8 Node Metrocluster ONTAP upgrade

[AmritaS]

Can you please clear out my doubt on 8 node Metrocluster ONTAP upgrade  with automatic non-disruptive procedure .Do we have to execute the "cluster image update" command on both clusters( one in site A & one in site B) simultaneously or we just have to execute this on one cluster in site A ,would it automatically upgrade the cluster on other site as it is Metrocluster ? Or we have to execute the update command on cluster in site A first ,after completion of site A, we will start site B ?

Thank you in advance .

[pedro_rocha]

Hello my friend,

Did you generate an Upgrade Advisor for this procedure?

Regards,

Pedro

[AmritaS]

yes I have generated and  gone through the document ,But No where it was mentioned my above query .

[akiendl]

you only execute the update command on one cluster.

It will then always update one node (Site A) and its DR partner (Site B) at the same time to ensure that a switchover would be still possible.

ak.

[AmritaS]

Thank you for the information .Moreover ,I would like to understand more to get the clarity further .suppose ,We have cluster1 with 4 nodes in site A and cluster2 with 4 nodes in site B .both are active-active .First I have to download the image to all 8 nodes on both clusters .Is it correct ? Then I have to run the update command on only cluster1 . update will start to dr-pairs (Cluster1_node1 and cluster2_node1) ,once its completed ,update will start for DR-pair(Cluster1_node2 and  cluster2_node2),after that DR-pair(Cluster1_node3 and  cluster2_node3) and then DR-pair(Cluster1_node4 and  cluster2_node4).is this the correct process  ? Thank you 

[pedro_rocha · Accepted solution]

hi

download the image to each cluster: https://docs.netapp.com/us-en/ontap/upgrade/task_download_and_install_ontap_software_image.html#for-manual-upgrades

system node image update -node * -package location -replace-package true -background true -setdefault false

https://docs.netapp.com/us-en/ontap/upgrade/task_upgrade_andu_cli.html#resuming-an-upgrade-using-the-cli-after-an-error-in-the-automated-upgrade-proce...

"For any MetroCluster configuration, except a 2-node MetroCluster system, the ONTAP upgrade process starts simultaneously on the HA pairs at both sites (the local site and the disaster recovery site) after the user initiates and provides confirmation on the command line. For a 2-node MetroCluster system, the update is started first on the disaster recovery site, that is, the site where the upgrade is not initiated. After the update is fully completed on the disaster recovery site, the upgrade begins on the local site.

"

[akiendl]

you use clister image commands and that will distribute it to all nodes. (both clusters)

the. you validate on both sites

then you trigger the update on one site

it then will update in the order you wrote.

you really should not use system node commands for the images.

ak.