Feed Retrieval Issue - ERROR Bee responded with HTTP 404 (Not Found)

[rampall]

⚠️ Support requests in an issue-format will be closed immediately. For support, go to Swarm's Discord.

Context

Bee version - 2.3.2-47c699e6 Swarm CLI 2.22.0

Upload to Feed works. But Errors while attempting feed retrieval:

$ swarm-cli feed upload file.txt -T "rp_file5" -i rp --curl --verbose
curl "http://localhost:1633/stamps" -H "accept: application/json, text/plain, */*"
? Please select a stamp for this action
curl -X POST "http://localhost:1633/bzz?name=file.txt" -H "accept: application/json, text/plain, */*" -H "swarm-postage-batch-id: cf0e4ef5ce2694f9cc92c6c65304dfb8fb75a9246ea6f02995c146c754938d0d" -H "swarm-deferr text/plain" --data @file.txt
Data has been sent to the Bee node successfully!
Swarm hash: 12dca0af8d4b5daaf8bcf724c4eb7a6bcdb8b21f72c25c48b499b56cd29b14f8
Waiting for file chunks to be synced on Swarm network...
Uploading was successful!
URL: http://localhost:1633/bzz/12dca0af8d4b5daaf8bcf724c4eb7a6bcdb8b21f72c25c48b499b56cd29b14f8/
? Please provide the password for this V3 Wallet [input is hidden] rp
curl -X POST "http://localhost:1633/feeds/351f184c7c469afec723b0b91c5e8c909d5b33a5/569cd4dfc67a8362596eaa82e019d92d440a8c301c7a08ed7548351193575ef7" -H "accept: application/json, text/plain, */*" -H "swarm-postagc65304dfb8fb75a9246ea6f02995c146c754938d0d"
curl "http://localhost:1633/feeds/351f184c7c469afec723b0b91c5e8c909d5b33a5/569cd4dfc67a8362596eaa82e019d92d440a8c301c7a08ed7548351193575ef7?type=sequence" -H "accept: application/json, text/plain, */*"
curl -X POST "http://localhost:1633/soc/351f184c7c469afec723b0b91c5e8c909d5b33a5/14ba09152e16d076f5aebb9b9d4b928dca7057ae7fc1031d5eed488cd06a1953?sig=63835944340e42aafc0bf6acd58f8d8284bef4fbecec5703a90bd18528ae77b7a51c6e506cd275d556c22dc3047c431c" -H "accept: application/json, text/plain, */*" -H "content-type: application/octet-stream" -H "swarm-postage-batch-id: cf0e4ef5ce2694f9cc92c6c65304dfb8fb75a9246ea6f02995c146c75
Chunk Reference: 12dca0af8d4b5daaf8bcf724c4eb7a6bcdb8b21f72c25c48b499b56cd29b14f8
Chunk Reference URL: http://localhost:1633/bzz/12dca0af8d4b5daaf8bcf724c4eb7a6bcdb8b21f72c25c48b499b56cd29b14f8/
Feed Reference: 8971d0748cc61e79c2d1fada34dd1591ca754cb8e686c255eeced29d35bedd87
Feed Manifest: 72f306aee8a414a5bd8082344d709f448e25472bac3564563268bbf7e8f4cdbf
Feed Manifest URL: http://localhost:1633/bzz/72f306aee8a414a5bd8082344d709f448e25472bac3564563268bbf7e8f4cdbf/
curl "http://localhost:1633/stamps/cf0e4ef5ce2694f9cc92c6c65304dfb8fb75a9246ea6f02995c146c754938d0d" -H "accept: application/json, text/plain, */*"
Stamp ID: cf0e4ef5
Usage: 30%
Capacity (immutable): 11.250 GB remaining out of 16.000 GB
Depth: 22
Bucket Depth: 16
Amount: 2958076800
Usable: true
Utilization: 19
Block Number: 38741032
Successfully uploaded to feed.

$ swarm-cli download 72f306aee8a414a5bd8082344d709f448e25472bac3564563268bbf7e8f4cdbf --verbose --curl
curl "http://localhost:1633/bytes/72f306aee8a414a5bd8082344d709f448e25472bac3564563268bbf7e8f4cdbf" -H "accept: application/json, text/plain, */*"
curl "http://localhost:1633/feeds/351f184c7c469afec723b0b91c5e8c909d5b33a5/569cd4dfc67a8362596eaa82e019d92d440a8c301c7a08ed7548351193575ef7" -H "accept: application/json, text/plain, */*"
curl "http://localhost:1633/bytes/72f306aee8a414a5bd8082344d709f448e25472bac3564563268bbf7e8f4cdbf" -H "accept: application/json, text/plain, */*"
curl "http://localhost:1633/bytes/72f306aee8a414a5bd8082344d709f448e25472bac3564563268bbf7e8f4cdbf" -H "accept: application/json, text/plain, */*"
curl "http://localhost:1633/bytes/8504f2a107ca940beafc4ce2f6c9a9f0968c62a5b5893ff0e4e1e2983048d276" -H "accept: application/json, text/plain, */*"
/
curl "http://localhost:1633/bytes/0000000000000000000000000000000000000000000000000000000000000000" -H "accept: application/json, text/plain, */*"
ERROR Bee responded with HTTP 404 (Not Found).

The error message is: Request failed with status code 404

There may be additional information in the Bee logs.

Summary

Expected behavior

Actual behavior

Steps to reproduce

Possible solution

[Cafe137]

OK, this seems to be a bug. It should not attempt to download the 00...00 entry, which is not a real chunk, just placeholder for metadata.

[Cafe137]

By upgrading to the new Bee-JS, this issue seems to have gone away, but there's a different one.

The feed indices now contain payloads instead of a reference to the payload, which is causing issues in BzzAddress class.

I already have a failing test for this case, fixing it right now.