NLB on 2 different switch running HSRP

Hi,
we have implemented NLB on the SGI Server running IRIX OS, it works fine. the connectivity is 2 NIC connected on 3750-1(ruuning HSRP),now if i connect 1 of NIC in to the other switch 3750-2(running HSRP) will the NLB workz & will i get the 2Gbps throughput, according to me, i don't think so.

Think, it should work, check out the following explanation to avoid single point of failure :
Create a subnet that spans two switches and home half of the NLB cluster to each switch. The objective is typically to create a LAN with no single points of failure. To this end you will also need two router uplinks and have one switch uplink to one router. You also need to run a redundant routing protocol such as HSRP for failover of the router links. Finally you will need two cross-over links between the switches (otherwise the cross-over cable is a single point of failure). Each switch has two paths off the network: one via the router it is linked to (which is used by default) and one through the other switch. In this configuration, the loss of a router (or link to a router) causes the affected switch to use the cross link to ship its traffic off the LAN. The loss of a switch cuts the cluster capacity in half.

Similar Messages

How to set Different applications running on the same computer to use different Gateways?

Dear All,
I would like to have different applications running on the same computer to use different Gateways (both gateways are in the same subnet).
Is it possible in Windows7 with / without a third-party utility?
Current scenario & requirements:
I have two internet connections. Both are in the same subnet (192.168.1.0/24) and connected to the same LAN switch
Gateway for the first connection is 192.168.1.1
Gateway for the second connection is 192.168.1.2
I need to set my chrome browser & google talk to use the gateway 192.168.1.1
I need to set my IE & skype to use the gateway 192.168.1.2
Is it possible? Answers are appreciated!!
Regards,
Anees

Hi,
I agree with Bill. Gateway is essentially an IP address for one network to other networks. For example, you have two networks: network A and network B, to make the two networks communicate with each other, you must go through Gateway. That is, as long
as you set Gateway, then you can make the two networks communicate with each other. The ip address for Gateway is the ip of device with routing enabled. One host machine can own multiple gateways.
Andy Altmann
TechNet Community Support

Running HSRP between N5k 's and vPC between N2k and server

Dear All,
I have the following existing setup, i would like to run HSRP between N5k1 and N5k2. Also would lilke to run vPC between N2k1,N2k2 to the server farm which are dual homed. Please find the diagram attached.
Please advise the step by step process along with the commands to be entered. Thanks for your help

Hi,
If you don't have many vlans, there is no need to load balance. Make one of the 5ks the primary for HSRP and the other one stand-by.
here is good link on how to configure it. Make sure to run HSRP version 2. This way you can mach the HSRP group IDs with your vlan IDs. Also give a higher priority to the active 5k and enable preemption.
http://www.cisco.com/c/en/us/td/docs/switches/datacenter/nexus5500/sw/unicast/7_x/cisco_n5500_layer3_ucast_cfg_rel_6x/l3_hsrp.html
HTH

Differenet VLAN's on different switches

In short, I have two SRW 2024 switches connected together. The first one goes to the router, ASA 5510 (supports inter-vlan routing), on the native VLAN and the second one is trunked on port 12 to the first one. I have been doing lots of research and have found ambiguous answers to my question. My question is can I have different VLAN's on different switches? Meaning can I have VLAN 10 on the first switch and VLAN 20 on the second but not have VLAN 20 on the first and VLAN 10 on the second? So far, I have heard that I HAVE to have identical VLAN's on both switches in order for them to be able to talk to each other and I have also heard that that is not true because I can setup routes on my router to make them talk to eachother and get on the internet... Does anyone have a definitive answer to my question? I am totally pulling my hair out on this one...

Well, reading this post now makes me wonder if we have the same understanding.
What do you mean with "have VLAN 10 on the first switch" etc.? What do you mean with "have"?
If you connect the ASA to switch 1, and switch 1 to switch 2. If you use VLAN 20 on the second switch and you want to give VLAN 20 access to the internet through the ASA switch 1 must know about the existence of VLAN 20. The switch will only forward frames for VLANs it knows of. If VLAN 20 does not exist on switch 1 VLAN 20 cannot pass through switch 1.
If you use VLAN 10 only on switch 1 and not on switch 2, you could omit VLAN 10 on the second switch as no VLAN 10 traffic has to go to switch 2. However, generally it is better to have all VLANs on both switches because it makes management easier.
This has nothing to do with routing, though, as the SRWs are only layer 2 switches. Routing allows you to connect a VLAN to another VLAN or LAN or internet.
Think of a single VLAN like a normal switched LAN. Different VLANs are just like different, physically separated LANs.
If you want to allow traffic between these separated LANs you'll need a router which routes traffic between them.
A managed switch with VLANs allows you to run these different LANs on the same hardware, making the individual VLAN assignments configurable.
A port on a managed switch usually is in on of two modes:
* access mode: an access mode port connects to a normal device like a desktop, printer, or similar. An access mode port can be member of a single VLAN only, i.e. you have to decide to which VLAN it is supposed to belong to. In your case, you configure an access mode port for either VLAN 10 or VLAN 20.
With a single switch things are clear now: some ports are VLAN 10 and some ports are VLAN 20. VLAN 10 can talk to each other. VLAN 20 can talk to each other. No traffic passes between VLAN 10 and VLAN 20.
Of course, now you want to connect this switch to some other network devices, in particular the second SRW because you need additional ports or you have an additional location. And there is the ASA which provides internet access for these VLANs.
* trunk mode: This is where trunk mode comes in. A trunk mode port can carry multiple VLANs on a single port. This is done using 802.1q tags. 802.1q tagged ethernet frames have an additional field for the VLAN to which the frame belongs to. With this, a switch can send frames for VLAN 10 and VLAN 20 through a single port to another switch or router. Each frame sent is tagged with 10 or 20 depending on which VLAN the frame belongs to. The receiver will accept each frame and assign it to the corresponding VLAN on the receiving side. This way the receiving switch or router is able to keep those VLANs strictly separated.
So let's say you want two VLANs 10 & 20 in your network. You would create VLANs 10 & 20 on your ASA and both SRWs. (Create only means that the device knows this VLAN exists and is able to handle traffic for this VLAN). You would configure LAN port 1 of your ASA as trunk with members VLAN 10 & 20. You configure port 1 & 24 of your first SRW in trunk mode with members VLAN 10 & 20. You configure port 1 of your second SRW in trunk mode with members VLAN 10 & 20. Now you wire port 1 of your ASA to port 1 of your first SRW. Then you wire port 24 of your first SRW to port 1 of your second SRW.
This creates the VLAN trunk through your network. Traffic in both VLANs can travel through this trunk between the switches and to the ASA and from there, if properly routed, into the internet.
In a very simple scenario you configure all remaining port in access mode. For each access mode port you define whether this port belongs to VLAN 10 or 20. If port 2 is in access mode and member of VLAN 10 then the device connected to port 2 is in VLAN 10.
You are completely free how to assign the VLANs. If you assign ports 2-24 on switch2 to VLAN 20 and ports 2-23 on switch 1 to VLAN 10 this is fine. In this case, you could reduce the VLAN configuration a little by not creating VLAN 10 on the switch 2 and not adding VLAN 10 on the trunk ports connecting switch1 and 2. However, as mentioned before, I would recommend not to do so. If at some point you decide to have a port in VLAN 10 on the second switch everything would already be set up if you created the VLAN 10 on the second switch and added it to the trunk.
You must create all VLANs on your ASA and the first switch in your case. VLAN 20 traffic has to travel through switch 1 (even if there is no end device connected to VLAN 20 on switch 1). Thus, VLAN 20 must exist on switch 1 and the trunk between the ASA, switch 1 and switch2 must carry VLAN 20 for traffic to pass through. If VLAN 20 did not exist on switch 1 no VLAN 20 traffic could travel trough switch 1.
As you only have two switches you will only have a few VLANs which you should be able to create in the beginning. If you really have to add a new VLAN you have to touch both switches and the ASA. But with some planning, it should not be necessary to add VLANs later. With two 24 port switches you won't have more then 48 VLANs anyway.
Your VLANs "terminate" on the ASA. The ASA is a 802.1q capable router. You can trunk your VLANs to the ASA. The ASA allows you to define gateway interfaces in each VLAN which will operate as gateways for each VLAN. Through that VLANs can talk to the internet. You can also configure the ASA to allow inter-vlan-routing, i.e. let specific traffic be routed from one VLAN to the other. For instance, if you have a printer in one VLAN you could allow traffic to this printer from the other VLAN while still blocking any attempt to access other devices on the other VLAN.

How to extend VLAN in 3rd and 4th 6513 switch with HSRP

Hi presently in One Data centre 02 Nos. of 6513 is configured in HSRP. Due to space constraint some servers need to be installed in other room and there 02 Nos of 6509 switches need to be configured in HSRP again.
One Data Vlan needs to be extended in expanded room. What should be the best practice for configuration in 6509 switches in HSRP. Please suggest.

Just as #1 and #2 are connected with (I would assume) trunking ports, similarly add connections to #3 and #4 Catalyst 6k. Define the layer 2 VLANs in the 6509 and assign access ports to those VLANs as required.
Personally I don't see much value to adding L3 interfaces (SVIs) for the VLANs (and adding them to the HSRP groups) on the 6509s. It depends in part on how your other connectivity into them is setup. If everything leaving the server VLAN goes to users, WAN, Internet etc. via the original pair of 6513s then keep the 6509s simple with only layer 2 VLANs.

Make ports on two different switches look like physically wired

I have two different switches in two different buildings and I need to make the ports look like there is a physical wire between them.
I have two disjoint networks that can talk to each other through a firewall but I need to be able to easily pass multicast traffic. I have a single PC in one building on my network (network A) and it needs to look like it is wired into the other network (network B) that resides in another building. My biggest problem is that the subnets on Network B (10.129.x.x/24) overlap with the management vlan (10.x.x.x/8) on my network. I have attempted to put all of the ports on Network A and Network B on the same vlan, however I could get no traffic to pass and I believe this is a problem with the overlap. Any help would be appreciated.
Sent from Cisco Technical Support iPad App

So is there a current connection between switch 1 and switch 2 and what vlan is it in or is it a trunk link.
If you want the PC to be in vlan 4 then -
1) the connection between switch 1 and switch 2 needs to be in vlan 4 at both ends or if it is a trunk link then it needs to allow vlan 4
2) the connection between switch 1 and the 6500 for both ends needs to be in vlan 4
then it should work because the PC has a direct path in vlan 4 to the 6500. The fact that there is an overlap in IP addressing doesn't matter because the PC is not routing off switch 2 but off the 6500.
Not sure how this relates to multicast traffic though. Do you mean there are multicast sources in vlan B and you need the PC to be able to receive them.
Jon

Port aggregation on different switches?

Hello all :)
I got a server PC with two fast ethernet NICs (NICS do support port aggregation)
I know you can connect two 100 Mbps ports of server to two fast ethernet ports of a switch, this way you get one 200 Mbps link. If one of the ports or one of the links fail, other link keeps working.
this is a vey nice redundancy.
My question is;
is it possible to bundle two fast ethernet ports of server to two different switches? (first ethernet port of PC to switch 1 and second ethernet port of PC to switch 2. switch 1 and switch 2 are connected each other with an uplink)
if possible, what is the CLI command for that?
thank you very much for helping.

Hello,
AFAIK, and unless something has changed very recently, the only way to link two (or more) switches together in a single Etherchannel is in a stack configuration.
Nortel appears to have a feature called Split MultiLink Trunking, which does multichassis trunking, you might want to have a look at the document below:
Layer 2 Trunking Availability Strategies using SMLT and DMLT
http://www.informit.com/articles/article.asp?p=169544&seqNum=3&rl=1
Regards,
GP

GBICS show up differently on different switches

Has anyone ever seen the exact same GBICs show up differently on a show interface status command on two different switches? I have a 4507 on one end of the link and a 3750 on the other end of the link. A show interface status on the 4507 indicates that a 1000baseLX is installed in the port. A show interface status on the 3750 is showing that a 1000baseLH SFP is installed on the port, but they are the exact same GBICs. It is throwing me off because I am having UDLD issues so I am trying to find a problem with the fiber or GBICs.

I'm curious, have you tried to swap the GBICs between the devices and see if the description follows the device?
The "LX" designation was for the original 5km reach specified standard and the "LX/LH" is for the 10km reach version. It might be possible that one of the GBICs is the old standard.
Are you operating this link over MMF (OM-?) or SMF and what is its distance?

Physical redundancy (CSS connected on two different switches)

Hello,
Is it possible for redundancy reasons to connect a CSS to two different switches (like we do for access switches) ?
I know that one option is to configure both interfaces in the same vlan and use the spanning-tree for the redundancy. However, this will not be transparent for the end-user due to the slow convergence of the spanning-tree (uplinkfast or rapid-pvst is probably not supported).
Are there any other options ? One interface active, the other backup ?
Thanks in advance for your reply.
Gaetan

Gaetan,
for the CSS we suggest to be connected to only 1 switch and if the CSS detect a failure with the switch, force failover to the redundant css.
If you have ASR configured, failover will be stateful.
Gilles.

Itunes 7- can two different ipods run on same itunes?

i have a 30gb ipod video and my husband has a nano. i have itunes on my computer twice in different user accounts for our ipods but since installing the new itunes 7 my husbands itunes will not open it gives me an error message. i uninstalled his itunes and reinstalled but then i couldnt open my itunes. i reinstalled itunes 7 and now i am back to my original problem.
Can i have my ipod and my husbands nano run from the same itunes? will my library be added to his nano and will his nano library be added to my itunes? please help this is frustrating me.

tean5289,
This article addresses using multiple iPods on the same computer.
http://docs.info.apple.com/article.html?artnum=300432
Method Two in the article is the method my wife and I have used in the past. I had mine set up to sync my playlists, plus the recently added playlist. That way if she added any music, I had 2 weeks that it would show up on my iPod, and give me time to move the song to one of my own playlists.
i have turned the fast switch for the user accounts
off. when i log into my husbands user account and
open his itunes it still tells me that it cannot be
opened an unknown error has occured.
anyone know a soloution? hence my original question
can ipod video and nano run from one itunes library?

KE30 Report differs when run at two different times

Hello,
I have an issue where a KE30 report is executed for a closed period occuring more than 6 months ago. When the report was run at the beginning of November, it had particular values for figures. When the report was run at the beginning of November, for the same exact month, some of the values are the same while others are different.
I drilled into the line items and sorted by Created On date to see if any documents were created from November to December, but unfortunately no documents were created.
I'm not sure where else to look, is there a historical view that I can drill into then do a comparison to determine where the difference is? Any help would be greatly appreciated.

Hi Johnnie
I assume the important bit is the bit within the second fix:
"New Forecast" = Consumption * "Bias %";
The way you have your fix statement set up it is looking at level 0 members in your Appliances and Batteries dimensions (I'm reading your post as if they are separate dimensions). I assume your issue is that the one piece of data is at level 0 interesections of appliances and batteries but the other piece is not possibly at level 0 of one dimension and level 1 or higher of the other, Bias% being the one at the higher level?
If that is the case then I think you need to look at a way of getting from the level 0 member to the correct intersection point(s).
Have you had a look through the technical reference at functions like @PARENTVAL, @ANCESTVAL or the multidimensional options for the same @MDPARENTVAL and @MDANCESTVAL, it sounds like your formula needs to be something along the following lines
"New Forecast" = Consumption * @PARENTVAL(Appliances, "Bias%");
What the above would do differently to your original code would be to get the Bias % value from the parent of the current level 0 member of the appliance dimension being calculated. While this might not be exactly what your requirement is hopefully it will steer you in the right direction.
Hope this helps
Stuart

Directly Connected Networks... Why can't I get to IPs on different Switches but in a common subnet???

Hi,
I've set up a pair of switches in a DMZ type environment that hangs off of our DMZ firewall.
The firewall has a static route to the supernet hosted on the switches via switch interface 10.1.191.12
   The static route is: 10.1.160.0 /19 via 10.1.191.12
   The firewall has the local interface address 10.1.191.11 /28
   Switches 01 and 02 have a floating HSRP address of 10.1.191.12 /28
   Switch 01 has a vlan interface address of 10.1.191.13 /28
   Switch 02 has a vlan interface address of 10.1.191.14 /28
The firewall is completely open at the moment. From my desktop on the other side of the firewall from the 10.1.160.0 /19 network I can happily ping all the addresses mentioned above. i.e. From my desktop (IP 10.0.33.50) I can ping 10.1.191.11, 12, 13, 14.
I've also configured another network across the switch pair: 10.1.160.0 /24
This network has a VLAN interface set up on each switch as follows:
   Switch 01 has a vlan interface address of 10.1.160.251 /24
   Switch 02 has a vlan interface address of 10.1.160.252 /24
From each switch I can ping the 10.1.160.25x address thus subnet connectivity is fine across the two switches.
From my desktop on the other side of the firewall from the 10.1.160.0 /19 network I can happily ping the 10.1.160.251 address.
NOW the Problem....
From my desktop on the other side of the firewall from the 10.1.160.0 /19 network I CAN NOT ping the 10.1.160.252 address.
i.e. I can only ping IP addresses hosted by the switch that the next hop for the firewall points to (i.e. 10.1.191.12)
I don't understand why I can only ping IP addresses on other subnets when the IP address is local to the switch that is hosting the next hop IP that the firewall is set to use!!
Please can somebody explain what the problem is here!?
* Please let me know if further clarification is required on this question!!?

Hi,
Just thought I'd update this as I figured out what the problem was.
I was using a default route instead of a static route to get back to the internal company network via the firewall. As soon as I put a static route on instead and enabled IP routing the routing started to work fine for all subnets this side of the firewall.

OATM : number of commands differs between runs

In testing OATM, first run generated 27000 commands.
Another test run generated many fewer commands.
Is this normal?
Are there something different to be done using OATM for 11gR2?
thx

Hi,
In testing OATM, first run generated 27000 commands.
Another test run generated many fewer commands.
Is this normal?Please elaborate more.
Are there something different to be done using OATM for 11gR2?There should be no difference across the different database versions.
Thanks,
Hussein

WAAS and Routers running HSRP

I am just in the process of deploying WAAS across our network. I have found that when the WAN router that is running WCCP has been configured for HSRP and is the active router i see the following
If I try and telnet to the HSRP standby address it does not connect and I get the following error message :
Jan 30 18:04:36.930 AU-Summ: %IP-3-LOOPPAK: Looping packet detected and dropped
src=172.16.33.254, dst=172.16.9.251, hl=20, tl=72, prot=47, sport=0, dport=0
in=GigabitEthernet0/0.9, nexthop=172.16.9.251, out=GigabitEthernet0/0.9
options=none
-Process= "IP Input", ipl= 0, pid= 82
-Traceback= 0x611B33C0 0x616A0FD4 0x616A11E0 0x616A1C14 0x616A1FE4 0x6168D09C 0x
6168E880 0x6168C384 0x6168C648 0x6168C704 0x6168C8A8
f I telnet to the actual router address then ok.
is this a Bug?

just seen other question about HSRP and tried the
"egress-method negotiated-return intercept-method wccp' command but made no difference
see router interface configuration below
Note if no HSRP then no problem
interface Loopback0
ip address 172.16.33.254 255.255.255.255
interface GigabitEthernet0/0
description trunk to Cat 6509
bandwidth 100000
no ip address
duplex auto
speed auto
media-type rj45
negotiation auto
ntp broadcast
interface GigabitEthernet0/0.1
encapsulation dot1Q 1 native
ip address 172.16.1.100 255.255.255.0
no snmp trap link-status
interface GigabitEthernet0/0.2
encapsulation dot1Q 2
ip address 172.16.2.250 255.255.255.0
ip wccp 61 redirect in
ip flow egress
no snmp trap link-status
interface GigabitEthernet0/0.3
encapsulation dot1Q 3
ip address 172.16.3.250 255.255.255.0
ip wccp 61 redirect in
no snmp trap link-status
interface GigabitEthernet0/0.5
encapsulation dot1Q 5
ip address 172.16.5.252 255.255.255.0
ip wccp 61 redirect in
ip pim sparse-mode
no snmp trap link-status
standby 5 ip 172.16.5.254
standby 5 preempt
service-policy input prec
interface GigabitEthernet0/0.6
encapsulation dot1Q 6
ip address 172.16.6.252 255.255.255.0
ip helper-address 172.16.5.228
ip accounting output-packets
ip wccp 61 redirect in
ip pim sparse-mode
no snmp trap link-status
standby 6 ip 172.16.6.254
standby 6 preempt
service-policy input prec
interface GigabitEthernet0/0.7
encapsulation dot1Q 7
ip address 172.16.7.252 255.255.255.0
ip helper-address 172.16.5.228
ip wccp 61 redirect in
ip pim sparse-mode
no snmp trap link-status
standby 7 ip 172.16.7.254
standby 7 priority 105
standby 7 preempt
service-policy input prec
interface GigabitEthernet0/0.9
description testlan for WAAS
encapsulation dot1Q 9
ip address 172.16.9.254 255.255.255.0
ip wccp redirect exclude in
no snmp trap link-status

Java application behaving differently when run from netbeans

Hi,
When i run my program from netbeans it receives packets from ip addresses which are not in my subnet
but when i run it from outside netbeans it doesn't receives packets from ip addresses which are not in my subnet.
Can anybody explains why my application is showing different behavior when run from netbeans and when run independently?
Thanks.

Bingo! My script started the rmiregistry and continued on to start the servers. Apparently, it was too quick. I put a 10 second delay after the rmiregistry startup and all is working as it should.
Thanks for your help. You get the Duke dollars if I can figure out how that works.

NLB on 2 different switch running HSRP

Similar Messages

Maybe you are looking for