1 Answer
- Newest
- Most votes
- Most comments
0
There seems to be many similar cases but no clear single root cause for these. As it works with fresh instance, you could try cleaning (or disabling) cache on the instance. If having filesystem access to data in S3 bucket is critical I would also consider more robust options than s3fs. One such would be FSx for Lustre. It might be an over-kill for small scale use-cases but it does offer a managed and reliable 2 way sync between filesystem and S3.
https://aws.amazon.com/blogs/aws/enhanced-amazon-s3-integration-for-amazon-fsx-for-lustre/
Relevant content
- asked 2 months ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 2 years ago
Thanks for that. AFAIK, the s3fs cache option isn't there. So from what I understand from the man page, cache should not be used. Is there another cache that could be involved ?