operate-first / ionos-openshift Goto Github PK

• create account
• change PW
• use bitwarden to handle PW

OpenShift creates a network load balancer for API and Ingress, specified as API and ingress IP

Setting these IPs as forwarding targets in the IONOS network load balancer does not work, therefore I have to create forwarding for each private IP of each node.

The internal VIP of the OCP loadbalancer is up and responds, so I guess this is a limitation of the IONOS LB.

There's a red warning in DCD

which github handles from IONOS are watching this repo?
add them to the operate-first community repo and grant access to this repo

getting reverse DNS setup in operate-first/apps#2214

according to the docs we also need to involve IONOS https://docs.ionos.com/cloud/compute-engine/networks/networks-faq#how-do-i-create-reverse-dns-entries

Find a common date to demonstrate the Data Center Designer
MUST: @tmicheli @schwesig StefanPusch
COULD: @tux1980

Hi,

we are trying to get the RHEL8 / RH CoreOS Standard qcow2 images to run in IONOS DCD but somehow we did not get cloud-init to run... instead the image always came up with uid/pw prompt but did not take the cloud-init config provided in the IONOS DCD.

A workaround was to baking in the uid/pw into the image using virt-customize, but the target config (providing ssh keys via cloud-init) somehow does not work with IONOS DCD... any ideas?

Thanks a lot in advance for your help.

Demosession for IONOS in April/May 2022.
How can I simplify using our RHEL KVM Images without being bothered through Workflow Policies.

The RH assisted installer requires machines to boot from an iso image, which bootstraps the machine. Then the machine should reboot from the internal storage.

I did not find a way to make the internal storage to be the default boot device either with ionosctl or the terraform provider. This has to be done in dcd by hand now

https://github.com/operate-first/ionos-openshift/tree/main/terraform-ocp#ionos-infrastructure

We're deploying with a network loadbalancer, i.e. we don't need ingress and api VIPs for the assisted installer.
We're currently not using them. The LB forwards to all nodes.

Then why do u deploy with vips? U can use umn (user managed network) and configure dns and lb (as u already do). Assisted installer supports it as well

https://github.com/openshift/assisted-service/tree/master/docs/user-guide

https://coreos.slack.com/archives/CUPJTHQ5P/p1661618892295609?thread_ts=1661519636.406069&cid=CUPJTHQ5P

now the assisted installer uses the default ssh key.
add the operate first key