Hello,

I love the design of S3QL so much that I want to use it in my homelab to 
replace my whole storage pool. But I have encountered a problem that when 
storing over 10 million files in a mount point, S3QL creates so many objects in 
a single bucket, without any prefix to divide these files into sub-directories. 
This behavior is causing my object storage backend (self-hosted MINIO) creating 
over 10 million files under a single directory, which is very slow on any 
operation that need list folder (any filesystem could not handle so many files 
in a single directory).
I have checked the document carefully, but I could not find any instruction on 
how could I change this behavior. What I want is let S3QL stores objects with 
prefix, like when it want to store s3ql_data_1197116, the directory structure 
would be s3ql_data_1/s3ql_data_1197/s3ql_data_1197116, so that any single 
directory would have no more than 1k files/sub-directories.
Could you please help to take a look and share some idea? I know this could be 
a backend specific problem, that AWS or other cloud provider do not have this 
issue. But I indeed appreciate the design of S3QL and want to use it in every 
place I could. Thanks a lot!

Best regards,
Moon

-- 
You received this message because you are subscribed to the Google Groups 
"s3ql" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/s3ql/C080486A-797E-45B0-928E-5FC08C1A6B6E%40gmail.com.

Reply via email to