Traceback (guilds not appearing)

[baka-yuki]

Expected Behavior

Application starts
...

Actual Behavior

See logs and notes
...

Steps to reproduce the problem

1. Download docker image
2. Attempt to run
3. (See notes)

Specifications

Environment

• Operating system: Linux (see comments
• CPU Architecture: x86_64 (PC)
• Flavor: Docker

DiscordChatExporter-frontend

• Release version or commit hash: docker hash - c31fe828b048 (most recent main)
• Can the issue be reproduced in the demo?: Don't know

DiscordChatExporter

• Version (CLI --version): v2.42.2

How did you export the data?

• Export format (CLI --format): Json
• Downloaded assets (CLI --media): Yes (and --reuse-media_
• Markdown processing (CLI --markdown): set to false
• Other CLI options (optional): ...

Logs

Is there any relevant information in logs? If so, please provide them:

2023-11-05 04:49:44.915173+00:00[Preprocess] took 0.64 seconds
2023-11-05 04:49:44.915217+00:00Traceback (most recent call last):
2023-11-05 04:49:44.915228+00:00File "/dcef/backend/preprocess/main_mongo.py", line 146, in <module>
2023-11-05 04:49:44.915368+00:00main(input_dir, output_dir)
2023-11-05 04:49:44.915413+00:00File "/dcef/backend/preprocess/main_mongo.py", line 119, in main
2023-11-05 04:49:44.915485+00:00channel_cache = ChannelCache()
2023-11-05 04:49:44.915507+00:00^^^^^^^^^^^^^^
2023-11-05 04:49:44.915521+00:00File "/dcef/backend/preprocess/ChannelCache.py", line 22, in __init__
2023-11-05 04:49:44.915568+00:00os.makedirs(self.cache_folder_path)
2023-11-05 04:49:44.915594+00:00File "<frozen os>", line 215, in makedirs
2023-11-05 04:49:44.915605+00:00File "<frozen os>", line 215, in makedirs
2023-11-05 04:49:44.915617+00:00File "<frozen os>", line 215, in makedirs
2023-11-05 04:49:44.915624+00:00[Previous line repeated 1 more time]
2023-11-05 04:49:44.915630+00:00File "<frozen os>", line 225, in makedirs
2023-11-05 04:49:44.915636+00:00OSError: [Errno 30] Read-only file system: '../../release'

Notes

This is actually running on a TrueNAS SCALE system, which is a debian linux appliance that runs a kubernetes cluster. I've ran into a few weird readonly errors prior to this but was able to resolve them. After I saw the most recent commit was to fix the docker image, I also made sure to pull the image by sha256 digest to make sure it was the most up to date.

/dcef/cache -> PVC Volume
/dcef/exports -> points to my DCE exports directory. Exports were generated with incremental chat exporter.
/var/lib/nginx -> PVC Volume (without this, /var/lib/nginx/body can't be written, NGINX throws an EMERG error and the container quits)
/var/log/nginx -> PVC Volume (without this, being unable to write access.log triggers an EMERG error and container quits)

The container is running as uid/gid 0/0 (root) as without this an error is thrown. Ports are mapped 21011 to 21011, which then feeds into a kubernetes ingress controller. Access works fine, webpage loads fine, it's only loading the guilds that fails.

I know this is probably beyond what was intended, so if there's anyway I can help let me know.

[baka-yuki]

Hey got it working. If i mount a PVC volume to /dcef/release/dcef/storage/cache/message-ids everything begins to work. I got this path from ChannelCache.py and the run_container script. Is there any way that this directory could be moved into /dcef/cache along with maybe nginx logs and nginx/body?

[slatinsky]

Hello, as mentioned in your pull request #44, project needs bigger folder restructure to fix this issue. Please use your workaround for now.