Multicast Group Questions

If I have 10 multicast streams going to flash media server how should I configure Group name and stream names ? Each stream will be watched by a different set of people but pushed to same FMS . Should giving each stream a seperate name and seperate stream name will be fine ?
Thanks

It depends on how do you want to implement. You can have different streams under one NetGroup or you can create one NetGroup for one set of people.

Similar Messages

Routed access design to support up to 2500 multicast groups

Hi all,
I'm doing routed access design with following:
- Routing protocol is OSPF.
- 3 layers with: Core/Distribution/Access design.
- Core: Catalyst 6500E Sup 2T.
- Distribution: ME 3800X / Catalyst 4500X.
- Access: ME 3400E.
We enable PIM from Core to Access layer.
At the moment, there are around 1000 multicast group in the network.
Cisco hardware limitation of multicast:
- Cisco ME 3400E can only support up to 1000 IGMP groups and multicast routes.
- Cisco ME 3800X can only support up to 2000 IGMP group and multicast routes with Metro IP Services license.
- Cisco Catalyst 4500X support up to 32K multicast routes.
- Cisco Catalyst 6500E Sup 2T support up to 128K multicast routes.
Requirement:
Design the network to support up to 2500 multicast groups.
So, are there any fine-tune in the design to meet above requirement?
Or I have to change access layer switch to higher model? (ex: 4500E)
Thanks,

hi there
replacing the hardware is one option
but the question here why are you going with routed access layer and in your case you have large Mcast routing table that going to be everywhere !!
have you considered using Cat65k or 45K in the distribution in VSS and use L2 from access to distribution
in this case you need to concern about the distribution routing table size and also VSS simplify the topology and managerially of the network rather than have it with a complicated routing design
routed access is a good and recommend as well for quicker convergence time and no reliance on STP or HSRP/VRRP timers
however it could comlicate the routing design !
while VSS will also eliminate the reliance on HSRP/VRRP and STP with more added simplicity to the topology and design
so you may go with 45K VSS in the distribution and 65K with VSS in the core as well and in the access using L2 uplinks to the distribution and the uplinks everywhere can use multi-chassis etherchannel MEC, for increased network capacity in terms of bandwidth and quicker convergence time in the case of a link failure as well, Plus you will be able to support the desired Mcast routing table in the distribution and Core !!
hope this help

Maximum number of multicast groups on a given port?

Using linux (Ubuntu 8.4 with jdk 1.6.10 and 1.5.12) if I try to join more than 20 groups to a MulticastSocket open on a given port I get the following exception:
FINE 14:45:45.883:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.0.1.118
FINE 14:45:45.883:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.12.1.113
FINE 14:45:45.884:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.20.0.43
FINE 14:45:45.884:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.0.0.42
FINE 14:45:45.884:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.115.246.43
FINE 14:45:45.885:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.125.246.43
FINE 14:45:45.885:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.125.246.75
FINE 14:45:45.885:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.12.1.120
FINE 14:45:45.885:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.215.246.45
FINE 14:45:45.886:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.12.0.43
FINE 14:45:45.886:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.125.246.20
FINE 14:45:45.886:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.1.1.120
FINE 14:45:45.887:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.12.1.118
FINE 14:45:45.887:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.115.246.42
FINE 14:45:45.887:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.115.246.20
FINE 14:45:45.888:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /233.215.246.47
FINE 14:45:45.888:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.12.1.112
FINE 14:45:45.888:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.1.0.43
FINE 14:45:45.888:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.1.1.118
FINE 14:45:45.889:net.c4i.socketcomms.MulticastSocketListener.bind 10: Adding Mcast Group: /226.1.0.75
WARNING 14:45:45.889:net.c4i.socketcomms.MulticastSocketListener.bind 10: Could not join multicast group /233.15.246.43
java.net.SocketException: No buffer space available
java.net.PlainDatagramSocketImpl.join(Native Method)
java.net.PlainDatagramSocketImpl.join(PlainDatagramSocketImpl.java:172)
java.net.MulticastSocket.joinGroup(MulticastSocket.java:276)
net.c4i.socketcomms.MulticastSocketListener.bind(MulticastSocketListener.java:143)
net.c4i.socketcomms.MulticastProtocolManager.createListener(MulticastProtocolManager.java:154)
net.c4i.socketcomms.MulticastProtocolManager.<init>(MulticastProtocolManager.java:100)
net.c4i.socketcomms.MulticastProtocolManager.getInstance(MulticastProtocolManager.java:67)
sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
java.lang.reflect.Method.invoke(Method.java:585)
net.c4i.socketcomms.SocketManager.createProtocol(SocketManager.java:245)
net.c4i.socketcomms.SocketManager.<init>(SocketManager.java:157)
net.c4i.socketcomms.SocketManager.getInstance(SocketManager.java:116)
net.c4i.infrastructure.C4IAdapterManager.<init>(C4IAdapterManager.java:172)
net.c4i.infrastructure.C4IAdapterManager.getInstance(C4IAdapterManager.java:263)
net.c4i.infrastructure.C4IAdapterManager.launch(C4IAdapterManager.java:492)
sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
java.lang.reflect.Method.invoke(Method.java:585)
Is there any known way to workaround this issue? I've searched but have not seen any similar issues. I also cannot reduce the number of multicast groups in use as I have no control of that.
Noah

Is there any known way to workaround this issue?It may be a tunable kernel parameter, and you may be able to tweak it somehow, but that would be a Linux question, not a Java question.
I also cannot reduce the number of multicast groups in use as I have no control of that.You may not have any choice. If the kernel will only give you 20, bad luck.

Multicast group 224.0.1.40 !

I came by in a forum and this question really attracted me, my topology is completely based on dense mode, then why the routers join this group by default, if they dont will the topology will ever run into any problem if in future sparse mode is configured ?

Hi,
AFAIK the 224.0.1.40 is a reserved multicast group for cisco's Rp descovery. All cisco routers are members of this grup by default and listen to this group for Cisco RP discovery messages advertised by mapping agent even if it is not configured.I dont think there would be any problems if you change your topology to Sparse mode. But you should take in to account all the pros and cons.
HTH,
regards,
Shri :)

Slm2005 not leaving multicast group

Hi,
I have this SML2005 smart switch I try to use surf over some video multicast group. I have 2 video sources streaming on different multicast addresses, and 1 video receiver with a single channel. All are connected directly to the switch.
When I toggle between both sources on the receiver end, I clearly see the igmp group leave for the address of the previous channel and the igmp report for the new address on the latest channel. This is observed while setting a port of the switch in mirror and capturing the ethernet traffic. So far so good.
What I observe is the the switch continues to send the stream of the previous channel to the receiver, even though there was an explicit igmp group leave.
The switch is configured as follow:
global:
-enable igmp snooping: on
-Enable Unregistered IP Multicast Flooding: off
vlan 1:
-enable igmp snooping: on
-Enable IGMP Querying: on
I expected some latency leave, but I let this run for 20minutes with the unwanted stream continuously being sent to the receiver.
What this causes is when both stream are present, the port of the receiver may exceed the allowed bandwidth and some packets are dropped.
Any help would be appreciated.
--Martin

Hi and welcome to the Skype Community,
You can leave a group chat any time without compromising any of the separate conversation history or 1-on-1 conversation with the contact who originally invited you to the group. Take a look at our article on group chats explaining its features:
https://support.skype.com/en/faq/FA1005/how-do-i-start-a-group-instant-message-in-skype-for-windows-...
P.S.: Just out of curiosity: When you open the Windows discussion board do you see the big green "New Topic" button on top:
If not do you see any other button at that location?
Follow the latest Skype Community News
↓ Did my reply answer your question? Accept it as a solution to help others, Thanks. ↓

SNMP query for multicast group count

I want to get the total number of multicast groups on a switch. Is there an oid that will give me the info (ie. the number that's given by "show ip igmp snooping groups count")? If there is no oid, is there a way query the groups (show ip igmp snooping groups)? I'm thinking with the 2nd option I might be able to count the number of entries returned by an snmpwalk or something and turn that into a number.
Right now i'm logging into the switch with a script and running the command but that might be a problem with all those logins (only every 5 minutes though) so I'd like to do this via SNMP if possible.
Any ideas?
Thanks

Works just fine!!!
Thanks!
LA$$E
Hi,
Well, at the very least, you can create a formula
column in your break group using a combination of the
same where clause as your original query and the
break values for the group so that you are getting
the correct records. If you have it in the break
group, then it will execute for each record in that
group.
Hope that helps,
Toby

Join multicast group on AIR DatagramSocket ?

Hello Adobe experts,
I'm currently testing Flex/AIR for a new software that we have to design to display data received from the network and I'm very excited about all possibilities Flash offers in terms of UI and how easy it is to quickly develop a nice-looking, portable application, compared to other technologies (like Java, though richer in terms of API but harder to write code reusable for desktop, Internet and mobile apps, and not so good looking ).
The application itself gets data from a multicast group and formats it on the screen (images, graphs, ...).
So I'm really willing to use Flash to build this app and after a few benchmarking/prototyping I'm using a DatagramSocket (AIR-only, that's fine) to read UDP datagrams and it's working fine except for two tiny things on which I couldn't find any answer:
I can't have the socket join a multicast group
I can't set the send/receive buffer size for the socket
I have a lot of code in C and Java for that and both are just a single call to "setsockopt" function:
In C:
setsockopt(sock, IPPROTO_IP, IP_ADD_MEMBERSHIP, &group, sizeof(struct ip_mreq)) - where "group" describe the multicast group to join
setsockopt(sock, SOL_SOCKET, SO_RCVBUF, &sz, sizeof(sz)) - where "sz" is the number of byte to use for the receive buffer
In Java it's even simpler, using a MulticastSocket:
sock.joinGroup(group)
sock.setReceiveBufferSize(sz)
I tried to use Alchemy to write simple functions to create an UDP socket and have it join a multicast group but alas, the "socket" call returns -1, as it seems that sockets cannot be created in Alchemy environment (though compilation works without problem...).
I find it hard to believe that I'm the only one needing these two tiny "setsockopt" calls... I found on Adobe forums people trying to set the TCP_NODELAY option on a socket (also using setsockopt in the end) but without success...
By the way, the data come from an electronic board and is somehow coded in VHDL so I can't use RTMFP on that, though it seems to implement the functionnalities I need (or at least the multicast group part, the receive buffer size would be a bonus).
Well, I'm rather disapointed by this as Flash fills remarkably 99.9% of the application needs but this tiny 0.1% makes it impossible for it to get the data, thus making it useless...
And we don't want that I turn to Silverlight, do we?
Thank you all for any help and answers on this topic!
Matthieu

Thanks, that's what I was planning to do but in C or Java (everyone has its favorites right? ): use another program as a kind of "proxy" to redirect the multicast streams to local unicast AIR sockets...
But I'd rather have a nice API modification from Adobe on that matter. This kind of proxy is clearly "hacking" your way out of an API miss. You also have to make sure that your "proxy" is running on the local machine.
It's not as if multicast was a new technology or something nobody uses... It's standard and it's simple when you can already handle UDP sockets so I don't know why they don't want to implement these 10 lines of code...
Anyway, thank you for your answer, I'll stick to something similar (but I'll do Java ).
Matthieu

RTPManager: problem joining multicast group

We've written an application that sends and receives audio and video streams over RTP to a multicast group. If two of these applications are both running on the LAN, it works fine (our router is multicast-enabled). If one application (A) is running on the LAN, and another (B) is running remote and connected to the LAN via VPN, then A receives B's audio and video, but B sometimes doesn't receive A's audio and video. If another application (C) is added on the LAN, C behaves just like A - C receives audio and video from both A and B, but B still doesn't see either A or C.
On some occasions, rebooting B and re-establishing the VPN connection fixes the problem, and on other occasions, this doesn't change the behavior at all.
Using Ethereal, we've confirmed that B doesn't receive any of the traffic from the multicast group, even though its successfully sending traffic to that group. Further research revealed that, on B, the IGMP join requests were being sent to B's normal network interface, and NOT to the VPN network interface. However, all the RTP and RTCP traffic was being sent to the VPN network interface.
Does anyone know how to control (or at least affect) which network interface is used by RTPManager when joining the multicast group? I'm not sure if this problem is a result of RTPManager, of the NIC driver, or something else. I think one option is to write an RTPConnector implementation that handles all the transmission and reception manually, but I hope there's a better, easier option. Thanks for any help.

Hi Eric,
I am doing exactly the same thing. I have the video running fine, and I am relying on VPN connection for remote monitoring of the video. My setup is a little more complicated - I am running a setup/start/stop/teardown session manager (like rtsp) and also a keep alive listener (like rtcp) for each user.
I have one little annoyance which is stoppping me from testing this over the VPN, on my remote machiens I get a startup error:
11:40:13.4 request = control_start
11:40:15.1 Video transmitted as:
11:40:15.1   JPEG/RTP, 320x240, FrameRate=15.0
Failed to realize: com.sun.media.ProcessEngine@7cbde6
Cannot build a flow graph with the customized options:
    Unable to transcode format: RGB, 320x240, FrameRate=15.0, Length=230400, 24-
bit, Masks=3:2:1, PixelStride=3, LineStride=960, Flipped
      to: JPEG/RTP, 320x240, FrameRate=15.0
      outputting to: RAW/RTP
Error: Unable to realize com.sun.media.ProcessEngine@7cbde6
11:40:16.1 response = 500I have run a debug on the two networks, the multicast runs fine on my machine, and the processor is realized, and the cameras have identical traces looping over the supported formats. Is this purely a format issue or could this be a router not configured for multicast?
Do you (or anyone!) have an idea about this? The program works on my setup of camera etc, and on a clone machine, with diff cameras (but the same if you know what I mean) it fails...
I'll let you know if I solve the VPN problem (after I fix this) I watched this topic.

NDP - Solicited node multicast group

Hello everyone,
I can`t understand the process of layer3 to layer2 address resolution in IPv6.
In order to make a Layer-2 (eg. Ethernet) delivery, it needs to know the target host's hardware address (eg. "Ethernet MAC address"). But in order to do this, it must first determine which hardware address to send it to. To do this, an IPv6 host will construct the Solicited-node Multicast Address related to the target address.
The IPv6 solicited-node multicast address has the prefix FF02:0:0:0:0:1:FF00:0000/104 concatenated with the 24 low-order bits of a corresponding IPv6 address.
For the target with address
fe80::2aa:ff:fe28:9c5a
solicated node multicast address would become
ff02::1:ff28:9c5a.
1- For example when a host has two unicast address 2001::AB:CDEA and link local fe80:2aa:ff:fe28:9c5a on its interface, does it join the associated solicited-node multicast group for every address to which it is assigned or only for link local adress?.
2- If only for link local address, how will the source device know the link local address of target to borrow last 24 bit? Can I use global address for target address while layer3 to layer2 address resolution?
3- Does all ND functions use link local address? Must a node know all neighbours link local address? If yes, what mechanism are needed to learn link local address of neighbours.
All replies will be appreciated.
Thanks
Gokmen...

Hi,
1) a node always joins the sollicited-node multicast for every IPv6 address it has configured on the ethernet interface including the link-local.
2) this destination multicast address is used for neighbour sollicitation( equivalent of ARP in IPv4) and DAD( duplicate address detection) which is the equivalent of Gratuitous ARP request in IPv4.
So when you want to communicate with a destination IPv6 address it must first do a NS to find out the L2 address and it will send it to the sollicited-node multicast corresponding to the IPv6 destination address
when the node is configured with an IPv6 address it will verify if it is unique by sending a NS to its sollicited-node multicast and if it receives a reply it considers the address is not unique.
3) no it must know how to send to link-local if it wants to communicate with link-local only
http://www.ietf.org/rfc/rfc2461.txt
Regards.
Alain

Multicast group capacity

We are using 3560 and 3750 PoE switches in a network which
interconnects a lot of clients (WIN XP professional) using a proprietary VoIP software
solution.
This Software uses Multicast groups for communication and now we are
running into the limitation of 1000 multicast groups supported by these
switches.
Does anybody know if there is any possibility to extend the maximum
number of Multicast groups on the 3560 and 3750 Cisco switches? IOS
update? Memory extension?
Thanks in advance

Hi,
Currently, we support only 1024 multicast routes+IGMP groups.
To extend the maximum number, we need code changes.
At the moment, no code changes planned.
Tom.

Can iAS multicast group cross clusters & machines

Say I have 2 clusters C1 & C2 deployed on different machines. Each cluster has 2 iAS instances. All components have Sticky flag.
Now I have 2 web servers iWS1 & iWS2 (on separate machines too).
A load balancer distributes requests between web servers using server affinity rule.
I would like the following to happen and am not sure of the settings that must be made
1) User1 comes through iWS1 and sets up a session on C1.
2) Subsequent requests from user1 will follow the same path because of server affinity at LoadBalance level and stickiness at web connector level.
3) If iWS1 is brought down, Load balancer would be forced to send User1 requests to iWS2. I would like iWS2 to start forwarding user1 requests to C1.
Also, it would be great if in some way, iWS2 could send any new users without existing session to C2 only. This would allow me to bleed users of C1 to perform an upgrade.
Thanks
GN

If C1 and C2 share the same ldap, then both the webservers will be able to forward requests to both the clusters (provided the application is deployed on all the servers). This is one solution to the problem.
Another solution, which is probably a hack is that you manually put the IP address of all the servers under the GUID of the application to be distributed.
This will allow the webserver to forward requests to any of the servers in C1 and C2.
In fact this method can be used to load balance even across clusters/multicast groups.
Make sure you make the load balancing option as web connector driven.
Note that the Multicast group doesn't come into picture while load balancing. You might see a number of iAS servers registered under \\6.5\GMS, but the webconnector doesn't forward requests to them unless they are also registered under the application GUID.
If you want that W2(webserver of C2) should forward requests to C1 only if W1 is down, then I don't know how to do that.
Also I don't know how to make sure that W2 should forward requests from all the new users only to C2 in that case.
-harjit

DMVPN w/ Multicasting setup/questions

Hello
I have a lot of questions, so bare with me as i puke them out of my head.
I have been doing some testing with DMVPN inconjuction with multicasting video (Hub and spoke, w/ no spoke to spoke). The test setup is using 2 cisco 2811 w/out the vpn module. I understand the performance hit with not having the module. With that being said here are my questions.
1. With encryption on both the HUB and spoke routers are using 90-97% cpu (8Mb multicast stream). With encryption off, the Hub is around 60%, and spoke around 75%. Here is where i'm confused. If i send that same stream as a unicast stream, w/ encryption on, both the Hub and spoke are only using around 30-35% cpu. Why is there so much more cpu need when its a multicast stream?
2. In the current config i'm seeing input, throttles, and ignore errors on the Hub and spoke. The Hub has these errors on the LAN interface, and the spoke has these errors on the WAN interface. All other interfaces are totally clean. I have checked and there are no duplex or speed mismatches. Any ideas?
HUB:
Current configuration : 1837 bytes
version 12.4
service timestamps debug datetime msec
service timestamps log datetime msec
no service password-encryption
hostname Hub
boot-start-marker
boot-end-marker
logging message-counter syslog
enable password
no aaa new-model
clock timezone Central -6
dot11 syslog
ip source-route
ip cef
no ip domain lookup
ip name-server 8.8.8.8
ip multicast-routing
no ipv6 cef
multilink bundle-name authenticated
voice-card 0
archive
log config
hidekeys
interface Tunnel1
bandwidth 100000
ip address 192.168.11.1 255.255.255.0
no ip redirects
ip mtu 1400
no ip next-hop-self eigrp 1
ip pim sparse-mode
ip nhrp map multicast dynamic
ip nhrp network-id 1
ip nhrp holdtime 450
no ip route-cache cef
ip tcp adjust-mss 1360
no ip split-horizon eigrp 1
delay 1000
tunnel source FastEthernet0/0
tunnel mode gre multipoint
tunnel key 100000
tunnel bandwidth transmit 100000
tunnel bandwidth receive 100000
interface FastEthernet0/0 (WAN)
ip address 216.x.x.x 255.255.255.192
ip pim sparse-mode
load-interval 30
duplex auto
speed auto
interface FastEthernet0/1 (LAN)
ip address 128.112.64.5 255.255.248.0
ip pim sparse-mode
load-interval 30
duplex auto
speed auto
router eigrp 1
network 128.112.0.0
network 192.168.11.0
auto-summary
ip forward-protocol nd
ip route 0.0.0.0 0.0.0.0 216.x.x.x
ip http server
ip http authentication local
ip http secure-server
ip pim rp-address 128.112.64.5 10
access-list 10 permit 239.10.0.0 0.0.255.255
snmp-server community public RO
Spoke:
Current configuration : 1857 bytes
version 12.4
service timestamps debug datetime msec
service timestamps log datetime msec
no service password-encryption
hostname Spoke
boot-start-marker
boot-end-marker
logging message-counter syslog
enable password
no aaa new-model
clock timezone central -6
dot11 syslog
ip source-route
ip cef
no ip domain lookup
ip multicast-routing
no ipv6 cef
multilink bundle-name authenticated
voice-card 0
archive
log config
hidekeys
interface Tunnel1
bandwidth 100000
ip address 192.168.11.2 255.255.255.0
no ip redirects
ip mtu 1400
ip pim sparse-mode
ip nhrp map 192.168.11.1 216.x.x.x
ip nhrp map multicast 216.x.x.x
ip nhrp network-id 1
ip nhrp holdtime 450
ip nhrp nhs 192.168.11.1
no ip route-cache cef
ip tcp adjust-mss 1360
no ip split-horizon eigrp 1
delay 1000
tunnel source FastEthernet0/0
tunnel destination 216.x.x.x
tunnel key 100000
tunnel bandwidth transmit 100000
tunnel bandwidth receive 100000
interface FastEthernet0/0 (WAN)
ip address 65.x.x.x 255.255.255.192
ip pim sparse-mode
load-interval 30
duplex auto
speed auto
interface FastEthernet0/1 (LAN)
ip address 128.124.64.1 255.255.248.0
ip pim sparse-mode
ip igmp join-group 239.10.10.10
load-interval 30
duplex auto
speed auto
router eigrp 1
network 128.124.0.0
network 192.168.11.0
auto-summary
ip forward-protocol nd
ip route 0.0.0.0 0.0.0.0 65.x.x.x
no ip http server
no ip http secure-server
ip pim rp-address 128.112.64.5 10
access-list 10 permit 239.10.0.0 0.0.255.255
snmp-server community public RO

Joe,
You ask the right question.
CPU ultization = CPU consumed by processes + IO operations (in a huge simplification - CEF)
Typically when a packet is processed by router we expect it to be be processed by CEF, i.e. very fast.
Packet is not processed by CEF:
- when there is something missing to route the packet properly (think missing ARP/CAM entry) i.e. additional lookup needs to be done.
- a feature requests that a packet is for processing/mangling
- Packet is destined to the router
(And several other, but those are the major ones).
When a packet is recived, but cannot be processed by CEF, we "punt the packet to CPU" this in turn will cause the CPU for processes to go up.
Now on the spoke this seems to be the problem:
Spoke#show ip cef switching stati Reason Drop Punt Punt2HostRP LES Packet destined for us 0 1723 0RP LES Encapsulation resource 0 1068275 0
There were also some failures on one of the buffer outputs you've attached.
Typically at this stage I would suggest:
1) "Upgrade" the device to 15.0(1)M6 or 12.4(15)T (latest image in this branch) and check if the problem persists there.
2) If it does, swing it by TAC. I don't see any obvious mistakes, but I'm just a guy in a chair same as you ;-)
Marcin

Multiple Artwork Covers Issue & Grouping Question

My favorite band is The Smashing Pumpkins. Their latest album +Teargarden by Kaleidyscope+ is being released one song at a time online for free and in 11 different limited EPs that will compose the whole album. My problem here is that there a two artwork covers, one for the entire album and a second for each EP (only one EP has been released, the first song for the second EP is out). The first EP displays with the EP cover but the one song from the second EP only displays the overall album cover. How do I get my Classic to display only one cover instead of picking randomly (as it seems)?
The following question deals with the same album. I want to group the EPs that are to be released in the future so that on my Classic I see only one big album instead of multiple EPs. Unlike their Machina II these EPs have subtitles for them. What I am wanting is one album cover to display (the overall album artwork), the name to display as +Teargarden by Kaleidyscope+ not +Teargarden by Kaleidyscope Vol. 1: Songs for a Sailor+ which is the name of the first EP in the album section of my Classic. Is all of this possible?

Just lke dropping support for 5 year old Tiger OS in iTunes while continuing support for 10 year old XP is an enhancement. Just like making the Library directory invisible in Lion is an enhancement. Just like dropping support for PPC generation applications in Lion is an enhancement. Just like, etc.
https://discussions.apple.com/thread/3496844?start=0&tstart=0
https://discussions.apple.com/thread/3650998?start=0&tstart=0
https://discussions.apple.com/thread/3529592?start=0&tstart=0
https://discussions.apple.com/thread/3503860?start=0&tstart=0
https://discussions.apple.com/thread/3498690
More with a web search using:
site:discussions.apple.com itunes artwork arrow

Authenticated Users Group Question

I have a quick question regarding the Authenticated Users "group". I used to be a systems administrator, but I'm a bit rusty since I've been a software developer for the last 10 years. A conflict with data center operations (DCO) group
at work lead me to get another opinion.
The question is this... is the authenticated users group a domain-level group or is there a local authenticated users group that would allow only users authenticated locally? We have a share that permits the authenticated users group access.
My opinion is that all domain users who have authenticated successfully have access to this share. The DCO group is telling me that this is the local (to the server containing the share of course) authenticated users group only.
Is there such a thing as a local-only authenticated users group? To me this doesn't even make sense, but I could very well be wrong.
Nathon Dalton
Sr. Software Engineer
Blog: http://nathondalton.wordpress.com

I apologize. I don't think I explained myself correctly. Let's consider the following...
SERVER: SERVER1
DOMAIN: DOMAIN1
SHARE: \\SERVER1\SHARE1
SHARE PERMISSIONS: Authenticated Users - Full Control
Given the above information, is it possible that the Authenticated Users group will allow ONLY users that are defined on SERVER1 to access \\SERVER1\SHARE1?
My understanding is that's not possible. There's one defined Authenticated Users group and that represents ALL users that are authenticated against DOMAIN1, whether added to local groups, shares, etc.
What I'm being told however is that SHARE1 having Authenticated Users assigned is okay since only those user accounts defined on SERVER1 will be able to access it. All the users in the domain will NOT be able to access it. I think this is bogus. Am I wrong?
Nathon Dalton
Sr. Lead Developer
Blog: http://www.nathondalton.com

A few SQL related protection group Questions - DPM 2012 R2

Hi - Here is my environment: DPM 2012 R2 backing up SQL 2012 DBs. Some of the DBs are simple recovery model, some are full recovery model in the same protection group.
1) Regarding the fact that DPM only looks to the recovery model when the DB is first added to the protection group, if I ever changed the recovery model in the future, would stepping through the "Modify Protection Group" wizard without changing
anything be the solution for DPM not changing its methodology on the backing up or not backing up the log files? Or would I have to remove the affected databases from the protection group and add them back in?
2) I have disk protection configured for 30 Days, sync every 15 minutes, and application recovery points once per day. Online protection is configured for 90 Days. Does this mean my disk consumption on the DPM side for full model databases will be the full
size of the databases, plus the amount of changes observed over 30 days? OR does it mean it will keep 30 full copies of the DB? I am fairly sure the former.
3) If I do decide to reduce the retention range on the disk side of this protection group to say 14 days, are there any specific issues with changing that after the PG has been working for some time?
4) My former backup solution had an option to run consistency checks after every SQL job. Is that still considered best practice? Would the "Run a daily consistency check" be a better choice than the "Only run if a replica becomes inconsistent"
option that I currently have in place? Would there be any ramifications to enabling this now after this PG has been in place for some time?
Thanks very much!

Hi
Q1) Regarding the fact that DPM only looks to the recovery model when the DB is first added to the protection group, if I ever changed the recovery model in the future, would stepping through the "Modify Protection Group" wizard without changing
anything be the solution for DPM not changing its methodology on the backing up or not backing up the log files? Or would I have to remove the affected databases from the protection group and add them back in?
A1)
Changing the Recovery Model of a Database
To change the recovery model of a protected database to the simple recovery model
Stop protection of the database, selecting the retain replica option.
Change the recovery model on the SQL Server database.
Add the database to a protection group.
Q2) I have disk protection configured for 30 Days, sync every 15 minutes, and application recovery points once per day. Online protection is configured for 90 Days. Does this mean my disk consumption on the DPM side for full model databases will be the full
size of the databases, plus the amount of changes observed over 30 days? OR does it mean it will keep 30 full copies of the DB? I am fairly sure the former.
A2) Your disk consumption on the DPM will be for size of model databases plus the amount of changes observed over 30 days.
Q3) If I do decide to reduce the retention range on the disk side of this protection group to say 14 days, are there any specific issues with changing that after the PG has been working for some time?
A4) Not really, but to gain back the potential extra space allocated for the original 30 day retention - you can try shrinking the recovery point volume once the retention range is reduced and the older recovery points are pruned. I would wait until
after midnight to be sure the old RP's are deleted, then try to shrink.
Q4) My former backup solution had an option to run consistency checks after every SQL job. Is that still considered best practice? Would the "Run a daily consistency check" be a better choice than the "Only run if a replica becomes inconsistent"
option that I currently have in place? Would there be any ramifications to enabling this now after this PG has been in place for some time?
A4) A consistency check will only run if DPM detects the replica is inconsistent. Even if you have the option selected to run nightly, if the replica is already consistent a consistency check will be skipped.
Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread. Regards, Mike J. [MSFT]
This posting is provided "AS IS" with no warranties, and confers no rights.

Multicast Group Questions

Similar Messages

Maybe you are looking for