Redshift 3PL using AWS portal problem

I’ve already put this question in my other thread about redshift issues, but this is better placed in it’s own thread for future googlers :slight_smile:

Anyways… Ive’ set everything up for using RedShift using the AWS portal and most things are working… what’s still a problem is the on demand 3th party licensing.

The docs that are there seem a bit disjointed with what’s actually happening on AWS portal and I’ve tried a lot of things but it just isn’t working and there are no logs, errors or other clues…

So what are the steps to make this work? I’ve got a redshift.pxf and then… ?

Hey Jonathan,

Do you have Tools > Configure Repository Options > Usage Based Licensing > Dynamic Licensing enabled? There may currently be a bug when having Dynamic Licensing and Redshift UBL enabled. If you do have both these enabled try disabling Dynamic Licensing and restarting the slaves. See if Redshift UBL starts to work.

Additionally please send over the job report, in case it’s something else. Also here are some troubleshooting steps to try out:

https://docs.thinkboxsoftware.com/products/deadline/10.0/1_User%20Manual/manual/aws-portal-troubleshooting/job-report-says-it-failed-to-checkout-third-party-license.html

Regards,

Charles

Hi!

I’m not using the dynamic licensing, just UBL for deadline which seems to work fine, the slaves are rendering fine… it’s just that I’m getting a redshift watermark. Other then this line there is not feedback:

STDOUT: Failed to load file /home/ec2-user/redshift/redshift-core2.lic (does not exist). Will have to communicate with server.

(I’ve dont have the farm running now, once I do some more test I’ll put the entire report here)

But I’m having a hard time figuring out the certificate…

I’ve downloaded the redshift.pfx file. and by the looks of it the gateway instance that is started for the portal is running a licensing forwarding server.

I guess i should upload the pfx to it using the ‘upload certificate’ button and that should be it right?

The docs mention having to set up a licensing forwarding yourself but I doubt that applies for the portal setup since it’s already running one.

(I did give it a try to setup my own which I coudnl’t make work… (errors on port 443, when fixed it gave errors about untrusted certificate chains and what not… the cert utility in mentioned in the docs in nowhere to be found, etc… but let’s save that for another day if it’s not required for the AWS Portal)

In the AWS Portal Panel > AWS Portal Settings > Advanced setup the Sync directory to point to the location of your UBL certificates. When you start an infrastructure the certs will be uploaded to an S3 bucket matching your infrastructures stack-id. You can access S3 from the AWS Management Consoles to view if your certificates are uploaded. You can also right click the infrastructure and upload the certs manually.

If you are not using Third Part UBL locally do not run a local License Forwarder. Stop that license forwarder and remove it from the Forwarder Panel. Since it was have certificate issue, it sounds like you have Web Forwarding enabled in Tools > Configure Repository Options > Usage Based Licensing. Double check that and make sure it is disabled. If you have to disable it, you will need to restart you License Forwarder so it picks up the new setting to not use Web Forwarding.

When you are able please send the full job report over.

If this is a custom AMI, when installing Redshift did you run the floating version of the installer. A floating installer would be indicated by “-flt.run”

1 Like

ahh crap… floating version redshift… that never occurred to me! :slight_smile: I’ve actually installed the demo version, assuming it was all the same binary and a found license would make it go no-demo. I haven’t bought Redshift yet since since I wanted to test properly in my pipeline/farm settings before committing to it so I only have access to the demo binary.

But in the end I’d like to be not depend on you guys for updating to the latest RS version, right now it’s lagging behind quite a lot ( 2.5.60 on AMI vs latest release 2.6.22 ). When do you guys usually update these kind of things?

Normally you’d reach out to Redshift support:

As far as when we upgrade, I’ll need to raise that with the team to find out.

Yeah, I’ve talked to RS people and they can supply me with the correct RS installers (that support floating lics) need for upgrading the AMI… but for those who use the base AMI without customization they relay on you guys for maintaining up to date versions… So I wondered how that would go and if you guys upgrade once a month or only when enough people ask for it :slight_smile: There is always that only nighly build that fixes that one bug that you run into… so having some control over the render slaves is important.

Once RS has given you access to the floating installers you should be able to get updated versions in the future at your leisure. So you should not be in the situation where you are stuck waiting for a RS installer or us to come out with the latest RS on our AMI. You will have the control to create your own AMI with the versions of software you need.