Incapability to run queries - MB

#1

Dear MB support,

Since last week, I'm unable of:

  • Running scripts with a size bigger than 15KB (Where supposedly the limit we have is 1MB)
  • Save queries where big tables are called

To me it seems that the something somewhere is trying to save computational/processing effort, but it comes from the product (MB) and not changes in the way we work.

Is this error familiar to you?

image
image.png1143×550 30.1 KB

image
image.png936×858 57.7 KB

Thank you,
Pere

#2

Hi @perfuses
The problem comes from your reverse-proxy in front of Metabase, not Metabase itself.
You are using Nginx, so you'll need to make sure the configuration can handle what you need.

#3

Hi Flamber. Thanks so much for the help here.

I'm working as data scientist and not data engineer, and I'm not fully familiar with the concept.

(1) Do you mean that MB runs behind the reverse-proxy, so the script size is maxed out via Nginx configuration? (2) Could you describe what is Nginx in simple terms?

(3) What about saving queries? Would the problem come from the same source (Nginx configuration)?

Thanks so much, Really!

Pere

#4

@perfuses You'll need to contact some IT folks to help you.

  1. I cannot tell, but since your other problems comes from Nginx, then I'm guessing it's also causing this
  2. Here's an explanation of a reverse-proxy: https://www.cloudflare.com/learning/cdn/glossary/reverse-proxy/
  3. The error comes from Nginx, as you can see from the error

But make sure you are using the latest release of Metabase: https://github.com/metabase/metabase/releases/latest

#5

Thanks so much!

After some checks, I've found that we use Beanstalk. This NGINX is set up automatically with the software that MB people offer on the website.

Therefore, agreed, the mistake does not seem to be from the product. Still it's still something that people from MB has done instead of coming from the configuration.

We use the last version of Metabase. Checked. We're investigating this further.

Thanks so much,
Pere

#6

@perfuses Logs are everything. You'll need to check the EB logs from both Nginx and Metabase, so you can understand what the problem is.

#7

Perfect, so you mean that in order to identify the problem we must look at the EB logs?

#8

@perfuses Yes, you are posting the errors returned, but that doesn't tell what the underlying cause is.