[Keep] Keepstore and keepproxy support HTTP Range requests
If a request has a Range header, read only the desired portion from the data volume, and return the segment with appropriate HTTP status (206) and headers (e.g., Content-Length: 10001, Content-Range: 1234-11234/100000).
Skip hash verification (this would be impossible without reading the whole block). If a client needs to verify content hashes, it can either retrieve the whole block ahead of time and compute the checksums of each segment, or retrieve ranges in order and compute the hash piecewise.
- Subject changed from Keep supports partial content to [Keep] Keep supports partial content
- Target version changed from Arvados Future Sprints to 2014-10-08 sprint
- Assigned To set to Brett Smith
- Target version changed from 2014-10-08 sprint to Arvados Future Sprints
- Target version changed from Arvados Future Sprints to 2014-10-29 sprint
- Subject changed from [Keep] Keep supports partial content to [Keep] Keepstore, keepproxy, and Python SDK use HTTP Range to retrieve partial content where appropriate
- Story points changed from 1.0 to 2.0
- Target version changed from 2014-10-29 sprint to Arvados Future Sprints
- Subject changed from [Keep] Keepstore, keepproxy, and Python SDK use HTTP Range to retrieve partial content where appropriate to [Keep] Keepstore and keepproxy support HTTP Range requests
- Description updated (diff)
- Assigned To changed from Brett Smith to Tom Morris
- Assigned To deleted (
Dear Team, Could we please assign a higher priority to this issue? Keepweb is extremely slow while keepproxy performs decently but is handicapped by the lack of Ranges support. It is quite a basic thing to have. Thank you in advance. Best regards, -- Stanislaw
Speaking strictly only in private capacity here. This issue is very high on my wish list.
Also available in: Atom