lab environment fails every time

[falko]

Hello, 

I do RH318 but since last week I´m not able to start any labs. I always get the following error:

"Verifying access to the RHV-M API ... failed"

I waited up to two hours for it to come up and recreated it two times but nothing helped.

What can I do here?

Thanks and best regards,

Falko

[Chetan_Tiwary_]

@falko in that case raise a support case to take help from our backend support team.

[vrait]

I'm having the exact same issue.

[Travis · Accepted solution]

@falko @vrait  -

Currently you are experiencing issues with the self-signed SSL certificate for the RHVM (ovirt-engine) service. This is preventing the WebUI from working as well as the RHVM-API from working. We have a solution in place and are rolling out new images for a new lab environment with certificates that have been renewed and should be good until 2029. Unfortunately, there were some other certificates expiring on those systems early next week and have renewed those as well.

We expect to have a new environment out early next week (hopefully by Tuesday) that has been tested and is fully working as we've had to manually rebuild the hypervisor images, the utility server image, and the RHVM VM.

For now, as a work-around, you should be able to set the time back by a week on the systems in order to make things working. You would need to SSH to the RHVM instance to change the date/time as well as the hypervisor hosts and the workstation machine so that the SSL certificates aren't viewed as expired.

I will follow-up on this post once the new images have been pushed to production and have been tested. At that point you will need to delete the existing lab environment so that the new images can be leveraged when creating a new RH318 environment.

@Chetan_Tiwary_ - FYI ....

[vrait]

Thanks for the update, I was able to find this out after digging through the ansible playbooks.

Setting the time back did work for me. For setting time back I used "date --set='1 month ago'"

[vrait]

Also, in Lab 4.7 I notice the labadmin user's password has expired and is unsable.

I plan to hopefully take the EX318 exam this month, do you know if there are any similar issues that might arise with that?

[Travis · Accepted solution]

The exam environment is completely separate build from the lab environments so it should not appear in the exam.

[Travis]

@vrait -

Can you confirm that you ran the "lab start" command and that it completed successfully? There have been issues with the IdM passwords in the past because they expire so quickly, but the lab start commands are supposed to reset the passwords and ensure they are current for the IdM users (which is something separate from the SSL in-product certificates that are expiring.

The issue with SSL certificates arise because they typically renew on a running environment, however, our lab environments revert back to the original images which reverts back to the original SSL certificates.

[falko]

I have just checked and it is still not working. And again was the ticket set to resolve. I have wrote them and hoping for help. Let me know if I should check something.

[Travis · Accepted solution]

@falko -

You will need to delete your lab environment and create a new one. You are correct in that the ticket was updated as we did build new images for the entire lab environment. It was necessary to fix SSL certificates on RHVM, but we also had to replace all the hypervisor certificates as well as updating IdM on the utility server. The problems were identified and fixed in the images which is why the ticket was marked to resolved after things were tested, but the images weren't fully pushed and available to all of production. At this time, I can confirm I've tested the course in the ROLE Production environment and both RHVM and the IdM server all have renewed certificates and the lab functions properly.

One more thing to keep in mind, because this is such a large environment, it can take up to 10 minutes for the RHVM machine to be up and the lab start command checks the RHVM API. If the machine isn't up, the script will still fail (like it did with the SSL issue), but this will be because the service hasn't started.

@Chetan_Tiwary_ ... FYI