First hop redundancy protocol

tecnodog7tecnodog7 Posts: 129Member
Sorry for noobie question but just went through FHRP and I had a questions.
1) Why would one ever use HSR over GLBP in a real life environment? Since can give you redundancy and also give you some LOAD balancing flexibility. I mean i know most big organization are running f5 or some load balance only designed for that but in a smaller organization doesn't it make sense?

Comments

  • RouteMyPacketRouteMyPacket Posts: 1,104Member
    tecnodog7 wrote: »
    Sorry for noobie question but just went through FHRP and I had a questions.
    1) Why would one ever use HSR over GLBP in a real life environment? Since can give you redundancy and also give you some LOAD balancing flexibility. I mean i know most big organization are running f5 or some load balance only designed for that but in a smaller organization doesn't it make sense?

    So what are the differences between HSRP and GLBP?

    With that said, I have never run into anyone using GLBP, ever!
    Modularity and Design Simplicity:

    Think of the 2:00 a.m. test—if you were awakened in the
    middle of the night because of a network problem and had to figure out the
    traffic flows in your network while you were half asleep, could you do it?
  • tecnodog7tecnodog7 Posts: 129Member
    From what i just learned the difference between
    HSRP and VPPR is it's almost exactly the same except the fact that in VRRP you can assign the interface IP address as the phantom(virtual) IP address

    With GLBP, It gives you the flexibility to create a group and add routers in that group and have one Manager Router AGR and that can assign the ARP request and if you suppose have 3 different routers and 3 different ARP request coming in it will give 1st ARP to 1st router mac address 2nd ARP to 2nd router mac address and so on
    Added to that if there is a fail over the Second becomes the new manager and also take all request for the fallen brother(first router) and it's own.

    This way you can assign how often it can accept request 40% 1st router 20% second router 30% third router and so on.

    This is my understanding. Please correct me if i'm wrong.
    So based on this I would think, it would be better to use GLBP over HSRP no?
  • OfWolfAndManOfWolfAndMan Posts: 923Member ■■□□□□□□□□
    tecnodog7 wrote: »
    So based on this I would think, it would be better to use GLBP over HSRP no?
    You could say that. However, there is an advantage to HSRP as well: Everyone that has ever managed an FHRP usually deals with HSRP, meaning they know the config in their sleep. I have only ever dealt with HSRP and VRRP (Only did VRRP because they believed everyone should be going the vendor neutral route. For now :D).
    :study:Reading: Lab Books, Ansible Documentation, Python Cookbook 2018 Goals: More Ansible/Python work for Automation, IPSpace Automation Course [X], Build Jenkins Framework for Network Automation []
  • tecnodog7tecnodog7 Posts: 129Member
    Everyone that has ever managed an FHRP usually deals with HSRP, meaning they know the config in their sleep.

    I mean the config for HSRP isn't that complicated just the standby command. And yes the VRRP makes sense for vendor netural router (but since i'm doing my cisco certification) why would we ever want to use anything else but cisco. Since cisco is so awesome (sarcasm) icon_lol.gif

    The only thing that I can think about why people wouldn't use a GLBP is, if your a big enough of an organization then you have MACHO BEEFY F5 load balancers sitting in the front doing the job of LOAD BALANCING and you wouldn't need a router to do that and it should just ROUTE.

    But for a small organization who aren't rolling in money wouldn't it just make sense to use GLBP over HSRP. I'm just curious.

    PS - I am loving this FHRP so i'm going to have ALOT of fun tonight with lab (especially after doing STP, my brain hurts because of STP. Literally had a dream about DR and RP icon_sad.gif)
  • OfWolfAndManOfWolfAndMan Posts: 923Member ■■□□□□□□□□
    Typically, FHRPs should be configured at the distribution layer (From dist to core, best practice is to use a routing protocol. EIGRP and OSPF both support ECMP). F5 load balancers wouldn't really be used in that situation.

    However, if it's a small company, they may implement a two-layer hierarchial model and not have justification for redundant links everywhere between the top to bottom of the hierarchy due to budget constraints, depending on where they put their money (Which isn't always in the right place :D. Just look at how the crooks spending my tax dollars every year act). While all these cool features make life easier, they also cost elsewhere in other aspects.
    :study:Reading: Lab Books, Ansible Documentation, Python Cookbook 2018 Goals: More Ansible/Python work for Automation, IPSpace Automation Course [X], Build Jenkins Framework for Network Automation []
  • tecnodog7tecnodog7 Posts: 129Member
    Oh ok, yes I forget we don't live in an ideal world so blah.
Sign In or Register to comment.