Print

Print


Hi Andy,

Some comments on your four points above:
a) yup, absolutely,

b) Ideally xrootd would select the requested algorithm(s) with the highest (non-zero) q value. However, I'm pretty sure selecting the first supported algorithm (ignoring any q-values) will not break anything. Just by-the-by, dCache (IIRC) builds the Want-Digest with descending q-value, based on the number of bits. If xrootd selects the first algorithm, it should even select the one dCache prefers :-)

c) This is perfectly fine behaviour. The xrootd could also return an empty "Digest" header; i.e.,

Digest: 

d) Returning an error is certainly OK (any HTTP request may return a 500 status code) but, in this case, I don't think it is necessary. The xrootd server could simply return an empty "Digest" header to indicate that none of the client's requested algorithms are supported.

Just to confirm: RFC 3230 defines the digest-algorithm values in terms of "token" (as per RFC 2616). A token may not have spaces, commas, semicolons or colons. Therefore, it's safe to use any of these to build the list.

HTH,
Paul.


Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.Message ID: <xrootd/xrootd/issues/1707/1377841608@github.com>

[ { "@context": "http://schema.org", "@type": "EmailMessage", "potentialAction": { "@type": "ViewAction", "target": "https://github.com/xrootd/xrootd/issues/1707#issuecomment-1377841608", "url": "https://github.com/xrootd/xrootd/issues/1707#issuecomment-1377841608", "name": "View Issue" }, "description": "View this Issue on GitHub", "publisher": { "@type": "Organization", "name": "GitHub", "url": "https://github.com" } } ]

Use REPLY-ALL to reply to list

To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1