EXECUTOR UPDATE ISSUE

Hi,
I recently updated the Knime server pre-installed on a VM by my colleague, to the latest version - 4.8.2 to get all the latest extensions. The update went alright, Then, when I tried updating the executor, I got the messages as shown in the attached file. Also, following this, I have lost the ./update-rmi.sh file itself. What could be the reason?

Is there an alternative to update the executor using the .zipped file on Linux?

Thanks in anticipation.update-log.txt (8.8 KB)

Hi there. I think that this happens if the update fails before the serverspace is reinstalled. I’ve opened a ticket to get to the bottom of the issue.

You might find the easiest option to take a pre-installed executor from the 4.8.2 downloads page:
https://www.knime.com/knimecom-product-downloads-2018-12

It’s then just a case of extracting the archive and updating the knime-server.config knime_exe option to point to the newly extracted executor.

Best,

Jon

Hi Jon,

Many thanks for your reply.

So, to clarify, should the archive be extracted in a particular folder?

Thanks

You can extract that to anywhere, but I’d recommend putting it in the enclosing directory for the current executor. Also be sure that the permissions for the whole extracted directory are set correctly.

Hi Jon,

I have now installed the pre-built executor and updated the server.config file. I can see that the new executor has following 3 files missing which I can see in the original executor -

  1. .VERSION FILE
  2. CONTENT.JAR
  3. ECLIPSE.INI

Can you please confirm if these files are not required in the latest version, or I should replace them in the new executor from the original?

Thanks

Hi Jon,

Also, now I see the below message on my URL -

image

Do I need to reconfigure something in the apache server as well?

Thanks

Hi,

We’d need to take a look at the log file to figure out why the server isn’t being deployed. Relevant information will be in the /logs/localhost..log and /logs/catalina..log files.

Hi Jon,

I just managed to turn it on. Looking at the log files I could see that the previous Apache process running at the same port was the issue. The issue is resolved after killing the earlier process.

Everything looks good now. Many thanks for the help.

Thanks

1 Like

Thanks for passing on the good news! Have a nice weekend.