is it to big of a file? iso is 10gb

big-judge-33880

05/07/2022, 11:36 PM

The harvester vip pool load balancer; is it possible to get real client IPs delivered down to workloads while using this?

clever-wire-57058

05/09/2022, 12:53 AM

(I also have 2x 4 TB drives on each node, but those are supposed to be data drives...)

sticky-activity-13255

05/10/2022, 3:22 AM

https://docs.harvesterhci.io/v1.0/dev/dev-mode/ I followed the documentation but failed. How to do it? Does the document need to be updated?

sticky-activity-13255

05/10/2022, 3:24 AM

How to install harvester in k8s?

cuddly-lion-65543

05/10/2022, 9:20 AM

In page https://docs.harvesterhci.io/v1.0/upgrade/automatic/ , there are some warnings about shutting down the VMs while upgrading the Harvester. Does it mean that we really need to shut down all VMs on every host and keep them all down while upgrading Harvester (causing potential downtime in the environment)? Is it still the case for upgrading from 1.0.1 to a future version?

elegant-ram-99031

05/10/2022, 8:41 PM

Hey all. Does an HA harvester run workloads on management nodes? Or are those sacrificed compute if too large?

wonderful-helicopter-16401

05/11/2022, 3:42 AM

Is anyone else experiencing "Failed to wait for processing start in 60 seconds" on image uploads? I have no issues creating images when using a URL, but they fail with uploads — regardless of which image (ISO) I choose. I'm not experiencing this on other hypervisors with the same images.

adventurous-oyster-5549

05/11/2022, 4:54 PM

My lab environment does not have a DHCP server. I am trying to deploy it using the following link mentioned in the document. https://github.com/harvester/harvester/issues/947 but unable to do so. Anybody tried it. It would be helpful if I get to know what configuration needs to be changed in the above YAML to get it working. Thanks in advance

wonderful-pizza-30919

05/13/2022, 2:02 AM

Hi, What is the latest version of Harvest at present? And when is the next release is scheduled to be released? I am still waiting for VXLAN support. Thank you.

bumpy-portugal-40754

05/13/2022, 1:56 PM

I'm playing a bit with my harvester virtual box image. No customizations yet. Plain image with 2 NICs. The update from 1.0.0 to 1.0.1 was never working. The update from 1.0.1 to 1.0.1-rc1 was working fine. Shorty after that I tried to update it to 1.0.1-rc2 and ran into a certificate error: Internal error occurred: failed calling webhook "virtualmachines-mutator.kubevirt.io": Post "https://virt-api.harvester-system.svc:443/virtualmachines-mutate?timeout=10s": x509: certificate has expired or is not yet valid: current time 2022-05-12T21:27:17Z is after 2022-05-12T20:20:27Z Anybody seen this? Such a thing should never happen.... I guess.

wonderful-helicopter-16401

05/13/2022, 9:31 PM

Is there a way to make public IPs available to VMs that are assigned to the management NIC? Switches available are not managed by us and are not VLAN aware.

wonderful-helicopter-16401

05/15/2022, 3:54 AM

I'm confused by what I am seeing in VM networking. When I create a new VM, DHCP is assigning the same IP to every new VM. First point of confusion is that there isn't a DHCP server available outside the cluster and I haven't installed one inside yet. So where is this DHCP address coming from? It's also a specific /24 cidr that I often use in lab networks, so I am curious if I set that subnet somewhere during install. Note that I am using one NIC with a public IP. Is this expected behavior?

narrow-twilight-18889

05/16/2022, 3:28 PM

Need some help. I'm trying to boot the RancherOS.ISO and the VeeamRecovery.ISO but I keep getting the UEFI Interactive Shell Environment. I'm going to startup a RancherVM and attempt to migrate a VM by restoring a backup using Veeam.

sparse-art-48043

05/17/2022, 5:11 AM

Is it going to break my system if I run Rancher in a VM inside harvester and then use that Rancher instance to manage harvester?

polite-megabyte-63476

05/17/2022, 2:26 PM

Any way to change the VIP address?

polite-megabyte-63476

05/17/2022, 6:50 PM

Wonder how hard it would be to add https://www.dell.com/support/home/en-us/drivers/driversdetails?driverid=rx6p0&oscode=sle15&productcode=poweredge-r740 into Harvester

wonderful-helicopter-16401

05/18/2022, 11:42 PM

How do you assign VIP to a different NIC than mgmt in an ISO install?

polite-megabyte-63476

05/19/2022, 7:21 PM

So what's the status on the harvester CSI driver for k3s?

numerous-barista-78936

05/19/2022, 11:35 PM

Title

numerous-barista-78936

05/19/2022, 11:35 PM

wonderful-helicopter-16401

05/20/2022, 2:00 AM

You may want to check out this Question and the walkthrough provided. https://github.com/harvester/harvester/issues/2280#issuecomment-1128672751

I opened that issue because I couldn't find good docs for troubleshooting. Because I didn't have workloads to consider, I just started over.

numerous-barista-78936

05/20/2022, 2:12 AM

Appreciated

bumpy-portugal-40754

05/22/2022, 10:17 PM

Looks like the GUI is really bad here. Users don't see whether or not the update is hanging. I'm watching the cluster events on one master node... and when the node is rebooted the update seems ok. The GUI does not update anymore. A reload is also not very helpful, because the update might hang somewhere inbetween. There are some troubleshooting links under https://docs.harvesterhci.io/v1.0/upgrade/v1-0-0-to-v1-0-1/ . They didn't work for me when I got an hanging 1.0.1 update, so I doubt they work for 1.0.2... but you can try. Suse has still a lot of work to do for their update experience.

numerous-barista-78936

05/23/2022, 12:25 AM

@bumpy-portugal-40754 I actually had some help from a community member, the solution was very easy. Check out that GitHub issue

Symptoms of early software development, but there's clearly dedication to it which is why I continue using Harvester

Here is the command that fixed it for me:

helm rollback rancher-monitoring -n cattle-monitoring-system

bumpy-portugal-40754

05/23/2022, 7:07 PM

Funny that a monitoring component can block a complete Harvester version update. Did you do this rollback from the GUI?

numerous-barista-78936

05/23/2022, 7:09 PM

No, from the terminal. Had to grab Helm for it

👍 1

View count: 6