Closed gabmacinas closed 2 years ago
Hi @gabmacinas, thank you very much for taking time to test v2 and provide me with some feedback. I really appreciate it!
I can't reproduce this issue, but the error is basically saying the the source file cannot be found (on the private
side) when trying to copy it. It's talking about both metadata and assets, may you please show me a couple of filenames in your private
folders in order to check they are actually correct?
Also, can you make sure the S3_PATH_PREFIX
is empty as well? (I see your folders seem to be in the root folder of your bucket)
Thank you.
Hello @liarco here's my file structure inside the private assets and metadata folder
also S3_PATH_PREFIX is by default to be null in my environment variables as well
also does the repository affects the buckets region? I certainly doubt that this would be the case since my bucket was deployed in NYC3 instead of FRA1 and changed the parameters to the environment variables.
@gabmacinas everything looks fine there. The region is absolutely important (as well as the bucket name and all the other configuration data related to DigitalOcean's spaces), but it should be included in the S3_ENDPOINT_URL
(each region has its own URL, https://nyc3.digitaloceanspaces.com
in your case).
If the endpoint URL is also correct I will try a deployment on the NYC3
region ASAP in order to see if I can replicate the issue on my side... I have everything set up for FRA1
and it works as expected there, it looks like the only difference is the region (I can't think of other configuration parameters that might be causing the 404 error).
@gabmacinas a little update:
I just saw that the error is [metadata-provider] [2022-06-10 10:45:39] NoSuchKey: null
, that null
there sounds a little strange to me... I'm gonna force a 404 error on my side to check that, but I think it should be telling you the exact key that couldn't be found. The fact that it says null
makes me thing there might be something wrong with the data from the contract.
Is it possible for you to share the contract address in order to run my tests on that as well?
heres the contract address that I've used rinkeby : 0xc52E65A3868642200442c38A855DDd737743cde3
@gabmacinas everything looks fine there. The region is absolutely important (as well as the bucket name and all the other configuration data related to DigitalOcean's spaces), but it should be included in the
S3_ENDPOINT_URL
(each region has its own URL,https://nyc3.digitaloceanspaces.com
in your case).If the endpoint URL is also correct I will try a deployment on the
NYC3
region ASAP in order to see if I can replicate the issue on my side... I have everything set up forFRA1
and it works as expected there, it looks like the only difference is the region (I can't think of other configuration parameters that might be causing the 404 error).
For this both the s3 and the App are in the same region and I've also noticed that the key is null. As for the contract I'm using the ERC721A collection.
@gabmacinas Ok, but I guess it's our own version of the contract (I mean, the one by the HashLips Lab, based on ERC721A
), because the default code here is made to work with that version, so the original ERC721
(or other customizations) might require some changes in the source code.
@gabmacinas I'm sorry but this was closed automatically by mistake.
I'm sorry if this took me a lot of time to get back to you, but finally I hope I found out what's happening. I had a situation recently where I got the same error as yours and it seems to be related to leading /
in object keys which are not working as expected.
The error was occurring depending on the way a user was configuring the paths (e.g. public/assets
or /public/assets
). Now I'm sanitizing all the keys in order to avoid this kind issues no matter what the configuration is.
May you please let me know if this fixes the error? Thank you.
@gabmacinas I'm sorry but this was closed automatically by mistake.
I'm sorry if this took me a lot of time to get back to you, but finally I hope I found out what's happening. I had a situation recently where I got the same error as yours and it seems to be related to leading
/
in object keys which are not working as expected.The error was occurring depending on the way a user was configuring the paths (e.g.
public/assets
or/public/assets
). Now I'm sanitizing all the keys in order to avoid this kind issues no matter what the configuration is.May you please let me know if this fixes the error? Thank you.
Hi @liarco, I'm closing this issue as I've tested your recommendation above by adding /
in S3_PATH_PREFIX
variable. Thanks a lot! It Works!.
I've check on my environment variable to see if all of the values we're in the same format but throws an error when copying the data from private to public folder
folder structure