Just revising CVOICE v8 and looking at QOS. The cisco model recommends 33% for RTP media, 5% for RTP signalling and then all the rest.
From my limited reading so far, would this be a valid config? (my router takes the config, but that does'nt mean it will work). Voice and video
sharing the same PQ, but treated differently based on their DSCP value for "dropping".
class-map match-any RTP-VOICE
match protocol rtp audio
class-map match-any RTP-VIDEO
match protocol rtp video
class-map match-any RTP-SIGNALLING
match protocol sip
match protocol h323
match protocol skinny
match protocol mgcp
policy-map QOS-POLICY
class RTP-VOICE
set dscp ef
priority percent 33
class RTP-VIDEO
set dscp af41
priority percent 33
class RTP-SIGNALLING
set dscp cs3
bandwidth percent 5
class class-default
fair-queue
random-detect