AWX Large Job Logs do not load in the webinterface

Get Help
1

In Jobs with a very long output, the logs often don’t load completely. This looks like this:

image
image1581×815 36.6 KB

These logs are from long running jobs that are run over 80 or more hosts and parse a lot of data (A lot being about a 10.000-40.000 lines of configuration per Host). My feeling is that AWX loads all these logs into the webinterface, or at least tries to do that, such overloading the page

In this pastebin are my awx-web logs: [pid: 21|app: 0|req: 95/952] 10.42.5.25 () {70 vars in 1417 bytes} [Wed Sep 25 1 - Pastebin.com

The only things I found that could have to with this issue are these:

Wed Sep 25 12:05:49 2024 - SIGPIPE: writing to a closed pipe/socket/fd (probably the client disconnected) on request /api/v2/jobs/24958/job_events/?order_by=counter&page=21&page_size=50 (ip 10.42.5.25) !!!
Wed Sep 25 12:05:49 2024 - uwsgi_response_write_body_do(): Broken pipe [core/writer.c line 341] during GET /api/v2/jobs/24958/job_events/?order_by=counter&page=21&page_size=50 (10.42.5.25)

But I am not sure what to make of that. Have I configured AWX falsely?

2

Hi.
I don’t think you have configured something wrong. I’m experiencing the same thing here. I find it better to download the logs and open them in a text-editor if I’m tshooting. But if there is a solution to this behavior, I’ll be happy to hear the answer too.

3

In my case, they don’t even download completely :confused:
I also cannot get them to load completely via the API. Do you have the same error codes?

4

No, I don’t have those problems.