Skip to content

Commit 5654991

Browse files
authored
Merge pull request #36418 from ktania46/mpio
[BZ: 2004921][MULTIARCH:1392][WIP]: Updates for IBM Power OCP 4.9
2 parents e60f061 + cae2871 commit 5654991

File tree

4 files changed

+95
-49
lines changed

4 files changed

+95
-49
lines changed

installing/installing_ibm_power/installing-ibm-power.adoc

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -79,6 +79,8 @@ include::modules/installation-user-infra-machines-static-network.adoc[leveloffse
7979

8080
include::modules/installation-user-infra-machines-pxe.adoc[leveloffset=+2]
8181

82+
include::modules/rhcos-enabling-multipath-day-1-power.adoc[leveloffset=+2]
83+
8284
include::modules/installation-installing-bare-metal.adoc[leveloffset=+1]
8385

8486
include::modules/cli-logging-in-kubeadmin.adoc[leveloffset=+1]
@@ -103,6 +105,7 @@ include::modules/cluster-telemetry.adoc[leveloffset=+1]
103105

104106
== Next steps
105107

108+
* xref:../../post_installation_configuration/machine-configuration-tasks.adoc#rhcos-enabling-multipath_post-install-machine-configuration-tasks[Enabling multipathing with kernel arguments on {op-system}].
106109
* xref:../../post_installation_configuration/cluster-tasks.adoc#available_cluster_customizations[Customize your cluster].
107110
* If necessary, you can
108111
xref:../../support/remote_health_monitoring/opting-out-of-remote-health-reporting.adoc#opting-out-remote-health-reporting_opting-out-remote-health-reporting[opt out of remote health reporting].

installing/installing_ibm_power/installing-restricted-networks-ibm-power.adoc

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -87,6 +87,8 @@ include::modules/installation-user-infra-machines-static-network.adoc[leveloffse
8787

8888
include::modules/installation-user-infra-machines-pxe.adoc[leveloffset=+2]
8989

90+
include::modules/rhcos-enabling-multipath-day-1-power.adoc[leveloffset=+2]
91+
9092
include::modules/installation-installing-bare-metal.adoc[leveloffset=+1]
9193

9294
include::modules/cli-logging-in-kubeadmin.adoc[leveloffset=+1]
@@ -115,5 +117,6 @@ include::modules/cluster-telemetry.adoc[leveloffset=+1]
115117

116118
== Next steps
117119

120+
* xref:../../post_installation_configuration/machine-configuration-tasks.adoc#rhcos-enabling-multipath_post-install-machine-configuration-tasks[Enabling multipathing with kernel arguments on {op-system}].
118121
* xref:../../post_installation_configuration/cluster-tasks.adoc#available_cluster_customizations[Customize your cluster].
119122
* If the mirror registry that you used to install your cluster has a trusted CA, add it to the cluster by xref:../../openshift_images/image-configuration.adoc#images-configuration-cas_image-configuration[configuring additional trust stores].

modules/installation-complete-user-infra.adoc

Lines changed: 4 additions & 49 deletions
Original file line numberDiff line numberDiff line change
@@ -153,62 +153,17 @@ command.
153153
If the pod logs display, the Kubernetes API server can communicate with the
154154
cluster machines.
155155

156-
ifndef::ibm-power,ibm-z,ibm-z-kvm[]
157-
. For an installation with Fibre Channel Protocol (FCP), additional steps are required to enable multipathing.
158-
+
159-
[NOTE]
160-
====
161-
When installing with multipath, it is strongly recommended to enable it at installation time, and not at a later time, which can cause problems.
162-
====
163-
+
164-
See "Enabling multipathing with kernel arguments on {op-system}" in the _Installing on bare metal_ documentation for more information.
165-
endif::ibm-power,ibm-z,ibm-z-kvm[]
166-
ifdef::ibm-power[]
167-
. Additional steps are required to enable multipathing. Do not enable multipathing during installation.
168-
+
169-
See the _Installing {op-system} and starting the {product-title} bootstrap process_ documentation for more information.
170-
171-
.. To display a boot list and specify the possible boot devices if the system is booted in normal mode, enter the following command:
172-
+
173-
[source,terminal]
174-
----
175-
$ bootlist -m normal -o
176-
sda
177-
----
178-
.. To update the boot list for normal mode and add alternate device names, enter the following command:
179-
+
180-
[source,terminal]
181-
----
182-
$ bootlist -m normal -o /dev/sdc /dev/sdd /dev/sde
183-
sdc
184-
sdd
185-
sde
186-
----
187-
+
188-
If the original boot disk path is down, the node reboots from the alternate device registered in the normal boot device list.
156+
ifndef::ibm-power[]
157+
. For an installation with Fibre Channel Protocol (FCP), additional steps are required to enable multipathing. Do not enable multipathing during installation.
189158
endif::ibm-power[]
190159
ifdef::ibm-power[]
191-
.. All the worker nodes are restarted. To monitor the process, enter the following command:
192-
+
193-
[source,terminal]
194-
----
195-
$ oc get nodes -w
196-
----
197-
+
198-
[NOTE]
199-
====
200-
If you have additional machine types such as infrastructure nodes, repeat the process for these types.
201-
====
160+
. Additional steps are required to enable multipathing. Do not enable multipathing during installation.
202161
endif::ibm-power[]
203-
204-
ifdef::ibm-z[]
205-
. For an installation with Fibre Channel Protocol (FCP), additional steps are required to enable multipathing. Do not enable multipathing during installation.
206162
+
207163
See "Enabling multipathing with kernel arguments on {op-system}" in the _Post-installation machine configuration tasks_ documentation for more information.
208-
endif::ibm-z[]
209164

210165
ifdef::restricted[]
211-
. Register your cluster on the link:https://console.redhat.com/openshift/register[Cluster registration] page.
166+
. Register your cluster on the link:https://cloud.redhat.com/openshift/register[Cluster registration] page.
212167
endif::restricted[]
213168

214169
ifeval::["{context}" == "installing-restricted-networks-vsphere"]
Lines changed: 85 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,85 @@
1+
// Module included in the following assemblies:
2+
//
3+
// * installing/installing_bibm_power/installing-ibm-power.adoc
4+
// * installing/installing_ibm_power/installing-restricted-networks-ibm-power.adoc
5+
6+
[id="rhcos-enabling-multipathday-1-power_{context}"]
7+
= Enabling multipathing with kernel arguments on {op-system}
8+
9+
In {product-title} 4.9 or later, during installation, you can enable multipathing for provisioned nodes. {op-system} supports multipathing on the primary disk. Multipathing provides added benefits of stronger resilience to hardware failure to achieve higher host availability.
10+
11+
During the initial cluster creation, you might want to add kernel arguments to all master or worker nodes. To add kernel arguments to master or worker nodes, you can create a `MachineConfig` object and inject that object into the set of manifest files used by Ignition during cluster setup.
12+
13+
.Procedure
14+
15+
. Change to the directory that contains the installation program and generate the Kubernetes manifests for the cluster:
16+
+
17+
[source,terminal]
18+
----
19+
$ ./openshift-install create manifests --dir=<installation_directory>
20+
----
21+
22+
. Decide if you want to add kernel arguments to worker or master nodes.
23+
24+
* Create a machine config file. For example, create a `99-master-kargs-mpath.yaml` that instructs the cluster to add the `master` label and identify the multipath kernel argument:
25+
+
26+
[source,yaml]
27+
----
28+
apiVersion: machineconfiguration.openshift.io/v1
29+
kind: MachineConfig
30+
metadata:
31+
labels:
32+
machineconfiguration.openshift.io/role: "master"
33+
name: 99-master-kargs-mpath
34+
spec:
35+
kernelArguments:
36+
- 'rd.multipath=default'
37+
- 'root=/dev/disk/by-label/dm-mpath-root'
38+
----
39+
40+
. To enable multipathing post-installation on worker nodes:
41+
42+
* Create a machine config file. For example, create a `99-worker-kargs-mpath.yaml` that instructs the cluster to add the `worker` label and identify the multipath kernel argument:
43+
+
44+
[source,yaml]
45+
----
46+
apiVersion: machineconfiguration.openshift.io/v1
47+
kind: MachineConfig
48+
metadata:
49+
labels:
50+
machineconfiguration.openshift.io/role: "worker"
51+
name: 99-worker-kargs-mpath
52+
spec:
53+
kernelArguments:
54+
- 'rd.multipath=default'
55+
- 'root=/dev/disk/by-label/dm-mpath-root'
56+
----
57+
+
58+
You can now continue on to create the cluster.
59+
60+
[IMPORTANT]
61+
====
62+
Additional post-installation steps are required to fully enable multipathing. For more information, see “Enabling multipathing with kernel arguments on {op-system}" in _Post-installation machine configuration tasks_.
63+
====
64+
65+
In case of MPIO failure, use the bootlist command to update the boot device list with alternate logical device names.
66+
The command displays a boot list and it designates the possible boot devices for when the system is booted in normal mode.
67+
68+
.. To display a boot list and specify the possible boot devices if the system is booted in normal mode, enter the following command:
69+
+
70+
[source,terminal]
71+
----
72+
$ bootlist -m normal -o
73+
sda
74+
----
75+
.. To update the boot list for normal mode and add alternate device names, enter the following command:
76+
+
77+
[source,terminal]
78+
----
79+
$ bootlist -m normal -o /dev/sdc /dev/sdd /dev/sde
80+
sdc
81+
sdd
82+
sde
83+
----
84+
+
85+
If the original boot disk path is down, the node reboots from the alternate device registered in the normal boot device list.

0 commit comments

Comments
 (0)