[asterisk-dev] [Code Review] SIP: authenticate OPTIONS requests just like we would an INVITE
David Vossel
dvossel at digium.com
Fri Aug 27 18:04:05 CDT 2010
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/881/
-----------------------------------------------------------
(Updated 2010-08-27 18:04:05.034717)
Review request for Asterisk Developers.
Changes
-------
This update addresses OEJ's comment on the dev-list which I have included below. Now there is an option to perform OPTIONS authentication. This option is on by default.
"Now, since you add a lot of extra processing, which people who only use OPTIONS as a "ping" don't want, we should propably have a configuration option for this new behaviour to be backwards compatible. I suggest that option is off by default and your new behaviour is the default.
/O"
Summary
-------
OPTIONS requests should be treated the same as an INVITE... which includes authentication. This patch adds the ability for incoming out of dialog OPTION requests to be authenticated before providing a response indicating whether an extension is available or not. The authentication routine works the exact same way as it does for incoming INVITEs. This means that if a peer has 'insecure=invite' in their peer definition, the same will be true for the processing of the OPTIONS request.
Diffs (updated)
-----
/trunk/channels/chan_sip.c 284033
/trunk/channels/sip/include/sip.h 284033
/trunk/configs/sip.conf.sample 284033
Diff: https://reviewboard.asterisk.org/r/881/diff
Testing
-------
I have tested this through a variety of sipp scenarios in attempt to verify that the OPTIONS request is treated the exact same as an INVITE request in regards to authentication. I also tested this patch with Asterisk back to back verifying that qualify=yes still works as expected. Asterisk's implementation to qualify peers as reachable using OPTIONS request does not care if it gets a"401" or a "200" response.
Thanks,
David
More information about the asterisk-dev
mailing list