Cisco WAN :: IP SLA And Object Tracking For Default Route On Nexus 7010
Mar 18, 2013
We have a Nexus 7010 running version 6.1(2).
I'd like to use IP SLAs and object tracking to define static routes for specific source/destination traffic across some WAN links we have. I've done this in IOS and it's worked fantastically, but I've not found where/how to do this on the Nexus 7010 platform (or any Nexus platform) as of yet. I could have sworn that this was going to be introduced in the 6.x code? Below is an example of how we do this in the IOS world:
track 11 ip sla 1 reachability
delay down 15 up 15
ip sla 1
[Code]....
Esentially this gives us the option of using a "failover" default route. I've attached a basic diagram to explain what we are trying to do with IP SLAs and object checking. The tracking should be configured against an SLA that uses icmp and the static routes should be configured against the tracking.
I've got two Nexus 7010's running HSRP north bound to a pair of ASA's, and BGP south bound to four 6509's. Is it possible to advertise default route to BGP neighbor (or prefer it via MED), only if the node is HSRP-active?
Essentially the goal is to create symmetry for inbound/outbound traffic. Only way I can think of so far is via an EEM script, so that when it sees HSRP go active via syslog, it would kick off an action to remove ASN prepend, or reduce MED, and the opposite if HSRP goes standby.
I cannot receive any OSPF route from Nexus to ASR1002 even they are both OSPF neighbour. I have attached the config for both, Both Nexus and ASR part of Area0.
Config ASR1002#sh ip ospf neighbor Neighbor ID Pri State Dead Time Address Interface10.165.117.12 1 FULL/BDR 00:00:35 10.231.175.226 GigabitEthernet0/0/0
I am working on Nexus 7010 with NX-OS 5.1.5. I have to delete the static route 10.10.0.0/16 via 10.16.0.21. [code] I try to remove the route with the command "no ip route 10.10.0.0/16 10.16.0.21" and I have the message below % Route not deleted, it does not exist..I don't understand why I have this message because the static route exist.
Ok I didn't setup my OSPF on my 7010. Today I found out that any static route I put into my 7010 gets sent into to my MPLS network. My 6509's you have to "Tag" the static rout for this to happen. Was under the impression the same was necessary for the 7010 or at least it had to "match" an access list. How can I fix the below so that by default all static routes are not resdistributed into OSPF? [CODE]...
I'm currently working on a plan to migrate our 6500's over to our new 7010's. At the time of the migration I want to tighten up our OSPF design and configure OSPF for "passive-interface default" then allow only those interfaces that should have OSPF neighbors to send the hellos. The issue is that the command is not showing up under the OSPF process. What's even more interesting is that the Nexus 5.x Unicast Routing Configuration Guide shows that the "passive-interface default" command should be an option to enter.
I'm currently running version 5.1(4) (though looking to upgrade to 5.2 during my migration testing). I would rather configure the passive-interface via the routing process versus having to enter it on every interface.
I was configuring route tracking at a client with several sites to route across GRE tunnels and being able to detect a failure of the main site. To my surprise when configuring a 2800 series router (after sucessfully configuring a 1800 series on the same infrastructure), a 2821 with IOS 12.4(24)T2 IPbase, the commands for ip sla object tracking don't show up.The feature navigator says the router supports this, but it just won't take the commands (also tried older versions of the commands such as "ip sla monitor.." and "rtr .." to no avail).
I am using a bunch of Cisco 1721 routers for my T1 lines. We recently purchased Digi cell modems as a backup for the T1. On configuring vrrp to work on both devices I discovered that IOS 12.3(6c) does not support the "vrrp track" feature. After reviewing the Cisco Feature Navigator I could not see an IOS that will support the vrrp object tracking. Is that correct? The routers have T1 WIC's installed. If it does work what is the latest IOS that will work on this end of life product?
I've enabled antispoof on all interfaces on asa 5510.If you start a traceroute to a network on the default route, everything works, since replies comes to an interface with route 0.0.0.0/0 defined.If you start a tracer route to a network that is NOT on the default route (let's assume coporate MPLS), you only get response from first carrier router, the other are discarded because of anti spoof violation.
I have ICMP inspection and icmp-error inspection enabled.
I am using two 7609 router in setup, HSRP enable on both router, and both have WAN connectivity to different PE end router we want to enable hsrp feature ip route traking, in case primary link goes down (Because of any reason link bgp failure, PE device issue).
Trying to implement PBR in N7K? I found that there is not track mechanism can use in "set next-hop ip", so if the next-hop is unreachable that the route might be died.
i have a couple of nexus 7010 (5(2)3a) connected to a checkpoint in HA(active/active)I have now to configure multicast mac..i found this commmand: [code]
I am trying to set up my ASA5510 the fail over of ISP when it can't ping three different IP. I create three different tracking to three different IP using sla monitor & track rtr. But when I do
the last route will replace the previous two and only the last route command takes effect.Is there anyway I can set up the fail over to ISP2 only when it can't ping three different IP from ISP1?
I am trying to determine why hosts off our Nexus 7010s are being picked up in UT. Since LMS 4.0.1, UT should be supported on these devices.When adding the Nexus devices to DCR, provide the netadmin SNMP RO credential.When other SNMP RO credential is provided, user tracking will not collect end host data.I think I have this setup correctly as the device center test passes when cehcking snmp ro credentials.Our 7010s are running NX-OS 5.0(3) - earlier than the recommended version - might that cause issues? We are not using VRFs other than the default and management.
Here is my snmp section:
sh run | sec snmp ip access-list copp-system-acl-snmp 10 permit udp any any eq snmp 10 permit udp any any eq snmp 20 permit udp any any eq snmptrap
I am trying to implement static route tracking on a Catalyst 3560G ( WS-C3560G-48PS, IOS version 12.2(35)SE5 and SW image C3560-IPBASE-M). The configuration is as follows:
track 101 rtr 1 reachability ! rtr 1 type echo protocol ipIcmpEcho 10.199.101.2 rtr schedule 1 life forever start-time now ! IP address 10.199.101.2 is reachable via ICMP (its the next-hop router).
The static routes configured are the following:
ip route 0.0.0.0 0.0.0.0 10.199.101.2 track 101 ip route 0.0.0.0 0.0.0.0 10.200.52.1 20
But only the secondary route(ip route 0.0.0.0 0.0.0.0 10.200.52.1 20) its being installed on the routing table by the switch.
There are times the Nexus CPU goes high around 70% but its not a constant occurance. Is there something to worry. Quite hard to find out which process caused it as it happens very brief. [code]
This is regarding Nexus core switch 7010. We are already running two nexus 7K with ten Nexus 5k. Currently we are going to add two new Nexus 5k in our DC. In the 7K we already running two VDC's.
In our LAN network design, we have two Nexus 7010 switches on the core connected via vPC. Then LAN access switches are directly connected to the Core Nexus switches via regular port channels on the 3750's and vPC on Nexus. The core Nexus switches will be linked to an existing LAN network and the applications will be progressively migrated from the old to the new network.In this scenario, three VDCs are planned to be configured on the Nexus - One for the Interconnect (and WAN at a later stage), one for the LAN/local services and one for the building facilities/local services.
We recently replaced our core switch from a non-cisco vendor with a Nexus 7010. With our old core switch, I had the ability to log changes to the ARP table. So if there was a dhcp conflict or a vMotion event, it would show up in the "show log" output. I've not found a way to do that with the Nexus switch - or at least no way to view the log. I have the command: logging level arp 6
I have to upgrade a Nexus 7010 with dual Sup engines from 4.2(4) to 5.2 and am hoping it could be an ISSU. We are fine with an outage window.To upgrade from 4.2(4) to 5.2(5) I'll have to do a multi hop upgrade from 4.2(4) - 4.2(6) - 5.2(5) and each hop would take 40-60 minutes.do I spend 40-60 minutes for each hop, or just do a disruptive upgrade straight from 4.2(4) to 5.2(5)? Like I said, we are fine with an outage window.
I have two Nexus 7010 in the data center. I'm unable to poll SNMP data from one of the NEXUS 7010s. The other Nexus 7010 is working fine. I have compared the SNMP configurations, and they are identical. When I do "show snmp" on the non-working switch, I have SNMP packets in "Unknown Context name", not sure why. I have done show vdc and it matches the working switch output. Here is output of show snmp from the non-working switch: 133 SNMP packets input 0 Bad SNMP versions 0 Unknown community name 0 Illegal operation for community name supplied 0 Encoding errors 0 Number of requested variables 0 Number of altered variables 0 Get-request PDUs 0 Get-next PDUs 0 Set-request PDUs 0 No such name PDU 0 Bad value PDU 0 Read Only PDU 0 General errors 0 Get Responses 133 Unknown Context name0 SNMP packets output 0 Trap PDU 0 Too big errors 0 No such name errors 0 Bad values errors 0 General errors 0 Get Requests 0 Get Next Requests 0 Set Requests 0 Get Responses 0 Silent drops.
I just deployed a nexus 7010 switch at a server farm. after deployment, it was notices that there are surges in latency across the network. The default gateway was then moved to the nexus, with this pinging from an host on the same subnet there is intermitent burst is latency
NEXUS>>>>>Server Ping of about 80ms and sometimes even times out. To me, this is strange. NX-OS Version is 5-02a
I'm looking to see if it is possible to run a vPC between to vDC's on a single 7010? We have a Production setup that runs dual 7010's with vPC's between the chassis but in our lab we only have a single 7010 with a 32 port 10gig module. I was thinking that maybe we could create 4 vDC's on the 7010 and run a vPC between the vDC's.
How to get a summary of netflow statistics on NX-OS? On IOS you could do sh ip cache flow which would show what I need? Can't find a similar command on the Nexus Platform.
We will install a new Supervisor Engine in our Nexus 7010.One Supervisor Engine is already installed an 1Year old.So the Problem is that both Supervisor Engines may have different NX-OS version.Could this lead to a problem?Does the installed Supervisor Engine "udate" the newer Supervisor engine?
i have a big problem because i configure a vlans with vrf and HSRP but, when i do "show hsrp brief", dont show this interfaces and, i can ping virtual IP. it seems hsrp dont work.
SWSERVSCAMILO_N7010_A# interface Vlan405 description smsc-fwatlas1 no shutdown [Code] ....
We have a couple of Nexus 7010's split into Core and Distribution VDCs. MGMT0 interfaces on each of the Nexus VDC's (including the Admin VDC) are configured with different IP address, but on the same subnet i.e 10.10.10.1/24 for admin, 10.10.10.2/24 for Core and 10.10.10.3/24 for Distribution. The MGMT 0 physical port on each Nexus is connected to a physical gig port on a 3750 X switch, and the 3750X has uplinks back to the Nexus configured for vPC.
When i ssh to the VDC MGMT0 IPs from the 3750X, i can access each of these VDCs without any problems. But if i enable routing on each of these links(OSPF) and advertise it to the WAN, i cannot see these routes advertised and also cannot see any of these routes in the local routing table.Just wondering if i have to enable these links on a VLAN and then advertise it to the WAN..But if this the case, VLANs cannot be created on the Admin(default VDC).
