Openshift 4.3 Installation

[Pawan_GSS]

Hi Experts,

I have a small query to ask, I want to install openshift 4.3 on oracle virtual box. I have one master and 1 infra and 1 worker machine configured with Red Hat 7.7. Is there any way to install openshift 4.3 without using bootstrap machine. If possible how can I achieve that, Please suggest, its urgent.

Help is appreciated!!! 

Thanks

Pawan Kumar

[Walid]

I am interested too in such use case, most likely using OKD and pre-provisioned infrastructure UPI, there are some videos, however, I have not watched them yet on how to install on bare-metal UPI method on RHEV, gcp. here is one link I search for quickly   Guide to installing an okd 4.4 cluster on your home-lab   and check this medium article it could of great help Openshift4.4 okd bare metal install on vmware home lab 

if you are just interested in the Dev side of things, Code Ready Containers could be your best option

[Pawan_GSS]

Thanks, Walid!!!

 Openshift4.4 okd bare metal install on vmware home lab  this article makes me understand something which I am missing earlier.

Now, I am stuck at bootstrap machine creation as when I completely boot then it shows error like "selinux:mount invalid. some superblock, different security settings for (dev mqueue, type myqueue)".  Not able to log in on the bootstrap machine to check the status of completion. Please suggest. 

Thanks

Pawan

[Walid]

there is a bug report for the above issue, can be tracked from an access article https://access.redhat.com/solutions/3348951  the bug report https://bugzilla.redhat.com/show_bug.cgi?id=1425278

I have not started yet @Pawan_GSS , however, I found that github repo helping creating a helper node  that has all the dependencies if it is a dependency issue, I am not very clear yet on OCP4 vs. OKD installation differences.  my install most likely would be using UPI static IP addresses.

[Pawan_GSS]

Can you please share the github link for helper node?

[Walid]

sorryI thought I already did, here it is https://github.com/RedHatOfficial/ocp4-helpernode

[Pawan_GSS]

Thanks Walid for the link!!

I have gone through the provided link and perform steps accordingly. However, when I run the " ./openshift-install wait-for bootstrap-complete --log-level debug" then we are getting below error "DEBUG Still waiting for the Kubernetes API: Get https://api.lab.okd.local:6443/version?timeout=32s: dial tcp: lookup api.lab.okd.local on 192.168.1.128:53: read udp 192.168.1.128:47881->192.168.1.128:53: read: connection refused"

And I logged in to bootstrap machine and run this "journalctl -b -f -u bootkube.service" getting below 

Apr 07 09:06:54 localhost podman[11304]: 2020-04-07 09:06:54.943394002 +0000 UTC m=+0.368446077 container create 7204f3857e1617ce79188015e5d6113e4f79082fa46339f2213fea2368278145 (image=quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:742b729094abbc4ebeaf62323e9393b0d6bf06606d4fe349e8458f9191d9905a, name=etcdctl)
Apr 07 09:06:55 localhost podman[11304]: 2020-04-07 09:06:55.336206545 +0000 UTC m=+0.761258456 container init 7204f3857e1617ce79188015e5d6113e4f79082fa46339f2213fea2368278145 (image=quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:742b729094abbc4ebeaf62323e9393b0d6bf06606d4fe349e8458f9191d9905a, name=etcdctl)
Apr 07 09:06:55 localhost podman[11304]: 2020-04-07 09:06:55.35317206 +0000 UTC m=+0.778223954 container start 7204f3857e1617ce79188015e5d6113e4f79082fa46339f2213fea2368278145 (image=quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:742b729094abbc4ebeaf62323e9393b0d6bf06606d4fe349e8458f9191d9905a, name=etcdctl)
Apr 07 09:06:55 localhost podman[11304]: 2020-04-07 09:06:55.35328216 +0000 UTC m=+0.778334051 container attach 7204f3857e1617ce79188015e5d6113e4f79082fa46339f2213fea2368278145 (image=quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:742b729094abbc4ebeaf62323e9393b0d6bf06606d4fe349e8458f9191d9905a, name=etcdctl)
Apr 07 09:07:00 localhost bootkube.sh[1961]: {"level":"warn","ts":"2020-04-07T09:07:00.370Z","caller":"clientv3/retry_interceptor.go:61","msg":"retrying of unary invoker failed","target":"endpoint://client-6bede497-f9aa-42bf-b82a-102f6db97928/etcd-0.lab.okd.local:2379","attempt":0,"error":"rpc error: code = DeadlineExceeded desc = latest connection error: connection error: desc = \"transport: Error while dialing dial tcp: lookup etcd-0.lab.okd.local on 192.168.1.2:53: no such host\""}
Apr 07 09:07:00 localhost bootkube.sh[1961]: https://etcd-0.lab.okd.local:2379 is unhealthy: failed to commit proposal: context deadline exceeded
Apr 07 09:07:00 localhost bootkube.sh[1961]: Error: unhealthy cluster
Apr 07 09:07:00 localhost podman[11304]: 2020-04-07 09:07:00.412560122 +0000 UTC m=+5.837612043 container died 7204f3857e1617ce79188015e5d6113e4f79082fa46339f2213fea2368278145 (image=quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:742b729094abbc4ebeaf62323e9393b0d6bf06606d4fe349e8458f9191d9905a, name=etcdctl)
Apr 07 09:07:00 localhost podman[11304]: 2020-04-07 09:07:00.460429288 +0000 UTC m=+5.885481198 container remove 7204f3857e1617ce79188015e5d6113e4f79082fa46339f2213fea2368278145 (image=quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:742b729094abbc4ebeaf62323e9393b0d6bf06606d4fe349e8458f9191d9905a, name=etcdctl)
Apr 07 09:07:00 localhost bootkube.sh[1961]: etcdctl failed. Retrying in 5 seconds...

Can you please guide me if I am doing wrong?

Steps what I performed

1 helper node---Os is Centos 7

1 Bootstrap node-- OS is RHCOS 4.3.8

2 master nodes-- OS is RHCOS 4.3.8

2 worker nodes-- OS is RHCOS 4.3.8

Steps:

1. DNS Server on helper node

2. Configure httpd on helper node

3. Run the following commands on helper node: (./openshift create manifest --dir=<install-confige path>, ./openshift create ignition-configs --dir=<install-confige path>), Copy created files to httpd directory

4. Boots the bootstrap machine using respective ignition file

5. Boots the master machine using there respective ignition file

Now facing the above problem

If I miss any steps please guide me.

[Walid]

so did you set api.lab.okd.local in DNS? if you did, is there something blocking the network traffice as you are getting connection refused? iptables or alike?

[Pawan_GSS]

Yes it resolving and there is no any firewall is between any node and somehow my etcd-0.lab.okd.local does not become ready 

 "localhost bootkube.sh[1961]: https://etcd-0.lab.okd.local:2379 is unhealthy: failed to commit proposal"  getting this error continuously. 

can you tell me at this phase it tries to find the master node and when I checked at master node there is no such port is listening so from which service it referring?

From the last 4-5 days, I am trying to figure it out? but no luck.