inline. [EMAIL PROTECTED] wrote: > > Reposting my query as there were no responses. Could someone > confirm whether this is an error in the RFC...and if not, the > reasoning behind the interpretation suggested in the RFC. > > Thanks in advance > Subhash Nayak. > Hughes Software Systems > http://www.hssworld.com > > ============================================================ > Hi, > I had a query with respect to section 16.9 of bis-09 > which talks about how proxies must handle transport errors. > This section states that any transport layer errors (like > ICMP errors) must be interpreted as equal to receiving a > 400 Bad Request. But is this the correct response code that > a network error must be equated to ? Because in this case, > neither is the request malformed, nor is it a client error. > I would assume that the correct interpretation in this case > should be a 503. Interestingly, the 503 interpretation is > prescribed for user agents as the following line from the > "General User Agent Behaviour" section shows: > > Line 1162: > "If a fatal transport error is reported by the transport > layer (generally, due to fatal ICMP errors in UDP or > connection failures in TCP), the condition MUST be treated > as a 503 (Service Unavailable) status code." > > Why is this difference of interpretation between UA's > and proxies. Shouldn't it be interpreted as a 503 at proxies > as well ? Was there some discussion/open-issue on this which > concluded that proxies must interpret network errors as 400 ?
I think this is an error. I don't recall that the choice of 400 in one case, and 503 in the other, was intentional. In practice it doesn't really make a difference, but they ought to be consistent. I'll flag it as a bug. Thanks, Jonathan R. -- Jonathan D. Rosenberg, Ph.D. 72 Eagle Rock Avenue Chief Scientist First Floor dynamicsoft East Hanover, NJ 07936 [EMAIL PROTECTED] FAX: (973) 952-5050 http://www.jdrosen.net PH: (973) 952-5000 http://www.dynamicsoft.com _______________________________________________ Sip-implementors mailing list [EMAIL PROTECTED] http://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
