Auto negotiation of codec in AS5350

My gateway (AS5350) can not do autonegotation of codec. Can anyone tell me how to do. I'm using c5350-is-mz.122-15.T14.bin in my gateway. I have config:
voice class codec 1
codec preference 1 g729r8
codec preference 2 g729br8
codec preference 3 g723r63
codec preference 4 g723r53
But I can not auto negotiate codec and always use g729.

Within Cisco's best practices for the large L3 switches (http://www.cisco.com/en/US/products/hw/switches/ps700/products_white_paper09186a00801b49a4.shtml), they recommend for 100 Mbps "As a rule, first use autonegotiation for any type of link partner." but also recommend if there's a problem configure for 100/full. For gig, they have "Enable Gigabit negotiation on all switch-to-switch links and, generally, on all GE devices."
My experience has been with current gen equipment, there's often less issues with auto working correctly vs. someone forgeting to properly configure both sides of the connection expecially when for some reason they connect to a different port (like when dealing with a hardware failure).

Similar Messages

  • Is it possible to disable auto-negotiation?

    Is there a way to fix the Ethernet link speed on a cFP-2220 (i.e. turn off auto-negotiation)?
    MTO

    I see I am not the only one that wonders about this...the wording made me overlook the other post though:
    http://forums.ni.com/t5/FieldPoint-Family/Unable-to-change-Preferred-Media-Mode-on-cFP-2220/td-p/165...
    MTO

  • Mac Pro not auto-negotiating link speed correctly

    We have found that none of our Mac Pros are auto-negotiating their network link speed correctly while running 10.4.10. This happens while booted into the production OS as well as netbooted to a 10.4.10 NetBoot image. The network card can be forced to negotiate at higher speeds either through the command-line or System Preferences but it will not do it on its own. We have gigabit to every desktop and imaging at the default 10 Mbit is horrible! Any help would be appreciated!

    There is a kext that is the culprit:
    /System/Library/Extensions/IONetworkingFamily.kext/Contents/PlugIns/AppleIntel82 54XEthernet.kext
    In 10.4.9 it's version is 1.1.7a3, in 10.4.10 it is 1.1.8a2, in 10.4.11 it is 1.1.8a6
    That doesn't matter much really, but that's the details.
    1. Get the 10.4.9 Intel Combo Updater
    2. Get OpenUp or Pacifist
    3. Open the DMG
    4. Show contents on the PKG
    5. Copy out the pax.gz file to the desktop
    6. un-Gzip the pax, and open that up (with Pacifist or OpenUp)
    7. Get the old kext
    8. Copy it into your /System/Library/Extentions
    Now, the permission on the kext need to be changed to root:wheel or if you in Finder system:wheel
    Also, you should either delete the kernel caches or:
    sudo touch /System/Library/Extensions
    this changes the date stamp on the directory and OS X will recreate the kernel caches...
    And yes, the 10.4.11 seed update corrects this.
    Also bonus Log files! These have been "prettied up" for readability, but you'll see, 10.4.10's driver is just spitting out Zer0s so the network switch/hub has no choice but to say "O K... I ' l l s p e a k s l o w e r t h e n..."
    10.4.9:
    Auto-Neg Advertise Reg (04d) = 0xde1,
    Link Partner Ability Reg (05d) = 0xc5e1,
    Gig Advertise Reg (09d) = 0xe00,
    Gig Link Partner Ability Reg (10d) = 0x7800
    10.4.10:
    Auto-Neg Advertise Reg (04d) = 0xde1,
    Link Partner Ability Reg (05d) = 0x0,
    Gig Advertise Reg (09d) = 0xe00,
    Gig Link Partner Ability Reg (10d) = 0x0
    (Much Prettier)
    10.4.11(seed):
    Auto-Negotiation Advertisement Register (04d) = 0xde1
    Auto-Negotiation Link Partner Ability Register (05d) = 0xc5e1
    Auto-Negotiation Gigabit Advertisement Register (09d) = 0xe00
    Auto-Negotiation Gigabit Link Partner Ability Register (10d) = 0x7800
    PHY Specific Status Register (17d) = 0xaf48
    If this info from 10.4.11 is NDA well bite me, I think people should know it'll be fixed in the future, right?

  • WRT320N and Auto Negotiation at 1Gig

    I upgraded my home network with a WRT320N router.  The cable modem, SB6120 should connect at 1 Gig on the LAN side.  For about 3 weeks it showed a blue LED indicating 1 Gig.  It went to orange the other day indicating 10/100 Meg.  I connected my PC directly to the modem & the LED came on blue saying it correctly auto negotiated the speed. When I reconnected the router it also now showed blue.  Today it is back to orange & again, connecting the PC gives blue, but when the router is connected it stays orange.  This leads me to believe the router's switch is not correctly performing auto negotiation.  I look at all the info in the router and in the Linksys EasyLink Advisor and see no way to determine the speed the LAN is connecting at.  I see nothing in the cable modem details that indicates that either.  Is there anywhere this can be checked, or some third party software I need to verify the LAN speed?  This has me concerned that it changed for no reason at all.
    Thanks
    Message Edited by TonyMan on 07-08-2009 09:25 PM

    I agree with control panel showing the  PC being at 1 Gig, but that is the PC to a LAN switch port on the router.  A speed test will not show what I am looking for.  I want to know if the port on the router labeled Internet and the Ethernet port on the modem are treally connected at 1 Gig.  The router has no way of showing this & the modem LED color indicates it.  When the modem is conneted to the 1 Gig port on the router and the modem says 10/100, then I want to know why.  I would have thought that both devices would have a way to check this via software and allobe able to turn auto negotiation off and let the user set the speed at 1 Gig.
    BTW, I changed to a new cable between the devices and it's stayed at 1 Gig for over a day.  Both were CAT 5e, but maybe that was the issue.  I'll just keep watching it.

  • NEW THREAD: 10G SFP+ Auto-Negotiating to 1G?

    Can a 10G SFP+ SR optic plugged into switch "A" AUTO-NEGOTIATE to 1G to support a 1G SFP SR optic plugged into switch "B" at the other end of the fiber cable?
    In short, what I am asking is whether a 10G SFP+ optic (SR/LR - doesn't matter) can be connected with fiber (of course) to a 1G optic and negotiate DOWN?
    If not, how do I connect that 1G sfp+ optic to a 10G sfp+ switch port?
    NOTE: I know Cisco SFP+ optics only operate at 10G and Cisco SFP optics operate only at 1G. But my question is a general question about SFP/SFP+. From my understanding, SFP+ CANNOT (regardless of vendor) negotiate down to 1G because the SFP+ standard defines a speed range between 4.25 Gbps to 16Gbps. Can someone verify?

    Here is one common Optic SFP that does do both 1G and 10G:
    Intel E10GSFPSR and
    Intel E10GSFPLR
    http://www.intel.com/content/dam/doc/product-brief/ethernet-sfp-optics-brief.pdf
    http://www.intel.com/support/network/adapter/pro100/sb/CS-030612.htm (expand section 1)
    I know for certain that they do support both 1G and 10G as for quite a while I ran these SFPs with Cisco GLC-SX-MMD, and eventually (now) Cisco SFP-10G-SRs.  No config changes or modifications were required to support this functionality.
    This was in a UCS server with a supported Cisco UCS Intel X520 card.

  • 3750 GLC-T SFP Auto Negotiation

    hi all
    I have a 3750 switch that have fiber sfps. I'm planning to buy GLC-t SFPs. I want to know that whether i will be able to connect GLC-T SFP to a normal 100BaseTx port ? Will GLC-T support autonegotiation ?
    Waiting for a response

    If you use a built-in port (e.g. a 10/100 or 10/100/1000 port) then it will negotiate 100Mb.
    It's just the SFPs that won't negotiate lower..
    Regards
    Aaron

  • Codec negotiation in cucm !

    Hi all.
    I have registered 2 CIPC with cucm 7. Now when i make call and press "i" twice, the codec i see is G.729. In my region i have selected call to be made using G.711. I know thats the highest bit rate actually. So is this due to codec negotiation ? can some one give me any highlights as to how codec negotiation works in cucm ? also if there is any cisco doc ?
    thanks

    Hi Everyone
    Thanks for a wonderful explanation from Aaron and Jaimie ,though has a bit doubt of the same .
    My scenario is hq and sb are 2 regions and
    hq<--->hq -------->G711
    hq<---->sb--------->g729
    sb<---->sb---------->g711
    and i put them in corresponding device pool .
    i have a phone in hq and 1 phone in sb .My sip trunk is in hq device pool.
    I have a pstn gateway simulated in which a dial peer is configued to accept incoming sip calls from the trunk.
    dial-peer voice 1 voip
    incoming-called-number .
    codec g711ulaw
    now when i call from hq phone call works and 2 ends negotiates g711 codec.
    when calling from sb phone call is terminated as codec negotiation fails
    then changed scenario of pstn gateway.
    dial-peer voice 1 voip
    incoming-called-number .
    codec g729r8   //default codec
    now calls from both hq and sb phone succeeds with negotiating  g729 codec.
    Now my doubt is why call from hq to pstn phone succeds negotiating g729 codec because
    hqphone to siptrunk_hq ---->g711 and pstn phone is specific to use of g729 codec ? why this call has not failed ?
    and comapring the fact other call failed
    ie sbphone to siptrunk_hq ----->g729 and pstn phone is specific to use g711 codec.
    So who negotiates here ? is the source side or is the destination side ?
    Any documents related to same will also help .
    Thanks,
    Harish

  • Speed negotiation settings for a 10G interface

    HI,
    We have the following setup:
    2 ports of CISCO 6509(also has xfp long range)  switch connected to 2(XAUI) ports of SPARC box T5220.
    When the SPARC box goes for a reboot, the link is UP at XAUI while ports are down at CISCO end.
    However either of the following brings the link up.
    . shut/no shut at 6509 or
    . unplug and replug the cable.
    No speed negotiation is configured at XAUI.
    I would like to know if 10G CISCO 6509  by default has autonegotiation on?
    From the document "Configuring and Troubleshooting Ethernet 10/100/1000Mb Half/Full Duplex Auto-Negotiation"
    I found the following line
    "By default, all devices are supposed to perform autonegotiation. 802.3z       does not specifically define a way to turn Autonegotiation off, for both       1GigabitEthernet and 10GigabitEthernet."
    Im seeing the following situation:
    "Assume that there are two devices, A and B. Assume that each device can       have Autonegotiation enabled, or disabled. The correct behaviour of link status       with autonegotiation in accordance to the IEEE Std 802.3z-1998 should be like       this:
    If A is disabled and B is enabled, then A should report link up and B           should report link down."
    Does it apply to 10 G inetrfaces also?
    PFB the config output of CISCO 6509
    Cisco IOS Software, s72033_rp Software (s72033_rp-IPBASEK9-M), Version 12.2(33)SXI5, RELEASE SOFTWARE (fc2)
    cisco WS-C6509-E (R7000) processor (revision 1.5) with 983008K/65536K bytes of memory.
    6  16  CEF720 16 port 10GE WS-X6716-10GE
    6  Distributed Forwarding Card WS-F6700-DFC3C
    interface TenGigabitEthernet1/6/16
    description BKUP:SSCPRODENG-NXGE4
    switchport
    switchport access vlan 101
    switchport mode access
    spanning-tree portfast edge
    end
    NOLACORE#sh int te1/6/16
    TenGigabitEthernet1/6/16 is up, line protocol is up (connected)
      Hardware is C6k 10000Mb 802.3, address is f866.f220.5027 (bia f866.f220.5027)
      Description: BKUP:SSCPRODENG-NXGE4
      MTU 1500 bytes, BW 10000000 Kbit, DLY 10 usec,
         reliability 255/255, txload 1/255, rxload 1/255
      Encapsulation ARPA, loopback not set
      Keepalive set (10 sec)
      Full-duplex, 10Gb/s, media type is 10Gbase-LRM
      input flow-control is on, output flow-control is off
      ARP type: ARPA, ARP Timeout 04:00:00
      Last input never, output 00:00:19, output hang never
      Last clearing of "show interface" counters 7w4d
      Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0
      Queueing strategy: fifo
      Output queue: 0/40 (size/max)
      5 minute input rate 0 bits/sec, 0 packets/sec
      5 minute output rate 81000 bits/sec, 7 packets/sec
         207547040601 packets input, 34173665463903 bytes, 0 no buffer
         Received 64391 broadcasts (0 multicasts)
         0 runts, 0 giants, 0 throttles
         36 input errors, 36 CRC, 66 frame, 0 overrun, 0 ignored
         0 watchdog, 0 multicast, 0 pause input
         0 input packets with dribble condition detected
         380437679022 packets output, 276421252994669 bytes, 0 underruns
         0 output errors, 0 collisions, 1 interface resets
         0 babbles, 0 late collision, 0 deferred
         0 lost carrier, 0 no carrier, 0 PAUSE output
         0 output buffer failures, 0 output buffers swapped out
    From the above config output can we say that 10Gb/s speed is forced on the link?
    Regards,
    Tejaswi A.

    Justin,
    Thanks very much for the tip!
    Do you know if Oracle 10g has any kind of built-in tool for editing data? I have heard of "The Vault", but that appears to be a tool for administering database users, reports, etc. I am looking for some kind of tool that shows the data in a grid layout (like Microsoft Excel) and allows someone to edit and save data in that grid. A web-based tool would be best, but a desktop tool (especially if it's already part of Oracle 10g) would also be good.
    Thanks again,
    Rachel

  • Missing codec in Media Player to play LabVIEW training video over Web

    I get a "missing codec 130" message when I try to run LV training video from NI website. The error link takes me to MS site but codec is not available for download (apparently). WM player is configured for auto download of codecs cant find it. I have re-downloaded WM player twice and also run windows update.
    Any ideas what to do next?

    Hi mg,
    After googling around, it appears as though "codec 130" is referring to an ACELP Sipro codec.
    If you enable codec download in WMP, the codec should be installed automatically when you play the video. However, if it does not, then it's possible that the codec may already be there and there may be an issue with the codec already installed.
    http://www.nwlink.com/~zachd/pss/pss.html#wmv is a link about installing/downloading this codec.
    I'm not sure if you've tried this yet, but is it possible to uninstall WMP first and then reinstall? Also, I would maybe look into finding ways to uninstall/unregister codecs to see if we can reload the codec.
    Jeremy L.
    National Instruments
    Jeremy L.
    National Instruments

  • Lan speed and duplex negotiation

    hello
    when you have speed and duplex set manualy would this stop the negotiation with the other site ?

    thank you for the reply
    i had the folowing problem which is solved
    router with manual setting speed 100, duplex full
    conected to an uknown switch witth increasing input errors.
    By changing all setting to auto,
    speed and auto duplex got the same settings 100 full duplex but errors disapeared
    My explanation to this (I have no visbility to the switch )by schanging settings to auto, negotiation started and switch changed the setting to 100 full from probably default 100 half.
    and from here comes my question would be by setting the ethernet to manual will disable nogatiations ?

  • Gigabitethernet negotiation problem with cisco 2960

    Hi
    I have 2 cisco 2960 switches. one is a WS-C2960-48TT-L Version 12.2(25r)SEE1 and the other WS-C2960-48TC-L Version 12.2(25r)FX.
    When i build a trunk between the gigabit ethernet ports, WS-C2960-48TC-L end shows me half duplex with autonegotiation whereas it is hard coded on the WS-C2960-48TT-L.
    Iam not able to set the speed and duplex for the port on the 2nd switch.
    Is this a bug with the 2960 or a compatibility feature between the 2 2960's
    Narayan

    Hello Narayan,
    Has the speed and dubplex settings on the 2960 been configured? In that case auto-negotiation frames are not sent and the switchport cannot determine the speed/duplex from it's link partner. It will automatically choose 1000 half-duplex.
    Refer to the following document: http://www.cisco.com/warp/public/473/3.html
    HTH
    --Leon
    * Please rate posts.

  • IPS 4240 & Interface Up\Down In Bypass-Mode Auto

    Hi. this is a strange one. We have a C7200R (FastEthernet) on one side and a C3500 (FastEthernet) on the other with an IPS4240 in the middle. When changing the IPS unit from "bypass-mode on" to auto the interface on the C7200 router goes down, ie no link activity. We have tried several combinations of interface speeds\duplex. The systems would normally be in speed auto\duplex auto but we have tried 100\full forced as well. When in "bypass-mode on" the all systems work fine in auto\auto negotiating 100\full. Any ideas. Thanks Alex

    I have a similar problem:
    ASA 5510 - E0/0 - Connects to E0/1 on C2800 - when we set it to FULL/100 the connection fails - when the interfaces are set to AUTO everything is fine. Any suggestions? Thanks

  • Why AS5350 won,t support G729A ???

    Need to test Different CODECs between AS5350 and Other vendors GW. I need to install G729A , This would normally be g729ar8, but is missing from my AS5350. Any ideas ?????
    Cisco Internetwork Operating System Software
    IOS (tm) 5350 Software (C5350-IS-M), Version 12.3(5), RELEASE SOFTWARE (fc1)
    cnibhco112(config)#voic cl codec 444
    cnibhco112(config-class)#cod pr 3 ?
    clear-channel Clear Channel 64000 bps
    g711alaw G.711 A Law 64000 bps
    g711ulaw G.711 u Law 64000 bps
    g723ar53 G.723.1 ANNEX-A 5300 bps
    g723ar63 G.723.1 ANNEX-A 6300 bps
    g723r53 G.723.1 5300 bps
    g723r63 G.723.1 6300 bps
    g726r16 G.726 16000 bps
    g726r24 G.726 24000 bps
    g726r32 G.726 32000 bps
    g729br8 G.729 ANNEX-B 8000 bps
    g729r8 G.729 8000 bps
    gsmfr GSMFR 13200 bps
    cnibhco112(config-class)#cod pr 3

    Teo,
    Have read this link previously,
    http://www.cisco.com/warp/public/788/voip/codec_complexity.html
    But as you can see by my config , I have no option for VOICE-CARD , only VOICE-PORT, and this does not allow CODEC COMPLEXITY command.
    cnibhco112(config)#voice-?
    voice-port
    cnibhco112(config)#voice-p
    cnibhco112(config)#voice-port 3/0:d
    cnibhco112(config-voiceport)#?
    Voice-port configuration commands:
    bearer-cap Specify the bear capability
    comfort-noise Use fill-silence option
    compand-type The companding type for this voice port
    connection Specify Trunking Parameters
    cptone Configure voice call progress tone locale
    default Set a command to its defaults
    description Description of what this port is connected to
    disc_pi_off close voice path when disconnect with PI received
    echo-cancel Echo-cancellation option
    exit Exit from voice-port configuration mode
    input Configure input gain for voice
    music-threshold Threshold for Music on Hold
    no Negate a command or set its defaults
    non-linear Use non-linear processing during echo cancellation
    output Configure output attenuation for voice
    playout-delay Configure voice playout delay buffer
    threshold Threshold [noise] for voice port
    timeouts Configure voice timeout parameters
    translate Translation rule
    translation-profile Translation profile
    trunk-group Configure interface to be in a trunk group
    voice-class Set voiceport voice class control parameters
    voicecap Apply voicecap
    cnibhco112(config-voiceport)#

  • Auto-nego or 100-Full

    We have a combination of Dell and Sun servers
    at my site and a fierce debate is ongoing as to whether all server switch ports should be set at auto or 100/full. My own preference is auto both ends and fix if that doesn't work.
    Does anyone else have an opinon about what's "best" or why they have chosen one method over another.

    I realize that you've already made up your mind but let me give you a practical prespective. I manage a network of roughly 17,000 nodes of which roughtly 2000 of them are in a data center (i.e. servers) and the rest are client devices such as desktops, some servers, and various medical equipment. The desktop environment is exclusively auto-negotiated while the server environment was hard-coded on both ends to 100/FULL with some exceptions. We've had more trouble worrying about duplex/speed settings on the 2000 machines while the auto-negotiated client networks have had almost unheard of problems due to negotiation. The problem isn't whether or not setting it one way or another is bad. Like others said, so long as both ends are the same then you generally shouldn't have any problem. The problem stems from people moving devices around or reconfiguring their devices and forgetting to set the port to 100/FULL. On a 2000 node data center environment this happens quite frequenty. The problem is procedural in nature and not really technical. We've recently switch to auto-negotiated across the entire network and we're already reaping the rewards. SysAdmin basically cut one process (i.e. hard-coding the ports) completely out of their procedures which also saved a lot of time in troubleshooting "network problems" if you catch my drift.
    So I'm only saying this to advise you to concentrate on process/procedures for how things get provisioned going forward in the DataCenter as this will be the biggest headache for you and having a good process in place will help you.

  • 4250XL auto negotiate or fullduplex

    Does anyone out there using 4250XL on:
    1. span port of switch configured as 10/100/1000 fullduplex?
    2. Toplayer taps?
    We are unable to get the sniffing port going as it does not recognize the tap nor the span port. Upon speaking with Cisco, we were advised that the span port has to be configured as "AUTO NEGOTIATE" for any Cisco IDS to work. We currently have a 4235 running without any problem.
    We were advised against making the config change on the sensor as it is not supported.
    A smart IDS should be able to recognize whatever mode on the switch, don't you agree?
    Are we missing something? Can anyone clarify please.
    Thanks,
    Simone.

    Does anyone out there using 4250XL on:
    1. span port of switch configured as 10/100/1000 fullduplex?
    MC> The sensor does not support configuration of the speed/duplex on the sensor or switch. It should be left auto negotiate for both speed and duplex.
    When hardcoding speed/duplex, you must either hard code both devices to the same speed/duplex, or hardcode one device to the default speed/duplex of the second device.
    In the case of the IDS-4250-XL running version 4.1 software, there is no support for speed/duplex setting of the interface so it is best to leave the switch in auto negotiation as well.
    (NOTE: In version 5.0 support is being added so the sensor interfaces may have their speed/duplex configured by the user. With 5.0 the command and control could be set to 10 or 100 or 1000 for the speed and half or full for the duplex. With 5.0 the sniffing interfaces of the IDS-4250-XL can be configured to 1000 Full duplex. 10/100 speeds and half duplex settings will not be supported on the Gig fiber sniffing interfaces.)
    2. Toplayer taps?
    MC> The IDS-4250-XL does not currently work with Fiber TAPs. The Fiber taps do not provide a valid link to the fiber sniffing ports of the IDS-4250-XL so the sniffing ports are not brought up for monitoring.
    We are unable to get the sniffing port going as it does not recognize the tap nor the span port. Upon speaking with Cisco, we were advised that the span port has to be configured as "AUTO NEGOTIATE" for any Cisco IDS to work. We currently have a 4235 running without any problem.
    MC> If you happened to set the switch to match the 4235's default settings then you got lucky and it worked (it won't always work). It is best to leave the switch configured for auto negotiate.
    We were advised against making the config change on the sensor as it is not supported.
    MC> Correct that this is not supported in version 4.1. (NOTE: The feature is being added in 5.0)
    A smart IDS should be able to recognize whatever mode on the switch, don't you agree?
    MC> No, When settings the speed and duplex on one device you need to set the speed and duplex on the other device as well. Or you could wind up with mismatches in speed and duplex settings that would prevent you from establishing a link. If the speed/duplex is only set on one device, that device does not advertise it's speed and duplex to the other device. Sometimes you get lucky and the default on the other device works, but you can't always count on this.
    Are we missing something? Can anyone clarify please.

Maybe you are looking for