Voice Gateways Hot Issues from Cisco TAC

CME 9.1 crashes with TLB (store) exception in CCSIP_SPI_CONTROL, Open CSCug81754

Symptom: Router c2800 with CME9.1 crashes with signal 10 TLB (store) exception in CCSIP_SPI_CONTROL process
Conditions: unknown
Workaround: unknown

Call fails as SIP implementation sets local rtp port to 0, Fixed CSCub06422

Symptom: Call flow: PSTN---pri---Voice Gateway---sip---SIP server After running fine for 6-7 days, calls through voice gateway fail (100% of the calls fail). On a call that comes in through the PRI, INVITE is sent with "m=audio 0..". Then, on getting "200 OK" from the other end, gateway disconnects the call.
Conditions: Router up and running for 6-7 days
Workaround: reload router.

ES Crashes after second 401 Challenge, Fixed CSCue39206

Symptoms: ES crashes after the second 401 challenge.
Conditions: This symptom occurs when the second 401 is received after SDP offer/answer with 183/PRACK is complete. This is a rare scenario.
Workaround: There is no workaround.

Ipad Jabber failed to register to CUBE, Fixed CSCud35416

Symptom: Ipad jabber failed to register to CUBE. Also seen where cube is not responding to the out-of-dialog option pings (TCP) being sent by Microsoft Lync server.
Conditions: Ipad jabber default register with TCP length longer than 536 and cause TCP fragment. 2nd case: Call Flow - Microsoft Lync Server >> sip >> Cube From the packet capture, have seen this TCP flow: LYNC Cube Syn>>>>>> <<<<<>>>>> OPTIONS>> <<<<< Workaround: configure ip tcp adjust 1400 or Downgrade to 15.2(3)T2.

ISDN PRI call failures with Resource unavailable, unspecified, Fixed CSCtz67980

Symptom: Outbound calls across a ISDN PRI on a Cisco 2921 voice gateway will start failing on random b-channels with "resource unavailable, unspecified". No DSP errors are observed. No DSP allocation failures are observed.
Conditions: Observed so far on Cisco IOS versions 151-3T3, 151-4M3 and 152-3T.
Workaround: Reloading the gateway will provide temporary relief.

AS5xxxXM and VGD: boot-time traceback %SNMP-3-DVR_DUP_REGN_ERR observed, CSCtj06522

Symptom: ======== The following traceback is observed at boot-time on the console of a VGD 1T3, AS5350XM, or AS5400XM Voice GateWay (VGW) configured for NFAS PRI: 000003: *Jul 26 05:03:18.839 EST: %SYS-6-CLOCKUPDATE: System clock has been updated from 04:03:18 EST Thu Jul 26 2001 to 05:03:18 EST Thu Jul 26 2001, configured from console by console. 000004: *Jul 26 05:03:18.963 EST: %SNMP-3-DVR_DUP_REGN_ERR: Attempt for dupe regn with SNMP by driver having ifIndex 61 and ifDescr Serial1/0:1:23-Signaling -Traceback= 0x60E15F14z 0x60DE9CF8z 0x601566F0z 0x60159388z 0x6068F0E8z 0x60695804z 0x60698B38z 0x606990A8z 0x608DE718z 0x608D8328z 0x608CFE94z 0x608DB5D4z 0x608DB7D8z 0x608DB860z 0x60646CF4z 0x6065ED40z 000005: *Jul 26 05:03:19.999 EST: %SNMP-3-DVR_DUP_REGN_ERR: Attempt for dupe regn with SNMP by driver having ifIndex 1069 and ifDescr Serial1/0:22:23-Signaling -Traceback= 0x60E15F14z 0x60DE9CF8z 0x601566F0z 0x60159388z 0x6068F0E8z 0x60695804z 0x60698B38z 0x606990A8z 0x608DE718z 0x608D8328z 0x608CFE94z 0x608DB5D4z 0x608DB7D8z 0x608DB860z 0x60646CF4z 0x6065ED40z 000006: *Jul 26 05:03:20.563 EST: %SYS-5-CONFIG_I: Configured from memory by console No SNMP features have been configured on the VGW. The tracebacks are observed each time the VGW is reloaded.
Conditions: =========== This behaviour has been observed on Cisco VGD 1T3, AS5350XM, and AS5400XM Voice GateWays (VGWs) configured for NFAS PRI voice trunks and installed with a version of IOS in the 15.0(1)M or 15.1T family of IOS releases. IOS release train 12.4(24)T is also affected.
Workaround: =========== This issue is not observed in the IOS 12.4(22)T release train. This is a cosmetic issue and does not affect voice services.

Received %MROUTE-4-SNMP_RADIX_ERR msg joining then leaving mcast group, Fixed CSCtz08416

Symptom: After upgrading from 12.4-22.T4 to 15.1-4M3, CU is getting "%mroute-4-snmp_radix_err:failed to add snmp node to snmp radix tree for multicast group (*, 224.0.1.40)" messages.
Conditions: There is no performance nor service impact. Just errors popping up, and CU is concerning of it.
Workaround: N/A

SIP GW does not transmit RTP after conference originator ends call, CSCtz60347

Symptom: One way audio expeerienced by Ip phone user after conference initiator drops off call and leaves PSTN user with other IP phone user
Conditions: IP phone - CUCM 8.0.3 - SIP - ISRG2/15.2.3T - PRI - PSTN
Workaround: Downgrade IOS to 151-3.t

Memory Leak in CCSIP_SPI_CONTROL @ sip_get_sipspi_message, CSCtz56248

Symptom: A Cisco2911 running 15.2(2)T may experience a memory leak in CCSIP_SPI_CONTROL
Conditions: Not fully understood
Workaround: None at this time

DSAPP Call-Waiting fails if translation rule or num-exp is used, Open CSCug92696

Symptom: Using DSAPP (Device control Session APPlication) to provide call-waiting for SIP->FXS calls returns busy signal.
Conditions: When using DSAPP (Device control Session APPlication) to provide call-waiting for SIP->FXS calls, if an inbound voice translation-rule or number expansion (num-exp) is used to translate the called number to a different number (for example, from the 10 digit DID number to an internal 4 digit extension) no call-waiting is provided for the second call and it will disconnect with a busy signal cause.
Workaround: Configure the destination-pattern of the POTS dial-peer for the FXS to be the same called-number that is being received through the SIP leg.

ISDN PRI Layer 1 Deactivated After Reload, Open CSCug36075

Symptom: layer 1 on the ISDN PRI does not come up after a reload.
Conditions: After Reload.
Workaround: A shut/no shut is required to bring back the PRI up.

T.37 - minor clock/frame slips may cause truncated pages in tiff image, Fixed CSCtf61563

Symptom: An IOS voice gateway acting as a T.37 Onramp gateway may generate tiff images with incomplete pages if a clock or frame slip occurred while receiving said pages. Pages preceding and following the page(s) where the clock or frame slip occurred will be complete.
Conditions: This happens with T.37 Onramp and PSTN circuits where clock or frame slips may occur on the local loop or in the PSTN network.
Workaround: None.

SG3 pass through not works with fax-relay sg3-to-g3 command, Open CSCug17385

Symptom: CM tone constant Playout
Conditions: When CM suppression is enable
Workaround: Disable CM supression

fax call failed in pvdm3 due to auto t30 flag spoofing too soon, Fixed CSCuc40252

Symptom: fax call fail
Conditions: the receiving fax machine take longer 2.2 second to respond
Workaround: no

Providing a CLI Knob for handling media loops in Voip RTP, Fixed CSCub45809

Symptoms: Cisco IOS configured for Voice over IP may experience stack corruption due to multiple media loops.
Conditions: This requires a special configuration of IP features along with disabling the recommended media flow-around command. IOS version 15.2(2)T
Workaround: Apply media flow-around command.
Further Problem Description: PSIRT Evaluation: The Cisco PSIRT has assigned this bug the following CVSS version 2 score. The Base and Temporal CVSS scores as of the time of evaluation are 5.4/4.4: https://intellishield.cisco.com/security/alertmanager/cvssCalculator.do?dispatch=1&version=2&vector=AV:N/AC:H/Au:N/C:N/I:N/A:C/E:U/RL:W/RC:C CVE ID CVE-2012-5044 has been assigned to document this issue. Additional information on Cisco's security vulnerability policy can be found at the following URL: http://www.cisco.com/en/US/products/products_security_vulnerability_policy.html

CUBE: Invite not forwarded from leg to leg because of pending 200, Fixed CSCtq07341

Symptom: CUBE may fail to forward buffered INVITE messages if it receives a different INVITE on a call leg where it is waiting for a 200 response. It should send a SIP 491 but does not do so.
Conditions: These was seen on the following topology: PhoneA,PhoneC----PSTN---SIP---CUBE---CUCM---PhoneB The problematic call flow is as follows: 1. PhoneA calls PhoneB, PhoneB answers, two way audio is heard. 2. PhoneB presses Transfer, which places PhoneA on hold. 3. PhoneB calls PhoneC, PhoneC answers, two way audio is heard. 4. PhoneB presses Transfer again, to complete the transfer (attended transfer). The call disappears from PhoneB's screen. 5. PhoneA and PhoneC have no way audio and the call drops approximately 25 seconds after the transfer is completed by BhoneB.
Workaround: None

Reload due to Bad chunk reference count with CCSIP_SPI_CONTROL process, CSCue48385

Symptoms: The router has an unexpected reload at the CCSIP_SPI_CONTROL process.
Conditions: This symptom is observed with Cisco IOS Release 12.4(26)T6.
Workaround: There is no workaround.

cube send 0.0.0.0 when 9971 has video enabled for hold/resume, Fixed CSCtz73157

Symptoms: CUBE sends 0.0.0.0 when 9971 has video enabled for hold/resume/conference from PSTN caller. CUBE sends correct IP address when 9971 has video disabled for hold/resume/conference from PSTN caller.
Conditions: The symptom is observed with the following conditions: - Cisco IOS Release 15.2(2)T1. - Current phone load sip99719.2.4-19. - Current CUCM version: 8.5.1.13900-5. - MCS7825I4-K9-CMD2A. - On the SIP trunk, the box "Retry Video Call as Audio" was checked. For the calls with video disabled, the CUBE is sending the 200OK with the C=IN ipX x.x.x.x address. Sent: SIP/2.0 200 OK Via: SIP/2.0/TCP x.x.x.x:5060;branch=z9hG4bK7322e28fb58f2 From: "name" ;tag=1171271~17954349-bc2a-4081-adb4-34491012bb45-24984725 To: ;tag=D99A474-A1A Date: Tue, 24 Apr 2012 18:26:17 GMT Call-ID: f9e43000-f961f049-61593-a28050a@x.x.x.x CSeq: 101 INVITE Allow: INVITE, OPTIONS, BYE, CANCEL, ACK, PRACK, UPDATE, REFER, SUBSCRIBE, NOTIFY, INFO, REGISTER Allow-Events: telephone-event Remote-Party-ID: ;party=called;screen=no;privacy=off Contact: Supported: replaces Supported: sdp-anat Server: Cisco-SIPGateway/IOS-15.2.2.T1 Supported: timer Content-Type: application/sdp Content-Disposition: session;handling=required Content-Length: 241 v=0 o=CiscoSystemsSIP-GW-UserAgent 9798 5431 IN IPX x.x.x.x s=SIP Call c=IN IPX x.x.x.x t=0 0 m=audio 25014 RTP/AVP 0 101 c=IN IPX x.x.x.x a=rtpmap:0 PCMU/8000 a=rtpmap:101 telephone-event/8000 a=fmtp:101 0-15 a=ptime:20 For the calls with video enabled, the CUBE is not sending the IP address correctly, as seen here: Sent: SIP/2.0 200 OK Via: SIP/2.0/TCP x.x.x.x:5060;branch=z9hG4bK734ab4c88ccb9 From: "name" ;tag=1171897~17954349-bc2a-4081-adb4-34491012bb45-24984949 To: ;tag=DA1D53C-2232 Date: Tue, 24 Apr 2012 18:35:25 GMT Call-ID: 39f7e280-f961f262-616f4-a28050a@x.x.x.x CSeq: 102 INVITE Allow: INVITE, OPTIONS, BYE, CANCEL, ACK, PRACK, UPDATE, REFER, SUBSCRIBE, NOTIFY, INFO, REGISTER Allow-Events: telephone-event Remote-Party-ID: ;party=called;screen=no;privacy=off Contact: Supported: replaces Supported: sdp-anat Server: Cisco-SIPGateway/IOS-15.2.2.T1 Supported: timer Content-Type: application/sdp Content-Length: 278 v=0 o=CiscoSystemsSIP-GW-UserAgent 144 2583 IN IPX x.x.x.x s=SIP Call c=IN IPX 0.0.0.0 t=0 0 m=audio 16654 RTP/AVP 0 101 c=IN IPX 0.0.0.0 a=rtpmap:0 PCMU/8000 a=rtpmap:101 telephone-event/8000 a=fmtp:101 0-15 a=ptime:20 m=video 0 RTP/AVP 126 c=IN IPX 10.5.40.14
Workaround: Disable video from CUCM phone page under the 9971.

CLI to allow MRCPv1 to support RFC 2616, Fixed CSCti50250

Symptom: VXML client is failing to parse Russian external grammer from Nuance Speech Server.
Conditions: Russian external grammer hosted on Nuance Speech Server 5.0.2 [ASR 9.0.3 - Russian Lang Pack; Realspeak 4.5 - Russian Lang Pack]
Workaround: None

Garceful termination for 401 requests after Early dialog establishment, Fixed CSCud99848

Symptom: CUBE crashes while handling 401 after early dialog is established and HA is enabled
Conditions: Crash happens when HA is enabled
Workaround: No work around

Crash at disc_tx_requeue_client, Fixed CSCts33018

Symptoms: The Cisco UC500 crashes at Dot11 subsystem after upgraded from SWP 8.1.0 (15.1(2)T2) to SWP 8.2.0 (15.1(2)T4)
Conditions: This symptom is observed when a Cisco 2900 PoE switch is connected to the Cisco UC540 with Cisco phones and an iMac is connected to the switch. An Apple laptop is also connected using wireless.
Workaround: There is no workaround.

Memory leak in CCSIP_SPI_CONTROL @ url_parseTelUrl, Fixed CSCuc42558

Symptoms: A Cisco router configured as the VXML gateway may experience a leak in the processor memory pool in CCSIP_SPI_CONTROL in the function url_parseTelUrl.
Conditions: This symptom occurs when a Cisco router is configured as the VXML gateway. There could be other triggers as well.
Workaround: Reload the router during a maintenance window to avoid an unexpected crash. You may also downgrade to Cisco IOS Release 15.1(4)M3, which is not affected.

Fax upspeed with SG3 fax machines fail, Open CSCtu80102

Symptom: Sg3 does not work
Conditions: When SG3 fax are enable
Workaround: None

CUBE: SIP to H323 doesn't complete T.38 negotiation w/o inbound OLC, Terminated CSCsy82117

Symptom: CUBE fails to send an OLC to initiate T.38 when operating in a SIP --> H323 call flow. Specifically, on an inbound SIP call, the H.323 will start T.38 negotiation. A H245 requestMode will be received and CUBE will respond with requestModeAck. The logical channels will be closed in both directions with CloseLogicalChannel and CloseLogicalChannelAck messages. However, CUBE will not originate an OLC for T.38 operation until the far end H.323 sends one first.
Conditions: This was seen specifically in this call flow: Fax-A---SIP---PSTN---SIP---CUBE---H323---CUCM---SIP---ATA---Fax-B Call is from Fax-A to Fax-B.
Workaround: If CUCM is interworking with CUBE on the H.323 side, load a version of CUCM with the fix for CSCsy55176. If CUCM is not in use, there is no workaround.

CUBE is not responding to Lync Server Option (TCP), CSCug74337

Symptom: The Cube is not responding to the out-of-dialog option pings (TCP) being sent by Microsoft Lync server.
Conditions: Call Flow - Microsoft Lync Server >> sip >> Cube From the packet capture, have seen this TCP flow: LYNC Cube Syn>>>>>> <<<<<>>>>> OPTIONS>> <<<<< Workaround: Downgrade to 15.2(3)T2

CM tone detector ON irrespective of SG3 fax / modem features disabled, Fixed CSCug44667

Symptoms: CM tone detector being turned ON irrespective of the fax and modem features being disabled
Conditions: CM tone detector being turned ON and being reported to the Host by the DSP.
Workaround: No workaround exist.

Modem calls fails between two MGCP gateways, Open CSCug78753

Symptom: Call Flow: Modem->VG202->MGCP->CUCM->MGCP->Voice gateway (PRI)-> PSTN(modem) Problem Description: Modem calls from VG202 to PSTN modem is failing. (V.8) phase completes but the call is not moving to the next phase i.e. Probing/Ranging.
Conditions: NA
Workaround: NA

No timer started on CUBE in SIP-SIP and both ends support session-timer, Open CSCta68880

Symptom: After successful negotiation of session timer through CUBE in SIP-SIP mode and both the endpoints supporting session-timer, there is no timer started on CUBE to keep track of session-expiration. When call establishes with session-timer, CUBE should start a timer with session-expires value and clear the call. This is an enhancement request for CUBE.
Conditions:
Workaround:

ISDN PRI L2 problem, GW reload while in mgcp fallback mode, Terminated CSCsb53322

Symptom: ISDN PRI L2 doesn't come up after GW reload while in SRST mgcp fallback mode.
Conditions: If gateway is reloaded while in SRST mgcp Fallback mode, ISDN PRI L2 fails to come up.
Workaround: Need manually delete 'mgcp' keyword under controller E1 configuration.

CUBE is not responding to the INVITE message received after 491 Message, Open CSCug45517

Symptom: Topology: ========= < -----(SIP Trunk A)-----CUBE-----(SIP Trunk B)-----> CUBE is not forwarding the REINVITE message received from Trunk A to the SIP Trunk B when 491 Request Pending is received from SIP Trunk B for the previous SIP transaction.
Conditions: When 491 Request Pending is received
Workaround: None

ASR Crashes, Open CSCug61688

Symptom: ASR Crash
Conditions: Crashd with trace back and core dump also generated
Workaround: none

Wireless clients can't connect to UC500 because stop sending beacons, CSCsx05336

Symptom: Clients can't connect to AP.
Conditions: UC500 running wireless. Wireless clients can't connect. UC500 doesn't show any clients trying to connect.
Workaround: None.
Further Problem Description:

Router crashes due to AFW processing tied to Tcl scripts, Terminated CSCuf86086

Symptom: Router configured for voice crashes while processing Tcl script.
Conditions: This may be tied to BACD Tcl scripts based on symptoms from the crash.
Workaround: None.

DTMF distorted by a voice coder may not be detected, Fixed CSCti18743

Symptom: Some DTMF signals are detected intermittently.
Conditions: This problem is most likely to occur when the DTMF signals are sent by a mobile device, which may encode (and therefore distort) the DTMF using a voice coder in the wireless link.
Workaround: There is no workaround.

CUBE creates new TCP connection for OPTIONS ping reply, Terminated CSCub75882

Symptom: RFC3261 specifies below: 18.2.2 Sending Responses The server transport uses the value of the top Via header field in order to determine where to send a response. It MUST follow the following process: 4:43 PM o If the "sent-protocol" is a reliable transport protocol such as TCP or SCTP, or TLS over those, the response MUST be sent using the existing connection to the source of the original request that created the transaction, if that connection is still open. When CUSP sends OPTIONS request to CUBE using TCP, CUBE will use the port in VIA header to establish new TCP connection to CUSP in order to send 200OK. CUSP rejects the connection due to RFC noncompliance, thus marking the CUBE server-group as down. This behavior can be fixed if a global SIP bind interface command is issued. Without global SIP bind (or with only dial peer level bind), CUBE will try to establish new connection rather than reuse open connection. Issue observed on both 15.2.2T and 15.1.3T3
Conditions: Without global SIP bind (or with only dial peer level bind), CUBE will try to establish new connection rather than reuse open connection.
Workaround: This behavior can be fixed if a global SIP bind interface command is issued

SIP source binding CLI removed after serial subinterface flaps, Open CSCth63196

Symptom: The sip source interface binding commands disappear after being configured and functional.
Conditions: The T1 subinterface which is bound flaps with an active call present on it.
Workaround: Reapply the CLI manually.

Router crashes due to missing memory associated with SIP, Fixed CSCtk55107

Symptoms: A router crashes due to SIP.
Conditions: This symptom is observed when SIP is configured.
Workaround: The only workaround is to disable SIP.

SIP crashing - bad block pointer errors, Terminated CSCub16808

Symptom: A router is having multiple unexpected reloads
Conditions: The router is working as a voice gateway with SIP protocol
Workaround: none

Traceback seen on 3x and 2x-platform for basic video call, Terminated CSCti58933

Symptom: Traceback is seen for video call when used sipp endpoints played the media
Conditions: Traceback is seen for video call when used sipp endpoints played the media.
Workaround: The problem is not seen in other platform.

Router crashed due to watchdog timeout on ISDN Process, Open CSCug87650

Symptom: Router crashed due to ISDN Process SYS-3-CPUHOG and SYS-2-WATCHDOG
Conditions: ISDN
Workaround: None.

ASR cvCallVolConnActiveConnection SIP not accurate, CSCua68587

Symptoms: cvCallVolConnActiveConnection.sip MIB count does not match what is seen on the CLI.
Conditions: This symptom is observed with the Cisco ASR 1006 running Cisco IOS XE Release 3.6.0S or Cisco IOS Release 15.2(2)S with the asr1000rp2-adventerprisek9.03.06.00.S.152-2.S image.
Workaround: There is no workaround.

Backhaul fails with MGCP GW as Layer 2 stays down, Terminated CSCtj96629

Symptom:2951 MGCP T1/pri version: 15.0(1)M3 This only happens with a newly added 2951 GW. All other MGCP GWs are working fine(38xx series). backhaul as L2 goes down to TEI assigned. CUCM: 6.1.5.11900-13 GW: 2951 MGCP T1/pri version: 15.0(1)M3 This only happens with a newly added 2951 GW. All other MGCP GWs are working fine(38xx series). backhaul as L2 goes down to TEI assigned. CUCM: 6.1.5.11900-13 GW: 2951 MGCP T1/pri version: 15.0(1)M3 This only happens with a newly added 2951 GW. All other MGCP GWs are working fine(38xx series). backhaul as L2 goes down to TEI assigned.
Workaround: Configuring H323 gets it to work. Even an FXO port registers as it does not need backhaul

CUBE with transcoding causes malformed RTCP packets, Fixed CSCua42104

Symptoms: Malformed RTCP packets are observed.
Conditions: This symptom occurs when DTMF interworking is enabled or SRTP/SRTCP is in use.
Workaround: Disable DTMF interworking if not required for the call.

2921 Router crashed after receiving 486 Busy, Fixed CSCue34694

Symptom: 2921 Router crashed after receiving 486 Busy
Conditions: Crashed observed when handling 486 Busy response
Workaround: No workaround

Cannot insert call history entry for SIP call, Fixed CSCtl55445

Symptom: CUBE logging the follwoing message: %SIP-3-INTERNAL: Cannot insert call history entry for callID
Conditions: Calling party cancels call before connection: INVITE --------------->---------------> 100 Trying <--------------<---------------- 180 Ringing <--------------<---------------- CANCEL ---------------->---------------> 200 OK <----------------<----------------- 487 Request Cancelled <------------------<--------------- ACK -------------------->--------------->
Workaround:

Memory Leak in CCSIP_SPI_CONTROL @ cc_api_get_call_active_entry, Fixed CSCtt26208

Symptoms: A Cisco 3845 running Cisco IOS Release 15.1(4)M1 may have a processor pool memory leak in CCSIP_SPI_CONTROL.
Conditions: Not known at this time.
Workaround: There is no workaround.

DSP Crash in 28.3.9, Fixed CSCue75966

Symptom: DSP Crash in 28.3.9
Conditions: AS5400XM IOS upgraded to 15.1(4)M5.20, after which the DSP Crashed once.
Workaround: None

ASR1K CUBE crashed with segmentation error at process=CCSIP_SPI_CONTROL, Fixed CSCug77212

Symptom: ASR1K CUBE RP may crash with Segmentation fault(11), Process = CCSIP_SPI_CONTROL when sip headers are manipulated using a sip profile for 200 response messages for KPML notify.
Conditions: Crash seems to be happening due to SIP profiles configs being wrongly applied to Notify response (this profile was meant for 200 OK Invite response).
Workaround: Do not configure sip profiles to manipulate the headers for 200 responses. More Info:

SIP NOTIFY with "480 Temporarily Not Available", Open CSCug91548

Symptom: IOS does not send "NOTIFY message with 480 Temporarily Not Available" to CVP.
Conditions: SIP NOTIFY with "480 Temporarily Not Available"
Workaround: No workaround.

DSP: DSP stucked in "BAD" state after a crash, not recovering properly, Fixed CSCtq71158

Symptom: DSP stucked in BAD state after a crash
Conditions: Running ios 124-22T5, dspware 23.8.6
Workaround: downgrade to dspware 9.4.9

Transcoder MTP changes sequence from different sources with same SSRC, Fixed CSCtz52795

Symptom: Telco ---- PRI --- VXML GW ---- CUCM/Transcoder-MTP --- Agent 1 ----Conference - CTI RP into Supervisor Q ---- Agent 2
Conditions: 1. External caller dials into main number and gets to agent 1 and request for Agent 2 2. Agent 1 dial cti route-point 3. Agent 1 is waiting in the supervisor Q waiting for Agent 2 4. Agent 1 is hearing the VXML prompt , then hears the ringback , then Agent 2 answers the call 5. Agent 1 cannot hear Agent 2 6. Agent 2 can hear Agent 1
Workaround: Use Software MTP instead of Transcoder MTP between Agent 1 and VXML/Agent 2

E1 E&M layer 1 down after upgrade to 151-4.M2 or higher, Fixed CSCud22148

Symptoms: The E1 (E&M) controller is down.
Conditions: This symptom is observed with Cisco IOS Release 15.1(4)M2 or later releases. This issue is seen with the Cisco 3945.
Workaround: There is no workaround.

TDM calls experience clicking noise with ecan v1.5.11 on PVDM3s, Fixed CSCud10219

Symptom: -Phone users experience static noise on PSTN calls. -PCM Capture shows clean SIN/RIN but the SOUT will have bad audio
Conditions: -Only affects PVDM3s (SP2600 DSPs) -Any dspware with the ecan library v1.5.11 will experience this issue. -To find the ecan library version issue the following command when there is an active call 'show call active voice echo-canceller port 0/0/0:23.1' (where channel 1 has an active call) Affected DSPWare ---------------- 30.3.2, 30.3.3 31.1.1, 31.1.2 32.1.0, 32.1.1 33.1.0
Workaround: Use a dspware which doesn't have ecan library v1.5.11

Middle buffer leak at skinny_service_moh_multicast process, Fixed CSCtx27284

Symptom: MIddle buffer leak with Multicast MOH packets
Conditions: Multicast MOH configured under ccm-call-manager. The multicast route needs to be on a SVI which is in the Down/Down state.
Workaround: 1)Remove multicast MOH configuration 2)Remove the interface that leaks the buffer from the multicast route list 3)Administratively shut the Vlan interface if it is not being used

Memory leak in CCSIP_SPI_CONTROL process, CSCud35394

Symptom: Malloc Failures in CCSIP_SPI_CONTROL process
Conditions: none
Workaround: none

MGCPAPP fails to send caller-ID from MGCP controlled FXO to CUCM, Terminated CSCtx98141

Symptom: Shut/No shut of MGCP controlled FXO port is required for the MGCP App to send caller-ID information to Call Manager.
Conditions:
Workaround: perform a shut/no shut on MGCP controlled voice-port

Memory Leak - "DSMP DSP REQ c" process, Fixed CSCtl51125

Symptom: Memory leak with Chunk Manager in "DSMP DSP REQ c" process
Conditions: Voice feature configured
Workaround: No known workaround at this time

CUBE - Cannot remove bind from HSRP interface once configured., Terminated CSCua91662

Symptom: Once we bind the SIP process to HSRP Interface (SIP gets bound to HSRP Virtual IP), we cannot bind SIP to any other interface. For example: Dial-peer voice 2911 Voice-class sip bind control source int fa0/0 Then Dial-peer voice 2911 Voice-class sip bind control source int lo0 SIP will still be bound to Virtual IP in fa0/0 and not the lo0.
Conditions: HSRP is configured. SIP is bound at dial-peer or global level to HSRP interface
Workaround: Reload the router to clear binding.

DSP Watchdog Timeout crash is seen in 26.8.4 dspware, Fixed CSCty98815

Symptom: Dsp Crash
Conditions: Dsp crashes maybe seem during transcoding sessions when High Complexity is configured
Workaround: none

2800 router running 12.4(20)T4 crashes at cdb_enqueue, Terminated CSCtl22077

Symptom: 2800 router running 12.4(20)T4 crashes at cdb_enqueue
Conditions: 2800 is used as a voice gateway and crashes with CPU hogs
Workaround: None

Mem Leak in CCSIP_SPI_CONTRO process, Fixed CSCua07791

Symptoms: A Cisco ISR G2 running Cisco IOS Release 15.2(2)T or later shows a memory leak in the CCSIP_SPI_CONTRO process.
Conditions: The leak is apparent after 3-4 weeks. The process is CCSIP_SPI_CONTRO.
Workaround: There is no workaround.

Voice gateway reports CPU hogs in 'gk process', Terminated CSCtq85757

Symptom: Logs full of CPU hog messages related to gatekeeper process: %SYS-3-CPUHOG: Task is running for (2004)msecs, more than (2000)msecs (3/2),process = gk process
Conditions: This has been observed on a C3845 running 12.4(24)T4.
Workaround: Workaround not known at the moment.

HW MTP not passing rfc2833 DTMF, Fixed CSCtq98385

Symptom: ++ When HW MTP is engaged, and instructed to pass the dtmf digits in rfc2833 tpe, DSP is not processing it. It gets intiated in express mode ( hpi_universal_transcoding_mode: repacketization and dsp decide mode (6)) which didn't support peer to peer messaging.
Conditions: HW MTP is engage using LSI dsp
Workaround: disable Hardware MTP

Router crash triggered by SIP timer expiry, CSCue66002

Symptoms: The router crashes due to a SIP timer expiry
Conditions: This symptom is triggered by a SIP generic timer expiry.
Workaround: There is no known workaround.

TLB (store) exception crash at cch323_h245_mr_notify, Fixed CSCtr29338

Symptoms: A router crashes.
Conditions: The symptom is observed after an %ISDN-6-DISCONNECT message from "unknown" followed by a couple of "Illegal Access to Low Address" messages.
Workaround: There is no workaround.

CUBE crash trying to save previous SDP context, Fixed CSCte86240

Symptom: If a CUBE receives a new early dialog offer, with a NULL SDP, when trying to save a copy of the previous SDP context, this can lead to a race condition, which can crash the CUBE router.
Conditions: Race condition, when the next early dialog offer is received before the CUBE has handled the previous SDP.
Workaround: use a SIP profile to prevent this early dialog offer race condition.

DSPs are getting hung when receiving an incoming Video call, Terminated CSCug63959

Symptom: DSPs are getting hung when receiving an incoming Video call
Conditions: When making the incoming video call On the AS5400XM gateway, the DSP's channels are not freed up after the call is disconnected. Because of this issue, if there is any incoming call (normal audio call) the calls fail with resource unavailable. We need to reboot the router to clear the DSPs.
Workaround: reload the router

IPIP Gateway crash ccsip_bridge, Fixed CSCsj81015

Symptoms: Cisco Multiservice IP-to-IP Gateway (IPIPGW) crashes during a stress scenario.
Conditions: This symptom occurs in a stress scenario with 100 SIP-H323 calls + 150 SIP-H323 DTMF interworking (rtp-nte to h245-alpha) calls.
Workaround: There is no workaround.

Incorrect 'Input rate' interface counters for rtp traffic, Fixed CSCtz48125

Symptom: Topology: G729 LAN A<->G729 WAN<->G729 LAN B or G711 LAN A<->G711 WAN<->G711 LAN B 1) in case of non-multilink (ethernet or serial) , the 'bits/sec' rate of 'show interface' output shows incorrect, lower value than the actual corresponding value in bytes/sec or output rate on same interface 2) in case of multilink, sum of the two input rate under the serial interfaces in multilink is more than the input rate on multilink interface itself, i.e 'show interface summary' shows less input rate on multilink interface which ideally should be sum of the two serial interface in multilink, however output rate is correct
Conditions: Issue only observed with RTP traffic and is not seen with non-voice/Data traffic
Workaround: None

CUBE H323-SIP sends OLC before it received MSDAck during transfer, Open CSCug78933

Symptom: CUBE in H323-SIP scenarion during call hold/transfer may send OLC before it receives MSDAck
Conditions: - can happen in any call flow where some network/processing delays might occur
Workaround: