Keith!!! Excellent video. Just love it. Searched so many sites but you nailed it. Some people are complaining about your quickness, i believe its your beauty not weaknes. Awesome Sir.
Hi keith..I honestly believe you are one of the best i've seen around. Thanks so much for all the hastle you take to make our cisco path little easier.
Great work Keith. My 2600 routers and layer 3 switches only have HSRP. It's great to see GLBP in action and I'm glad to learn the configuration is so similar to HSRP.
Keith Barker is Champ always unique person in technology and very kind and help full out of the way he helps really very fantastic person helping people by his great knowledge
Using virtual MAC addresses allows for moving things around, transparently. For example, if a router fails, one of the other forwarders can jump in and start supporting that virtual address (in addition to its own and other virtual addresses it is assigned.
Hi keith, Again good video with good analogy. I just like to learn things from you even if i know the technology. Your knowledge gives me very deep understanding. Good work keep it up. Regards, Praveen
Good point. The weight controls the number of clients getting the forwarders address in an ARP reply, but the actual load depends on how much traffic is sent by which devices. I don't know of an GLBP feature that can/will balance on true load.
Hi Keith, I just hit HSRP, VRRP, & GLBP in my CCNP SWITCH studies. You explain so well. Jeremy C. is great but you make it sound so easy. I see now that I can use GNS 3 for some L3Switching labs. Obviously no VTP or 802.1x Switch stuff. I'll get by. Th Boson Netsim 8.0 looks awesome. The Demo is very limited so I can't really tell. Thanks as always. Evan
I've thought about that myself. According to the Cisco Press CCNP text, for HSRP the determining factor when all priorities are equal is who has the highest ip address on the HSRP enabled interface. I suspect that this is also the case for GLBP. The Cisco Press books don't specify what happens in this equal priorities condition for GLBP or VRRP. Source: CCNA, and CCNP in training.
Keith, Thanks for the video. I'm a little confused as to what the "track 1 rtr 1 reachability" command does. Could you explain? Also, with the SLA probes, I see that everytime fa0/1 on R1 was shut, R3 took over as the AVG, even though it had a priority of 90, compared to R2, which was set to 100. It seems to me that weighted has precedence over priority. Am I correct in saying this or did I misunderstand? thank you.
Great video. I was thinking GLBP would allow me to load balance or load share two ISPs (two cable modems). What would you recommend doing such? I have (2) 3750's, and (2) 2950 switches.
Hi Keith; I really enjoy your videos. Aren't you using a switch between the PC and the GLBP routers, and between the GLBP routers and R4? I am asking because, in the video you stated that if you were using a switch, one should remember to use portfast to accelerate the convergence time. Thanks for the great work of helping a lot of people.
One Question please, do i need to directly connect the two GLBP hosts or they will be communicating through layer two swtiches. Please reply as i am going to configure it by this week.
Hi, great video and explanation. Thank you for the time spent! I am having trouble building this topology in GNS3. Can you help me out? How can i do the big connection that combines the routers on both sides? Thanks.
Keith at can you give some more detials on "Track 1 rtr1 rechability". how does that link to the IP SLA statements and to the Glbp 1 weighting track 1 decrement 5 statement.
Hi keith, great video. I just have one question, i understand what portfast does but why would i disable portfast on the interfaces that is going to my router? (19:45)
In the video I said that if PortFast is NOT enabled, it may take up to 30 seconds for traditional spanning tree to allow forwarding (on the switchport) after an interface comes up.
Great video Keith. I do have a question. If we have 2 routers and the WAN link for one that is being tracked goes down and the wieght wasnt decremented to zero(was decremented by 10 only and the wieght of both routers initially was 100 (default) , the other router's virtual mac would be forwarded.Would it be forwarded at all times or for every 100 times the other still gets 90 and the best way to avoid this is to decrement to zero?Thank you
With HSRP there is only 1 Active router for each HSRP group at a time (winner take all). If you want to do proportional and/or weight based load balancing, Gateway Load Balancing Protocol (GLBP) also from Cisco, has that option.
I do understand that GLBP loadbalances but my question is if one of the uplinks went down and traffic should not go through this circuit , would it still go if its weight wasnt down to zero upon tracking?
wael A. Depends on the thresholds set. If the value falls below the lower threshold, it will cause the router to give up its role as a virtual forwarder. Here is some additional information: www.cisco.com/en/US/docs/ios/12_2t/12_2t15/feature/guide/ft_glbp.html
Hi Keith, Great explanation. Can we do the following? 1-create 2 glbp groups: group 1 for Vlan 2 and group 2 for vlan 3 Like this we can load balance the traffic in both forward and return flow. 2- Will this cause any arp issues. 3- if this solution is feasible, do we still need ip sla tracking ? 4- I see that shifting from active to standby , we are using weight. What about the priority parameter you configured at the beginning, for what it is used? Awaiting your help reply :) Bernard
Hi Bernard- Yes we can use GLBP in both directions, with a group for each VLAN and associated subnets. ARP shouldn't be a problem, and we would still want to do tracking. Priority is for who will be the AVG, and the weight can be used to influence the load balancing (or how many assignments 1 forwarder is given compared to another). Cheers, Keith
great video sir. But when I am trying to configure GLBP on two routers in GNS3 none of them are showing their group members, Each one is behaving as if it is the only member of the group
Hi keith, I have a small doubt in VRRP. We create groups in VRRP along with virtual IP addresses, but we no where configure it in the client system as default gateway as we know that VRRP uses physical interfaces as their gateways. What is the use of creating multiple groups with multiple virtual IP for each group. Correct me if I am wrong in framing my question. Thanks, Rajiv R
Really like the video, it has thot me a lot.my quiz is that- i have 3 internet links (of different ISP) that are coming to cisco 2901 router. I have an EHWIC card tat i have connected the 5AP for users. 2internet links goes to router G0/0 & G0/1 iterface.1 other link goes to the EHWIC 1port. is it posible to failover & loadbalance this links in this one router????plis help
Greetings, Could you also track an interfaces via the "track object" (global config) command and track the line-protocol of the interface instead of using SLA?
Keith, I hope you are doing well, Suppose that many computers to the left of VLAN2 are pinging to R4->fa0/0, We know the traffic is being distributed by R1->fa0/1, R2->fa0/1 and R3->fa0/1 respectively in normal conditions; but suddenly the links R1->fa0/1 and R2->fa0/1 were came down. Will be the traffic redirected to R3->fa0/1 to reach the R4->fa0/0 interface?? Thanks in advance for your support/comments.
We can add tracking so if there is a failure on the right side, the GLBP router with the failed link can step down, and another router will service the MAC address so the customer will still have a path. That is one option.
Keith, I've already added tracking in to the 3 routers (R1, R2 and R3); but when many PCs on the left are pinging to R4->fa0/0 only one part preserve connectivity and the remaining lost connectivity. I think that just add tracking is not enough, how the router with the failed link can step down to another GLBP router?? Best regards and thanks.
Julaos Cisar The tracking could be combined with weight, so that the weight is reduced to 0 for the router with the failed link. www.cisco.com/en/US/docs/ios/12_2t/12_2t15/feature/guide/ft_glbp.html
Keith, So when the weigh is reduced to 0 for the router with the failed link, this path is dropped and the others two continue being active so that they can route the traffic, isn't it? Thanks.
Julaos Cisar From the link I posted: GLBP Gateway Weighting and Tracking GLBP uses a weighting scheme to determine the forwarding capacity of each router in the GLBP group. The weighting assigned to a router in the GLBP group determines whether it will forward packets and, if so, the proportion of hosts in the LAN for which it will forward packets. Thresholds can be set to disable forwarding when the weighting falls below a certain value, and when it rises above another threshold, forwarding is automatically reenabled. The GLBP group weighting can be automatically adjusted by tracking the state of an interface within the router. If a tracked interface goes down, the GLBP group weighting is reduced by a specified value. Different interfaces can be tracked to decrement the GLBP weighting by varying amounts.
Hi , I have two router R1 and R2 R1 bandwidth is 100 Mbps and R2 's speed is 50 Mbps , my load balancing is not happening on the routers,Plz help to configur GLBP so that I can used my both router as Active active and utilize the max bandwidth , shashi
You will want to configure the "weight" that can proportionally distribute the load. www.cisco.com/en/US/docs/ios/12_2t/12_2t15/feature/guide/ft_glbp.html#wp1027129
Why does glbp use virtual macs to hand out to hosts. why cant it just hand out the actual macs of the router interface ?? sorry if its a dumb question.
