Bug: Import is not working

[TheZoker]

What happened?

I setup the v4 version following the documentation here: https://zipline.diced.sh/docs/migrate

I exported from v3, setup v4 and then tried to import into v4. But no file is created in v4.

I also do not see any error in the docker log:

Version

Latest v4 release (ghcr.io/diced/zipline or ghcr.io/diced/zipline:latest)

What browser(s) are you seeing the problem on?

Chromium-based (Chrome, Edge, Brave, Opera, mobile chrome/chromium based, etc)

Zipline Logs

[2025-02-17T09:21:15 INFO   db] connecting to database postgres://postgres:postgres@postgres/zipline_v4
[2025-02-17T09:21:17 INFO   server] starting zipline mode="production" version="4.0.0" argv=[]
[2025-02-17T09:21:17 INFO   server] reading settings...
[2025-02-17T09:21:21 INFO   server] server started hostname="0.0.0.0" port=3000

Browser Logs

Additional Info

No response

[diced]

Not sure what you mean by "no file is created in v4", did you follow the steps completely through?

[TacticalTechJay]

OP, did you use the same datasource you had in v3?

[DoctorDinosaur]

Same issue for me, my v3 import is not working but no debug logs are produced.

Clientside, I get this error:

Looking in the Networking tab, the /settings/ API endpoint gives 200 success on importing settings, but the /import/v3 endpoint gives a 502 error:

All the Debug logs show is the settings import.

If I turn off the Settings import and try only to import the data, the logs don't show anything at all.

Additional info:

• I'm behind Cloudflare
• The json file I'm importing is 6mb, so I don't think its a chunking error
• I'm behind an apache reverse proxy, but its logs only show AH01084: "pass request body failed" which indicates the issue stems from Zipline, further down the stack (I think)

[DoctorDinosaur]

Bypassing cloudflare and directly connecting via IP, I get "Request body is too large"

Possibly related to #756 but Cloudflare was hiding that error and showing a 502?

[DoctorDinosaur]

I tried 4.0.0 as per #756 and it works fine, so it's the same issue with request body too large. Could be duplicate, but not my issue to close.

[diced]

closing as stale, the op hasn't responded in months