Link-state protocol on blade switches

Hi,
Can anyone confirm is after an uplink is reestablished the down stream interfaces are automatically restored when using link-state protocol?
The documentation states the downstream ports in a group are put in error disable state but doesn't say if they are automatcally made up when the upstream interface comes back up. This refers to Cisco blade 3020 switches running 12.2.25
http://www.cisco.com/c/en/us/td/docs/switches/blades/3020/software/release/12-2_55_se/configuration/guide/3020_scg/swethchl.html#wp1346176
Thanks
Ken

Link state tracking, also known as trunk failover, provides Layer 2 redundancy in the network when used in conjunction with server network interface card (NIC) adapter teaming. Link-state tracking is used to mirror the state of the ports that carry upstream traffic from connected hosts and servers, and to allow the failover of the server traffic to an operational link on another Cisco Ethernet switch. Check out the following link for more information on link state tracking :
http://www.cisco.com/en/US/products/ps6406/products_configuration_guide_chapter09186a00805a75e0.html#wp1285238
Hope this helps.

Similar Messages

LSAs in link state protocols

Lets say we have a stable network and we are using a link-state routing protocol. Then something fails. Will the routers reflood ALL LSAs when one LSA changes or only refloods the changed LSAs?

Francisco
Only the LSA's that have changed will be advetised out. That is one of the big advantages of OSPF in that only affected links are advertised rather than the entire OSPF database.
Jon

Blade Switch Management

I am deploying HP BladeSystem C Class with Cisco 3020 Ethernet blade switches connecting to Catalyst 6500 at the aggregation layer. What are my best options for managing layer 2, VLANs, etc, across the Cat6K and 3020 switches?

Matt,
Great catch, thank you!
Switch#show link state group detail
Link State Group: 1 Status: Disabled, Up
Upstream Interfaces : Po1(Up)
Downstream Interfaces : Gi0/1(Up) Gi0/2(Up) Gi0/3(Up) Gi0/4(Up) Gi0/5(Up)
Gi0/6(Up) Gi0/7(Up) Gi0/8(Up) Gi0/9(Up) Gi0/10(Up)
Gi0/11(Up) Gi0/12(Up) Gi0/13(Dwn)
Link State Group: 2 Status: Disabled, Down
Upstream Interfaces :
Downstream Interfaces :
(Up):Interface up (Dwn):Interface Down (Dis):Interface disabled
Switch#conf t
Enter configuration commands, one per line. End with CNTL/Z.
Switch(config)#link state trac
Switch(config)#link state track 1 ?
Switch(config)#link state track 1
Switch(config)#end
Switch#show link state group detail
Link State Group: 1 Status: Enabled, Up
Upstream Interfaces : Po1(Up)
Downstream Interfaces : Gi0/1(Up) Gi0/2(Up) Gi0/3(Up) Gi0/4(Up) Gi0/5(Up)
Gi0/6(Up) Gi0/7(Up) Gi0/8(Up) Gi0/9(Up) Gi0/10(Up)
Gi0/11(Up) Gi0/12(Up) Gi0/13(Dwn)
Link State Group: 2 Status: Disabled, Down
Upstream Interfaces :
Downstream Interfaces :
(Up):Interface up (Dwn):Interface Down (Dis):Interface disabled
Switch#
Scott
www.xpresslearn.com

Routing loop avoidance in Link State

"The link state method is not susceptible to routing loops which is an advantage over the distance vector method" Why? How it avoid routing loop?

Link-state protocols like OSPF use reliable flooding mechanisms to ensure that all routers are acting on the basis of the same information. Loops occur when routers act on the basis of inaccurate or old information. That is what link-state protocols avoid. All routers in a link-state database have the same view of the network.
Paresh

Link State Routing Protocol Question

"In LSP, one router in each area is designated as the authoritative source of routing information (called a designated router). Each area router receives updates from the designated router" Why need designated router? How it work? Why can?t it just broadcast LSP and leant the routing information without the need of designated router? Is designate router the same as backbone or root area in OSPF? Is "area" concept only be used in Link State Routing Protocol OSPF?

hi...
you will find area topology in IS IS also...
here we are using the Area as well as DR and BDR for reducing the LSA flooding in the area... each router in ospf area will send update to the DR on multicast address and then DR will send the multicast update to all other router in the area... here each and every router in the area have the full adjucancy with DR but they are not in the full adjucancy with any other router ...
hope this will help you
rate this post if it helps
regards
Devang

Etherchannel and STP on HP Blade Switches

Hi Guys,
I have a 6509 with and etherchannel connection to an HP Blade switch... the channel is made up of two phy ports between the two switches...
When I try ti bring up the link, the 6509 ports go into err-disable and i get the message below.
I have done all (I think?) checks that there is to ensure etherchannle compatibility and STP stuff but still no luck...
can anyone suggest soemthing I do ro check that might shed light on this...
thanks
2005 Jun 26 02:53:06 AEDT +10:00 %ETHC-5-PORTTOSTP:Port 10/24 joined bridge port 10/24-25
2005 Jun 26 02:53:06 AEDT +10:00 %SPANTREE-6-PORTLISTEN: Port 10/24-25 (agPort 14/45) state in VLAN 144 changed to listening
2005 Jun 26 02:53:20 AEDT +10:00 %SPANTREE-6-PORTLEARN: Port 10/24-25 (agPort 14/45) state in VLAN 144 changed to learning
2005 Jun 26 02:53:35 AEDT +10:00 %SPANTREE-6-PORTFWD: Port 10/24-25 (agPort 14/45) state in VLAN 144 changed to forwarding
2005 Jun 26 02:54:23 AEDT +10:00 %SPANTREE-2-CHNMISCFG: STP loop - channel 10/24-25 is disabled in vlan/instance 144
2005 Jun 26 02:54:23 AEDT +10:00 %SPANTREE-2-CHNMISCFG2: BPDU source mac addresses: 00-03-fe-af-d0-b5, 00-11-f9-21-56-00
2005 Jun 26 02:54:23 AEDT +10:00 %ETHC-5-PORTFROMSTP:Port 10/24 left bridge port 10/24-25

Hi ankur,
thanks for your response. currently the etherchannel operatiosn mode on the 6509 is set to manual, which is the case globally for the 6509 and we have 50 pairs of this blade switch running on this 6509... I don't think that that is the cuase of this issue since this is the only one not forming the channel and goes into err-disable...
the blades also have similar configs in them...
the only thing I see different on ports 10/24-25 is that it has bpdu-guard set to disable while the rest are set to default... could this make a diffrence?
also, does the order of enabling the cisco end and the HP blade end make a difference? i.e. should I bring up one first than the other?
the thing is I cannot "experiment" fully as this is a full prod network...
Any input si appreciated...

Cisco 3020 Blade Switch gecrashed

Hi,
One of our Blade switches has crashed, error log as follows,
*Mar 1 05:31:48: %LINEPROTO-5-UPDOWN: Line protocol on Interface Vlan550, changed state to up
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: System previously crashed with the following message:
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Cisco IOS Software, CBS30X0 Software (CBS30X0-LANBASE-M), Version 12.2(25)SEF1, RELEASE SOFTWARE (fc3)
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Copyright (c) 1986-2006 by Cisco Systems, Inc.
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Compiled Tue 23-May-06 15:26 by antonino
.Dec 26 22:21:20: %PLATFORM-1-CRASHED:
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Program Exception (0x0700)!
.Dec 26 22:21:20: %PLATFORM-1-CRASHED:
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: SRR0 = 0x003C38AC SRR1 = 0x00029210 SRR2 = 0x00451BDC SRR3 = 0x00021200
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: ESR = 0x08000000 DEAR = 0x00000000 TSR = 0x84000000 DBSR = 0x00000000
.Dec 26 22:21:20: %PLATFORM-1-CRASHED:
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: CPU Register Context:
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Vector = 0x00000700 PC = 0x003C38AC MSR = 0x00029210 CR = 0x20000008
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: LR = 0x003C38A4 CTR = 0x0027C814 XER = 0x60000020
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R0 = 0x003C38A4 R1 = 0x014D6650 R2 = 0x00000000 R3 = 0x00000004
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R4 = 0x013A1E60 R5 = 0x00000001 R6 = 0x00433EDC R7 = 0x00000004
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R8 = 0x00029210 R9 = 0x00000002 R10 = 0x01220000 R11 = 0x00F60000
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R12 = 0x0070D567 R13 = 0x00110000 R14 = 0x00000002 R15 = 0x00000000
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R16 = 0x00000000 R17 = 0x00000000 R18 = 0x01220000 R19 = 0x01220000
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R20 = 0x00F0CF5C R21 = 0x012D27D0 R22 = 0x01220000 R23 = 0x0187F530
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R24 = 0x00000000 R25 = 0x0004073F R26 = 0x01220000 R27 = 0xEF180E50
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: R28 = 0xEF19007A R29 = 0x00000000 R30 = 0x01220000 R31 = 0x00000000
.Dec 26 22:21:20: %PLATFORM-1-CRASHED:
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Stack trace:
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: PC = 0x003C38AC, SP = 0x014D6650
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 00: SP = 0x014D6678    PC = 0x003C38A4
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 01: SP = 0x014D6680    PC = 0x003C3964
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 02: SP = 0x014D66A0    PC = 0x004B97AC
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 03: SP = 0x0126264C    PC = 0x00451060
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 04: SP = 0x0126265C    PC = 0x00451008
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 05: SP = 0x0126267C    PC = 0x0045342C
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 06: SP = 0x01262684    PC = 0x004538F8
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 07: SP = 0x012626D0    PC = 0x00451110
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 08: SP = 0x0187F298    PC = 0x0000D3CC
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 09: SP = 0x0187F2A0    PC = 0x002659D8
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 10: SP = 0x0187F4D0    PC = 0x00423870
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 11: SP = 0x0187F4E8    PC = 0x00AAF21C
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 12: SP = 0x0187F528    PC = 0x006C9400
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 13: SP = 0x0187F5F8    PC = 0x006C8E04
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 14: SP = 0x0187F610    PC = 0x006C8A64
.Dec 26 22:21:20: %PLATFORM-1-CRASHED: Frame 15: SP = 0x0187F638    PC = 0x006C848C
.Dec 26 22:21:20: %PLATFORM-1-CRASHED:
Dec 27 09:10:14: %SYS-5-CONFIG_I: Configured from console by vty0 (172.22.25.22)
Please help to fixed this issue,
Asa,

Signal = 4, Vector = 0x700
Thank you very much for the output. They were very helpful.
Here's the story, the blade has been crashing because of the error "Signal = 4". I have never seen something like this so the only thing I can tell you is that this is a code that simply means "Illegal Opcode Exception". It could mean that you are running an old IOS and you're hitting a bug.
You can contact Cisco TAC and get them to analyze the crashinfo files. You can put this thread into the TAC Case so you don't need to repeat your story all over again.

Stacking with Cisco Blade Switches

Hi,
We have a HP p-class Blade System with two Cisco Gigabit Ethernet Switch Modules (Cisco IOS Release 12.2(25)SE) for every enclosure. My question is how I can stack the two blade switches (in the same enclosure) so that every HP blade server to be connected with the two switches in the active/standby state? In other words, can and how I make a port-channel which includes ports from the two switches?
Unfortunately I couldn't found any documentation for the purpose.

Hi,
With the CGESM switches you have you cannot use 802.3ad teaming against the two blade switches. This is possible with the Cisco 3120 blade switches for HP c-class.
Now, in your case, you can use standard Active/Standby NIC teaming, where the active NIC of each server connects to one switch and the standby NIC connects to the other switch. Should a blade switch completely fail, all of the servers with their active NIC connected to the failed switch should see the link go down and switch over to the standby NIC on the second switch. This should work very well.
One problem that can happen however is that the blade switch itself does not completely fail. Instead maybe the blade switch uplink fails, or the upstream switch it is connected to fails. In this situation the server NIC does not experience a failure and continues to send traffic to the blade switch that does not have anywhere to send the traffic, a black hole. This is perhaps the problem you are seeing?
The fortunate thing is that Cisco has a solution to this problem called Trunk Failover with Link State Tracking. What this allows you to do is to tell the blade switch that if it experiences a failure on its uplinks that it should also bring down the links to the servers as well. This will cause the servers to see a failure on their active NIC and perform a switchover to the second blade switch.
I looked up your switch and found that your 12.2(25)SE1 code does not support this feature. You will need to upgrade to 12.2(25)SEE4 and you will be able to configure this on your switches.
Here is a link to see how this configuration works:
http://www.cisco.com/en/US/docs/switches/blades/3020/software/release/12.2_25_sef1/configuration/guide/swethchl.html#wp1346176
Hope this helps.
If so please rate my posts.
Thanks,
Brad

"Cisco Catalyst Blade Switch 3020 for HP" Best Practices

Hi all,
Here we use this blade switch cluster for the Data Center server farm. Here we noticed some network traffic delay. when I monitor the up link I can see there is not much traffic in the link. So doubt about the traffic delay. If anyone can help me on this it would be great...
Thanks in Progress.....
Chandana

Here in the "blade switch for HP" a request processing application is running in a server. when we use the same type server stand alone with a separate switch it processes 80,000 requests per 4 hours. But when this blade server in the blade switch it only processes near 6,000 requests per 4 hours.
That what I mean like "Network delay"...
That is what I cannot understand.
I have attached the current configuration of the blade switch so you can go through it.
Is there any best practices specific to that type of blade switch that should enable?
Thanks for your help
regards
Chandana

MDS FC Blade Switch for IBM - Licensing

Hi,
our customer has got installed the 20-port-switch of the "Cisco MDS Fibre Channel Blade Switch for IBM BladeCenter". He wants to change some of the links now. For that reason he asked me:
- Is there a fix relationship between a port and a license?
- Can we use any port as long as we fall below the number of licenses?
Thanks in advance and regards
Matthias

Hello Matthias,
Cisco Fabric Switch for IBM BladeCenter (20 ports; 14 internal 2/4 Gbps, and 6 external full-rate ports) licenses for internal ports are allocated as the ports come up. Licenses for external ports are allocated sequentially.
Refer "On-Demand Port Licensing" section on CLI guide here :
http://www.cisco.com/en/US/docs/storage/san_switches/mds9000/sw/rel_3_x/configuration/guides/cli_3_3/ondport.html
Customer can select the specific external ports he wants licensed.
switch# config t
switch(config)# interface fc1/1
switch(config-if)#port-license acquire
Thanks,
Prateek

Catalyst Blade Switches 3020 For HP Blade Chassis

I was trying to access Catalyst Blade Switch 3020 through serial port to reset the lost password. This Switch is plug-in to HP c7000 Chassis. I followed the instruction from Cisco documentation for this password re-set process as follows.
{ Step 1 Connect a terminal or PC with terminal-emulation software to the switch console port.
Step 2 Set the line speed on the emulation software to 9600 baud.
Step 3 Push the release latch on the front of the switch to the open position.
Step 4 Pull the switch partially out of the module bay until the power disconnects, and then push it in again. The switch restarts when it reseats in the server chassis. The switch performs POST after power is applied.
Step 5 Push the release latch on the front of the switch to the closed position.
Step 6 Within 15 seconds, press the Mode button while the System LED is still flashing green. Continue pressing the Mode button until the System LED turns briefly amber and then solid green; then release the Mode button.
------- Step 6 did not work. System LED neither flashing nor in Solid state. I can not connect through Serial port any more! Any help.....

Here in the "blade switch for HP" a request processing application is running in a server. when we use the same type server stand alone with a separate switch it processes 80,000 requests per 4 hours. But when this blade server in the blade switch it only processes near 6,000 requests per 4 hours.
That what I mean like "Network delay"...
That is what I cannot understand.
I have attached the current configuration of the blade switch so you can go through it.
Is there any best practices specific to that type of blade switch that should enable?
Thanks for your help
regards
Chandana

LACP or Link State Tracking over VPLS?

Hi all!
I have 2 sites connected with VPLS. Both sites are now having a 2nd VPLS circuit installed (with a different carrier) for redundancy/failover. I've got a Catalyst 3750 at each end to work with.
My question: what's the best way to configure the 3750's? I was thinking either LACP with 2 physical interfaces (one for each VPLS line) - in which case traffic would be balanced across them, which is fine.
OR I could use Link State Tracking, such that if 1 link fails it would failover. Though I'm new to Link State Tracking so I don't know if this would actually work over VPLS.
Your thoughts are very appreciated.

Link state tracking, also known as trunk failover, provides Layer 2 redundancy in the network when used in conjunction with server network interface card (NIC) adapter teaming. Link-state tracking is used to mirror the state of the ports that carry upstream traffic from connected hosts and servers, and to allow the failover of the server traffic to an operational link on another Cisco Ethernet switch. Check out the following link for more information on link state tracking :
http://www.cisco.com/en/US/products/ps6406/products_configuration_guide_chapter09186a00805a75e0.html#wp1285238
Hope this helps.

VIF link state down

Hi
When I configured service profile to a blade then deploy ESXi, It works correctly for a time then I found the following error " VIF 716 link state is down" as in the attached and the server management IP is pingable also the ESXi IP is pingable during the error
I did the Same to another blade but no problem

Hi Amr
This has been discussed many times before
see eg. https://communities.vmware.com/message/2183957
But about the HW failover configuration I really do not know why not recommended to use the HW failover I think it's better than failover of VMware VSS
Agreed that it should work, but I'd prefer to enable and configure failover behavior in ESXi rather than UCS. In my research on this issue, I found a thread over on the Cisco UCS forum that has a clear statement regarding this configuration: https://supportforums.cisco.com/thread/2187653.
"Never use fabric failover with a hypervisor vswitch. Let the failover mechanism of the vSwitch handle the network outage."
So while it may be a valid design, it looks like the consensus is to let your vSwitch handle the failover.
Cheers
Walter.

Link-State Tracking?

Attached is a diagram of my fabric I will be deploying (data center environment). The access switches are 2960G (24port), and the distribution switches are 2960G (48port). A total of 18 access switches will be used.
I will be using Rapid PVST+ throughout the fabric. With that in mind, is it necessary to implement link-state tracking on the access switches as well? I am having difficulty finding out if Cisco recommends link-state tracking for performance improvment, or if it is absolutely necessary in my environment.
What exactly are the pros/cons of enabling link-state tracking on my access switches?
Thanks,

Link state tracking, also known as trunk failover, provides Layer 2 redundancy in the network when used in conjunction with server network interface card (NIC) adapter teaming. Link-state tracking is used to mirror the state of the ports that carry upstream traffic from connected hosts and servers, and to allow the failover of the server traffic to an operational link on another Cisco Ethernet switch. Check out the following link for more information on link state tracking :
http://www.cisco.com/en/US/products/ps6406/products_configuration_guide_chapter09186a00805a75e0.html#wp1285238
Hope this helps.

Two forwarding root ports on cisco 3120 blade switch.

Hi,
 
 After upgrading blade switch 3120 to 12.2(46)SE, I could see two root ports in which both are forwarding.
 
 Te1/0/1 Root FWD 2000 128.27 P2p
 Te2/0/1 Root FWD 2000 128.81 P2p
 
 Where as STP should've kept the other root port as blocked state and named it has Alternate port.
 Also started seeing Mac flap messages.
 As soon as IOS was downgraded to 12.2(40)EX3, Problem was solved, even messages were stoped.
 Seems to be bug, but couldn't get any.
 

Hi Rob,
I have faced this issue with many enclosures so far and the only option left here is, you need to reboot the enclosure.
This is an issue with the IOS code and you have to open a case with HP to get the appropriate IOS ver. Probably 15.0 code has this issue resolved and I haven't seen this issue further after the Upgrade.
Thanks,
Sagar

Link-state protocol on blade switches

Similar Messages

Maybe you are looking for