[10G/100G etherchannel]

Hi, all:
Just curious if on an XR box is possible to bundle a 10G and a 100G interface together. Will XR support it?
Is there a way to have, say a 100G uplink, and if growth is needed use 10G interfaces (several)? I'm thinking I could do this by using ABF (no bundles anywhere) and try to balance traffic based on source. Can anyone offer a different solution that's 100% dynamic?
Thanks,
c.

hi Carlos,
The XR release 4.2.1 introduced the 10x (mixed) bandwidth. This means that a giga ethernet/10 giga ethernet interfaces can be aggregated in a bundle ethernet interface, or a 10 gig/100 giga.
you can find more information in the link below:
http://www.cisco.com/c/en/us/td/docs/routers/crs/software/crs_r4-2/general/release/notes/reln_421crs.html#concept_F9C0AFB0070D4B6191B3EF55161A6982
Regards,
Gregorio Bueno

Similar Messages

10G wan etherchannel bundling between 12410 routers or between CRS-1 router

How etherchannel/bundle ether will behave on the N*10G wan links between the 12410 routers and when running between the 12410 and CRS-1 routers and also when running between 2 CRS-1 routers ??
How it is advantageous compared to the normal CEF based load balancing on these above routers without etherchannel or bundle ether configured on them.
Any suggestions on this ??

Hello Rajesh,
we are speaking of high end routers that are expected to process tens of thousands of flows per second even just differentiating on source and destination ip addresses there should be enough variety in your traffic to use the parallel CEF links in a fair way.
CEF uses an EXOR of ip source and ip destination and an hash seed that changes only when the system is reloaded.
etherchannel can use EXOR of different fields:
MAC SA exor MAC DA
IP SA exor IP DA
L4 SA port exor L4 DA port
but only one of them applies to a type of traffic at a specific time.
Or you expect a few very high volume flows on the links (DR or DB synchronization just to say)
So to answer your question, unless used for connecting two data centers, I don't expect to gain in fairness by using a bundle.
Hope to help
Giuseppe

10G etherchannel between 4510 and 3850 - NO Link

I'm pulling my hair out on a simple issue, and I could use some advice. Thank you, in advance.
I've got a core 4510 switch with 2 12-port 10GE SFP+ modules and my closet has a stack of 3 3850-48p with 2 4-port C3850-NM-2-10G modules.
I'm using 4 SFP-10Gb-SR SFP's to create an ether-channel between these two switches.
The fiber between them is OM4 cable, and OM3 patch cables.
The distance is about 90 Meters.
I've got po8 created on the 4510 through te1/8 and te2/8. On the 3850, i've got po8 created on te1/1/4 and te3/1/4. My problem is that I cannot get a link light at all between these ports. Absolutely nothing indicating when i hook it up. I've tested all the patch cables using loop backs and they all respond. I can even loop back through the complete length of the fiber and i get a link light then also. When i hook them up together, the log on the 4510 starts throwing Rx power low alarms with an Operating value of -29.5 dBm.
I've had Cisco's support go through and verify my port-channel configuration, and they found it to be correct.
I'm starting to wonder if there is larger forces at work here. Am i bumping into a compatibility issue?
Any help would be appreciated.

I've removed the interfaces from the etherchannel and tried cross looping. On the 4510 I get no indication from the LEDs. On the 3850 I get immediate indication on the LEDs.
One might conclude, Cisco support included, that the SFPs on the 4510 side were bad and should be replaced. I hesitate to conclude this because, while members of the etherchannel, i get link lights when looping immediately. Its only when setting the 4510 interfaces back to default and outside the etherchannel that i don't get a response.

Oracle 10g - AMD cpu

Will oracle 10g under windows 2000, run on an amd box?

gtcol wrote:
Hello allhi
Why is it that data pump is extremely slow in RAC environment? A 100G schema takes hours to export in RAC environment where as I can do the same in few minutes in standalone server. there could be zilion reasons why , but in your case there is propably higher resource usage on machines where RAC instance are than on the standalone one.
I am on RHEL 4 update 5, Oracle 10g, 8 cpu Dual-Core AMD Opteron with 32G physical memory. Does it need any tweaking that I'm overlooking here. I appreciate any advise or suggestions.Maybe you could start reading again those data pump documentation you already read, like this one for instance
http://www.oracle.com/technology/products/database/utilities/htdocs/datapump_faq.html
>
Thankscheers
gtcol

LACP using 2x 10G ports not showing 20G?

I currently have etherchannel configured to 2x 10g ports.
Cat_6509#sh run int ten8/1
Building configuration...
Current configuration : 156 bytes
interface TenGigabitEthernet8/1
switchport
switchport mode trunk
channel-group 42 mode desirable
end
Cat_6509#sh run int ten9/1
Building configuration...
Current configuration : 156 bytes
interface TenGigabitEthernet9/1
switchport
switchport mode trunk
channel-group 42 mode desirable
end
Cat_6509#
Cat_6509#sh int t8/1
TenGigabitEthernet8/1 is up, line protocol is up (connected)
Hardware is C6k 10000Mb 802.3, address is fc99.4736.a2f0 (bia fc99.4736.a2f0)
Description: 10 Gig GEC to MADDC (Spanned VLANS)
MTU 1500 bytes, BW 10000000 Kbit, DLY 10 usec,
reliability 255/255, txload 5/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 10Gb/s
Cat_6509#sh int t9/1
TenGigabitEthernet9/1 is up, line protocol is up (connected)
Hardware is C6k 10000Mb 802.3, address is fc99.4736.a238 (bia fc99.4736.a238)
Description: 10 Gig GEC to MADDC (Spanned VLANS)
MTU 1500 bytes, BW 10000000 Kbit, DLY 10 usec,
reliability 255/255, txload 8/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 10Gb/s
Cat_6509#sh int po42
Port-channel42 is up, line protocol is up (connected)
Hardware is EtherChannel, address is fc99.4736.a2f0 (bia fc99.4736.a2f0)
Description: 10 Gig GEC to MADDC (Spanned VLANS)
MTU 1500 bytes, BW 20000000 Kbit, DLY 10 usec,
reliability 255/255, txload 7/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 10Gb/s <<<<<<<<<<<<<<<<<<<<<<< Should this be 20GB if the ports are bundled?
Cat_6509#show etherchannel 42 summary
Flags: D - down P - bundled in port-channel
I - stand-alone s - suspended
H - Hot-standby (LACP only)
R - Layer3 S - Layer2
U - in use N - not in use, no aggregation
f - failed to allocate aggregator
M - not in use, no aggregation due to minimum links not met
m - not in use, port not aggregated due to minimum links not met
u - unsuitable for bundling
d - default port
w - waiting to be aggregated
Number of channel-groups in use: 28
Number of aggregators: 28
Group Port-channel Protocol Ports
------+-------------+-----------+-----------------------------------------------
42 Po42(SU) PAgP Te8/1(P) Te9/1(P)
Last applied Hash Distribution Algorithm: Fixed
Cat_6509#
Thanks
ST

Hi WannaB
Oops you are right the configuration is in PagP. However I have another port channel from the 7706 to Fabric Interconnect which is configured as LACP that is still operating on 10G.
ethernet# show int po30
port-channel30 is up
admin state is up
Hardware: Port-Channel, address: 8480.2da5.c36c (bia 8480.2da5.c36c)
Description: UCS Fabric Interconnect A
MTU 1500 bytes, BW 10000000 Kbit, DLY 10 usec
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, medium is broadcast
Port mode is trunk
full-duplex, 10 Gb/s
ethernet# show port-channel summary
Flags: D - Down P - Up in port-channel (members)
I - Individual H - Hot-standby (LACP only)
s - Suspended r - Module-removed
S - Switched R - Routed
U - Up (port-channel)
M - Not in use. Min-links not met
Group Port- Type Protocol Member Ports
Channel
1 Po1(SD) Eth NONE --
2 Po2(SD) Eth NONE --
3 Po3(SU) Eth LACP Eth1/2(P) Eth2/2(P)
30 Po30(SU) Eth LACP Eth1/5(P) Eth2/5(D)
ethernet# sh run int eth1/5
!Command: show running-config interface Ethernet1/5
!Time: Thu Dec 11 02:43:46 2014
version 6.2(8a)
interface Ethernet1/5
description UCS Fabric Interconnect A
switchport
switchport mode trunk
channel-group 30 mode active
no shutdown
ethernet# sh run int eth2/5
!Command: show running-config interface Ethernet2/5
!Time: Thu Dec 11 02:43:48 2014
version 6.2(8a)
interface Ethernet2/5
description UCS Fabric Interconnect A
switchport
switchport mode trunk
channel-group 30 mode active
no shutdown
ethernet#

Pvlan (promiscuous port) not permitted on etherchannel

This is probably an often-asked question: I've read in the 7K v5.0(2) NX-OS docs, the 6500 12.2SXF/SXH IOS docs and the 5K docs that the pvlan feature cannot co-exist with etherchanneling on the same uplink. This would assumedly include promiscuous trunk ports.
This seems so counter-productive and self-defeating. Especially where the 7K vPC feature is cfg'd requiring LACP etherchannels from the single switch into both 7Ks.
What am I missing here? Is there a work-around short of new dedicated links? The cost of the 10G optics for that approach is prohibitive. Is it possibly on the road-map for some future upgrades?
Maybe I could configure an unconditional port-channel with no protocol, no negotiation? I'm waiting on the vDC licensing to let me direct-connect the lab 5Ks.
Thank you,
Ken Johnson

Required info:
system image file is: bootflash:///n5000-uk9.6.0.2.N2.2.bin
system compile time: 10/4/2013 12:00:00 [10/04/2013 22:23:49]
interface port-channel4001
description SERVER1
switchport mode trunk
switchport access vlan 201
switchport trunk native vlan 201
storm-control broadcast level 1.00
storm-control multicast level 1.00
storm-control unicast level 1.00
vpc 4001
interface Ethernet107/1/47
description SRV1
no lldp transmit
no cdp enable
switchport mode trunk
switchport access vlan 201
switchport trunk native vlan 201
storm-control broadcast level 1.00
storm-control multicast level 1.00
storm-control unicast level 1.00
channel-group 4001 mode active
no shutdown

Suggestions for Oracle 10g Installation using VMware in Windows XP.

Hi All,
I am using Intel(R) Core (TM) 2 Duo CPU P8600 @2.4 Ghz with 1.98 GB RAM with Windows XP with 70 GB disk space. I would like to install a 2 node RAC with ASM and if possible also configure a DG for learning and testing purpose.Please advice if this will work using a VMware . Is RAC VM possible only Linux or with Windows as well ?
I think 11g is out of scope since i don't have sufficient memory.
I am new to VMware and have not done any installations or used it till now.
It would be appreciated if you could help with these doubts.
1. How much memory can be allocated to each instances?If i am installing Linux how much memory is required ?
2. what will be the load on the CPU and will i be able to use it without any resource crunch ?
3.Once memory is allocated to VM, can it be claimed back , is it like if i am not starting the VMware services.. i have the full 2 GB in discretion?
4. Can 10g Standard editions be used for learning purposes?
If the above hardware is not sufficient, please suggest a suitable configuration
Hari
Edited by: Hari2012 on May 27, 2012 12:24 AM

You do not have sufficient hardware - you will likely need 6GB of RAM or more, plus 100G of disk or more and a multiprocessor machine. Each VM will need at least 2GB of RAM, and once all of the VMs/RAC are installed, the system will be pretty busy running 3 OS's (host and two guests) and two databases. If you do not start up the VMs, all of the memory is available for the host to use. 10g is no longer available for download, you will have to use 11g - yes SE edition is sufficient for learning purposes.
Re: Oracle RAC Desktop Setup
HTH
Srini

EtherChannel between stacked VSS Catalyst 4500X and Stacked 2960X.

I can have only one link in the EhterChannel up but the 2nd one goes in to suspended mode.
I've tried recreating channel group on different ports but with the same results. Am I doing something wrong or pagp cannot be transmitted over stacks on both ends (I have quite a few EtherChannels but all to single/non-stacked switches)?
2960X stack config
Version: 15.0 (2r)EX
interface TenGigabitEthernet2/0/2
description HR-Link1
switchport mode trunk
switchport nonegotiate
channel-group 2 mode on
interface TenGigabitEthernet1/0/2
description HR-Link2
switchport mode trunk
switchport nonegotiate
channel-group 2 mode on
4500X Stack (VSS) config
Version: 03.04.00.SG (fc3)
interface TenGigabitEthernet1/1/12
description MO-Link1
switchport mode trunk
switchport nonegotiate
shutdown
channel-group 12 mode on
interface TenGigabitEthernet2/1/12
description MO-Link2
switchport mode trunk
switchport nonegotiate
shutdown
channel-group 12 mode on
Thank you,
T

Thanks for reply!
I have tried desirable mode as well but no luck. I can work on this only during the weekend as downtime is not an option.
These two 10G links were just installed last weekend and one patch cable I had to replace due to complete lack of connectivity (small shop so we no fiber testing equipment).
I did not do a lot of troubleshooting last weekend but I will dig in to that this weekend. I was hoping to get some insight from people who work with this on daily basis.
Post from about a year ago got me worried: https://supportforums.cisco.com/discussion/12072281/4500-x-vss-mec-2960-x-stack
Thanks again.
T

Layer 3 Etherchannel on C2960X

Hello all,
Does Cisco 2960X catalyst switches support Layer 3 Etherchannel ?
I went throught the datasheet, where it mention of supporting Etherchannel technology only.
Can any body confirm it to me please that does it support layer 3 etherchannel or not ?
Following is the exact part number of device from BoQ
WS-C2960X-48FPD-L
Catyalyst 2960-X 48 GigE PoE 740W, 2 x 10G SFP+
Regards,

2960-X only has LAN Base. This has very limited Layer 3 functionality.
2960-XR, however, has IP Lite and can do some routing stuffs. In my opinion, if you want Layer 3 etherchannel, 2960-XR is your choice.

EtherChannel problem on Nexus 7000

Dear NetPro gurus,
One of my customer is trying to setup an EtherChannel (LACP) on a pair of Nexus 7000. However, doesn't matter what we do, the port Eth 1/17 always become suspended. We have tried swapping fiber cables and also swapping SFPs, but no help.
The 1st Nexus 7010 - called 'VIWLRCA'
The 2nd Nexus 7010 - called 'VIWLRCB'
Originally port eth 1/17 are left as 'normal' trunk port, and we can see eth 1/17 shows up fine under 'show interface brief'
viwlrca-PROD# sh run int eth 1/17
interface Ethernet1/17
switchport
switchport mode trunk
udld disable
no shutdown
viwlrca-PROD# sh run int eth 1/18
interface Ethernet1/18
switchport
switchport mode trunk
udld disable
channel-group 20 mode active
no shutdown
viwlrca-PROD# sh int brief
Ethernet      VLAN    Type Mode   Status Reason                   Speed     Port
Interface                                                                    Ch #
Eth1/17       1       eth trunk up      none                        10G(S) --
Eth1/18       1       eth trunk up      none                        10G(S) 20
Eth1/19       --      eth routed down    SFP not inserted           auto(S) --
Eth1/20       --      eth routed down    SFP not inserted           auto(S) --
Eth1/21       --      eth routed down    Administratively down      auto(S) --
Eth1/22       --      eth routed down    Administratively down      auto(S) --
Eth1/23       --      eth routed down    Administratively down      auto(S) --
Eth1/24       --      eth routed down    Administratively down      auto(S) --
Eth2/25       --      eth routed down    Administratively down      auto(D) --
Eth2/26       --      eth routed down    Administratively down      auto(D) --
Eth2/27       --      eth routed down    SFP not inserted           auto(D) --
Eth2/28       --      eth routed down    SFP not inserted           auto(D) --
Eth2/29       --      eth routed down    SFP not inserted           auto(D) --
Eth2/30       --      eth routed down    SFP not inserted           auto(D) --
Eth2/31       --      eth routed down    SFP not inserted           auto(D) --
Eth2/32       --      eth routed down    SFP not inserted           auto(D) --
viwlrca-PROD#
But as soon as I add the Eth 1/17 back onto PortChannel 20
The Eth 1/17 becomes "Suspended" straight away
viwlrca-PROD# sh int brief
Ethernet      VLAN    Type Mode   Status Reason                   Speed     Por
t
Interface                                                                    Ch
Eth1/17       1       eth trunk down    suspended                  auto(S) 20
Eth1/18       1       eth trunk up      none                        10G(S) 20
Eth1/19       --      eth routed down    SFP not inserted           auto(S) --
Eth1/20       --      eth routed down    SFP not inserted           auto(S) --
Eth1/21       --      eth routed down    Administratively down      auto(S) --
Eth1/22       --      eth routed down    Administratively down      auto(S) --
Eth1/23       --      eth routed down    Administratively down      auto(S) --
Eth1/24       --      eth routed down    Administratively down      auto(S) --
Eth2/25       --      eth routed down    Administratively down      auto(D) --
Eth2/26       --      eth routed down    Administratively down      auto(D) --
Eth2/27       --      eth routed down    SFP not inserted           auto(D) --
Eth2/28       --      eth routed down    SFP not inserted           auto(D) --
Eth2/29       --      eth routed down    SFP not inserted           auto(D) --
Eth2/30       --      eth routed down    SFP not inserted           auto(D) --
Eth2/31       --      eth routed down    SFP not inserted           auto(D) --
Eth2/32       --      eth routed down    SFP not inserted           auto(D) --
viwlrca-PROD#
viwlrca-PROD# sh port-channel summary
Flags: D - Down        P - Up in port-channel (members)
        I - Individual H - Hot-standby (LACP only)
        s - Suspended   r - Module-removed
        S - Switched    R - Routed
        U - Up (port-channel)
        M - Not in use. Min-links not met
Group Port-       Type     Protocol Member Ports
      Channel
20    Po20(SU)    Eth      LACP      Eth1/17(s)   Eth1/18(P)
viwlrca-PROD#
Config on Primary Nexus:-
viwlrca-PROD# sh run
!Command: show running-config
!Time: Tue Mar 22 06:04:26 2011
version 5.1(1a)
hostname PROD
cfs eth distribute
feature udld
feature interface-vlan
feature lacp
feature vpc
feature vtp
username admin password 5 $1$pkJaKHZW$Sx4wpDG5xXYkD.QfDk/Cg. role vdc-admin
no ip domain-lookup
ip domain-name vfc.com
crypto key param rsa label viwlrca-PROD.vfc.com modulus 2048
snmp-server user admin vdc-admin auth md5 0x05f7328e3b39a70be09abc3056ec2819 pri
v 0x05f7328e3b39a70be09abc3056ec2819 localizedkey
vrf context management
spanning-tree pathcost method long
spanning-tree port type edge bpduguard default
spanning-tree loopguard default
spanning-tree vlan 1-3967,4048-4093 priority 4096
interface Vlan1
interface Vlan161
ip address 172.30.161.2/24
interface Vlan162
ip address 172.30.162.2/24
interface Vlan163
ip address 172.30.163.2/24
interface Vlan164
ip address 172.30.164.2/24
interface Vlan165
ip address 172.30.165.2/24
interface Vlan190
ip address 172.30.190.2/24
interface port-channel20
switchport
switchport mode trunk
interface Ethernet1/17
switchport
switchport mode trunk
udld disable
channel-group 20 mode active
no shutdown
interface Ethernet1/18
switchport
switchport mode trunk
udld disable
channel-group 20 mode active
no shutdown
interface Ethernet1/19
interface Ethernet1/20
interface Ethernet1/21
interface Ethernet1/22
interface Ethernet1/23
interface Ethernet1/24
interface Ethernet2/25
interface Ethernet2/26
interface Ethernet2/27
interface Ethernet2/28
interface Ethernet2/29
interface Ethernet2/30
interface Ethernet2/31
interface Ethernet2/32
interface Ethernet2/33
interface Ethernet2/34
interface Ethernet2/35
interface Ethernet2/36
interface Ethernet3/25
interface Ethernet3/26
interface Ethernet3/27
interface Ethernet3/28
interface Ethernet3/29
interface Ethernet3/30
interface Ethernet3/31
interface Ethernet3/32
interface Ethernet3/33
interface Ethernet3/34
interface Ethernet3/35
interface Ethernet3/36
line vty
viwlrca-PROD#
Config for Secondary Nexus 7000
VIWLRCB-PROD# sh run
!Command: show running-config
!Time: Tue Mar 22 09:19:22 2011
version 5.1(1a)
hostname PROD
cfs eth distribute
feature interface-vlan
feature lacp
feature vpc
feature vtp
username admin password 5 $1$Lc486EOm$EtKhZWuxGjWWokfeuUsMk. role vdc-admin
no ip domain-lookup
ip domain-name vfc.com
crypto key param rsa label VIWLRCB-PROD.vfc.com modulus 2048
snmp-server user admin vdc-admin auth md5 0xeb607b54234985ed6740c5fdbb8d84c6 pri
v 0xeb607b54234985ed6740c5fdbb8d84c6 localizedkey
vrf context management
spanning-tree pathcost method long
spanning-tree port type edge bpduguard default
spanning-tree loopguard default
spanning-tree vlan 1-3967,4048-4093 priority 8192
interface Vlan1
interface port-channel20
switchport
switchport mode trunk
interface Ethernet1/17
switchport
switchport mode trunk
channel-group 20 mode active
no shutdown
interface Ethernet1/18
switchport
switchport mode trunk
channel-group 20 mode active
no shutdown
interface Ethernet1/19
interface Ethernet1/20
interface Ethernet1/21
interface Ethernet1/22
interface Ethernet1/23
interface Ethernet1/24
interface Ethernet2/25
interface Ethernet2/26
interface Ethernet2/27
interface Ethernet2/28
interface Ethernet2/29
interface Ethernet2/30
interface Ethernet2/31
interface Ethernet2/32
interface Ethernet2/33
interface Ethernet2/34
interface Ethernet2/35
interface Ethernet2/36
interface Ethernet3/25
interface Ethernet3/26
interface Ethernet3/27
interface Ethernet3/28
interface Ethernet3/29
interface Ethernet3/30
interface Ethernet3/31
interface Ethernet3/32
interface Ethernet3/33
interface Ethernet3/34
interface Ethernet3/35
interface Ethernet3/36
line vty
VIWLRCB-PROD#
Cheers,
Hunt

Quick troubleshoot:
Default all interfaces in newly created port-channel as well as the port-channel interface, then delete port-channel interface. Recreate port-channel without the LACP protocol:
interface e1/17,e1/18
switchport
channel-group 20 mode on
no shutdown
exit
interface port-channel20
switchport
switchport mode trunk
no shutdown
exit
show port-channel summ
show int trunk
HTH,
Sean

What Do I Need to Accomplish 10G at 450 feet?

Will multi-mode be able to accomplish this or will I need single-mode?

Below is the data sheet for SFP+ transceiver modules. You can reach 220M/660FT on standard OM1 fiber (62.5um) using a 10GB-LRM module. If you are installing new fiber, as was said earlier, an OM3/4 fiber will easily reach your 450FT at 10G.
http://www.cisco.com/en/US/prod/collateral/modules/ps5455/data_sheet_c78-455693.html
If you are installing new multimode, using OM4 will enable you to scale to 100G using the same fiber at those distances (see link below).
http://www.cablinginstall.com/articles/2012/02/commscope-demos-om4.html

Adding another link with etherchannel-PAGP

Hi all,
I have a 10g trunk link between my two 4510 coreswitches. Now the utilization on this link is shooting up to 9.5gb which is forcing us to put another 10g link between these two switches.
We are planning to configure Etherchannel-PAGP. There should be any sort of downtime on the current link while adding another link to it.
How do i add another link with etherchannel -PAGP with out having downtime on the current link? could you please guide me on this?
Thanks
Shibu

Shibu,
Jon is 100% correct, that since this is not already and Etherchannel, there will be downtime, but it should be minimal. Now, what you could do, is depending on your physical topology, you should connect these two switches together with a GigbitEthernet link in the meatnime, it will be blocked by spanning tree (if it's a L2 port), since the 10G link will be give priority by default. Then you could afford downtime on the 10 gig link, while you create an etherchannel, If these two links are L2, you could configure UplinkFast, so the switchover is VERY fast.
I've done this same setup before when making changes, but it all depends on if you can acutally physically do that. If not, you will have some downtime involved.
You can also, just configure all this on notepad, and copy and paste.

EtherChannel over WDM MUX on 7600

Dear All,
I am trying to increase bandwidth between my 7600s running as PE/P
routers with 1G line cards terminating dark fibre links.
I am exploring the following two options:
Option I
- replace the 1G lines cards with 10G on the 7600s
- terminate the dark fibre links on the 10G line cards
Option II
- terminate the dark fibre links on passive CWDM MUX
- EtherChannel multiple 1G interfaces over CWDM MUX
My main concern is about the efficiency of different load-balancing
algorithm offered in the EtherChannel technology in terms of maximum
achievable bandwidth.
Also, is there any issues of running Etherchannel over passive CWDM MUX ?
I will be really grateful for any comments or inputs on this.
Regards,
Zahid

hi zahid
I feel if you are conservative on spending lines you can you add on one more 1G port on both the sides and start doing load balancing but keeping the future in mind and also considering the traffic requirements you better think off going on for 10G line card.
What are the services being offered to your clients using this network ?
How many such boxes are required to be assessed and requires an upgrade ?
do you also have some kinda traffic projection done so that you can take up a concrete decision.
regds

Etherchannel link selection

Hi all,
Question: By default, how does etherchannel decide which link it will use when sending/receiving ? I've bundled two 10g links (i am not load-balancing at the moment) and I'm trying to determine which link at any given time it will use because I understand it doesn't split the traffic equally. Any help would be appreciated.
Thanks

Hi,
Look at http://www.cisco.com/en/US/tech/tk389/tk213/technologies_tech_note09186a0080094714.shtml
Very interesting.
I hope this helps.
Best regards.
Massimiliano.

Etherchannel vs Stack

Dear all,
I want to know about the exact difference between Etherchannel and Stacking with some configuration
Which is recommended on which situation ?

Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
Normally you would want both.
I.e. Etherchannel with links to different stacked switch ports.
PS:
As Leo notes, unfortunately, many (all?) of Cisco's stackable have rather shallow buffers. However, this doesn't guarantee that they are totally unacceptable to be used with servers, but they can be. It really depends on your expected traffic patterns. (It also depends what ports you use. On the 3750X, uplink ports have more buffer memory to share than the copper edge ports. Also the later stackable switches often support 10g ports too.)

[10G/100G etherchannel]

Similar Messages

Maybe you are looking for