Not able to launch MapR hadoop cluster

Hi Team,

I am trying to launch a MapR hadoop cluster using AWS Marketplace but at final time it is getting rolled back. attaching screenshot for reference.

Hi kushal,

Can you please provide the details of cluster you are using, and please attached the screenshot which you are referring.

Thanks

I do wanted to attach screenshot but there are approximately 13-14 screenshots. I created a pdf file having all those screenshots but when I am uloading on this forum. Website gives me an error that “new user cannot uplaod files”.
I am using below link to launch a MapR Hadoop cluster of 3 nodes using “Mapr Enterprise Database Edition with Spark” image.
But in the end stage cluster creation gets rolled back.

But when I am trying to launch a cluster using “MapR Community Edition”
I am successful

Please suggest

Hi,

Few things to check, if you are using the IAM account to launch the cluster check you have necessary permissions to launch the instances with required type.

Check that you have IAM permission to launch the paid MapR cluster.

As you are a new user, you dont have permissions to upload the attachments. As time passes and based on the interaction with the forum you will get the access to upload.

I asked my Forum team to update your permission levels, now you can upload the attachments.
Please check.

Thanks

Thanks…
Please refer attachment, we are the owner and I have complete access to AWS account.Marketplace_POC_screenshot.pdf (1.6 MB)

Hi,

I have gone through the attached document, in last screenshot it seems there is an error while creating an autoscaling group. It would be helpful if you expand the error part and take the screenshot and reply with that screenshot or the error text.

Also try to stop rollback on failure by selecting the Rollback on failure to No in Advance section of Options part while creating the Stack. That would be helpful in troubleshooting.

You can also try to launch it in other region and see whether it works or not.

Thanks.

HI Team,

I tried with disabling “Rollback on failure to No” the it is going into endless “Create in Progress” state.

Attaching error screen shot

Hi,

It is okay if the create progress state is endless. You can delete the stack later which whould terminate the instance. For now, while it is in progress, you can log into or SSH into the EC2 instance and analysis the logs file. You will find the detailed logs at /var/log/cfn-init.log, and try to figure out the cause of failure.

And please try to launch one instance instead of three and check.

Thanks

Before doing this do I need to buy M5 or M7 separately.
Is that the issue?

Hi,

You can also try to launch a MapR cluster using the Amazon EMR services.
When you launch a cluster, in Software Configuration it will ask you to select the Vendor, there you have to select MapR and after that select the MapR Release you want to install and fill up the required details and create cluster.

Thanks