Just stood up foreman on a CentOS 7 box in AWS. When I log into the web UI, I’m seeing errors about:
Failed to apply catalog: Could not render to json: source sequence is illegal/malformed utf-8
This is my first trial of foreman, so I’m not familiar with the tool or with Puppet. Trying to see if this will be a solution for us to manage our heterogeneous Linux infrastructure.
Any help with this error would be appreciated.
Thanks!
Hi,
could you provide a bit more information? What foreman version do you have? Is the whole UI down and the error is the only thing you can see? Are there any errors in /var/log/foreman/production.log?
I’m using Foreman 1.16.1. The web UI is running and I can log in, but when I log in, it shows that the foreman server is getting errors every 30 minutes. When I drill down, that’s where I’m getting the error:
Failed to apply catalog: Could not render to json: source sequence is illegal/malformed utf-8
I get the same error from running ‘puppet agent -t’ from the command line.
I do not see anything in the production.log that appears to be an error.
I’m also working with AWS support as what little info I’ve been able to find on this has pointed toward the EC2 user-data being gzipped. I have confirmed that the user-data is, in fact, gzipped and I can’t even view it from the EC2 console.
My guess is that this is the root of the issue, so I will post whatever info is relevant as I find out more. In the meantime, if someone can beat AWS support to the punch, there are always bragging rights for that.
I stood up a new EC2 instance outside of OpsWorks. Apparently, OpsWorks adds user-data and gzips it (hence it is binary) which seems to cause this issue.
Also, the installation instructions I had been following referenced specific versions of the various yum repos. For the ones I could do so, I changed the specified version in the URL path to “latest” (instead of, for example, “1.15”).