I’m having a huge issue with the backups. All the backup tar.gz that have been uploaded to AWS S3 contain only 1 single file with no extension, just the domain name as the name of the file, nothing else. That 1 file is the size of the entire account, for example 3.1GB in size. We don’t know how long this is been happening as weekly backups upto 7 days and Monthly backup upto 2 months seem to all have the same issue. What could possibly be the issue?
File extensions in Linux don’t matter. You can try to rename one of your backups adding .tar.gz extension to it. You can check on the backup file by running the following command:
@Ilia sorry I wasn’t clear earlier. The backup file itself has the extension, but inside it only 1 file instead of all the files and folders.
However, this turned out to be an edge case of only happens when downloading via incongnito mode directly from AWS S3 console.
When downloading the file via something like cloudberry/API, it has all the files and folders as expected directly under the main tar.gz backup file. However, when downloaidng via a browser from s3 cosole, it comes as a tar.gz of another tar.gz.
When using NON-incognito mode to download, the backup has another tar.gz backup inside the main tar.gz instead of the files and folders as usual. You are able to click through and see the files and folders. Not sure why the s3 console is doing this. And when you download in INCOGNITO mode, the backup folder still has the tar.gz in it but this time, it’s without the extension of tar.gz, so it looks like just one file. but when you extract that file and then rename it with extension of .tar.gz, you can now see the folders and files in it.
I’m absolutely puzzled as to why this is differently when downloading directly from S3 console and how it differs as well when you are incognito or non-incognito.
