Cisco Bug: CSCun44159 - CUCM H.323 state machine unable to keep up with rapid SIP re-INVITEs
Last Modified
Mar 05, 2018
Products (1)
- Cisco Unified Communications Manager (CallManager)
Known Affected Releases
8.6(2.21900.5)
Description (partial)
Symptom: If CUCM receives a rapid series of SIP reINVITEs as follows, then the H.323 state machine is unable to keep up. The result is that the H.323 side is torn down with "Bearer capability not implemented". In the CallManager trace, we see that H245Interface sends 2 'MXInterfaceOpenLogicalChannelIncoming' to the MX which in turn forwards the same to the MTPAgenaInterface. MTPAgenaInterface rejects the second 'MXInterfaceOpenLogicalChannelIncoming' which is forwarded to the H323 side causing the tear down of the call. Later, when the MTPAgenaInterface opens its port and the info is sent to MX, it ignores the same as the call is being torn down. On the SIP side we see a rapid series of SIP messages for the same call, which the H.323 state machine is attempting to keep up with: e.g. 13:56:16.667 Invite(inactive)--------> 13:56:16.726 Invite(DO)-----------> 13:56:16.886 Invite(inactive)---------> 13:56:16.938 Invite(DO)------------> Conditions: The situation where this was experienced was seen as a result of the SIP REFER performed by a MeetingPlace server (after the intial greeting and prompt, when the call is transferred into the meeting). The topology was: PSTN -> E1 -> Voice Gateway -> H323 -> CUCM -> SIP -> CUCM (SME) -> SIP -> MeetingPlace.
Bug details contain sensitive information and therefore require a Cisco.com account to be viewed.
Bug Details Include
- Full Description (including symptoms, conditions and workarounds)
- Status
- Severity
- Known Fixed Releases
- Related Community Discussions
- Number of Related Support Cases