Not all subnets are balancing

Greg Sloop <gregs@sloop.net> gregs at sloop.net
Fri Oct 10 03:53:23 UTC 2014


I seem to recall there being some serious issues with peer fail-failover on
4.1.x versions. 4.2.x works well for us. [A search of the list might be
helpful to confirm that my recollection is correct.]

That's not to say there isn't an issue with your config. [I didn't see
anything with my weak-dhcp-foo, quick look.]
 On Oct 9, 2014 1:30 PM, "Norman Elton" <normelton at gmail.com> wrote:

> I should add that this is ISC DHCP 4.1.1-P1, packaged by RedHat
> (dhcp-4.1.1-38.P1.el6.x86_64)
>
>
> Norman
>
> On Thu, Oct 9, 2014 at 4:28 PM, Norman Elton <normelton at gmail.com> wrote:
> > I've got two DHCP servers configured for failover:
> >
> > failover peer "wm-dhcp-01-02" {
> >   primary;
> >   address 1.2.3.4;
> >   port 647;
> >   peer address 1.2.3.5;
> >   peer port 647;
> >   max-response-delay 30;
> >   max-unacked-updates 10;
> >   load balance max seconds 3;
> >   mclt 1800;
> >   split 128;
> >   min-balance 60;
> >   max-balance 300;
> > }
> >
> > I've noticed that our DHCP log file shows helpful information when
> > subnets get balanced:
> >
> > balancing pool 7f9da82e7960 INIT-WM-HALL  total 240  free 88  backup
> > 96  lts -4  max-own (+/-)18
> >
> > But this doesn't seem consistent. That is, there are some subnets that
> > balance, others never show up in the log. Presumably there's some
> > small difference between the two, but I can't figure it out. Here's a
> > subnet that is getting balanced:
> >
> > group {
> >     option subnet-mask              255.255.255.0;
> >     option domain-name-servers      128.239.29.9, 128.239.20.9;
> >     option domain-name              "wm.edu";
> >     default-lease-time              3600; # 1 hour
> >     max-lease-time                  3600; # 1 hour
> >
> >     shared-network INIT-WM-HALL {
> >         subnet 10.45.170.0 netmask 255.255.255.0 {
> >             option routers                  10.45.170.1;
> >
> >             pool {
> >                 failover peer   "wm-dhcp-01-02";
> >                 range           10.45.170.11        10.45.170.250;
> >             }
> >         }
> >     }
> > }
> >
> > And here's a seemingly similar subnet that is never showing up in the
> log file:
> >
> > shared-network ADMIN-NET {
> >     subnet 128.239.96.0 netmask 255.255.224.0 {
> >         option subnet-mask              255.255.224.0;
> >         option routers                  128.239.102.1;
> >         option domain-name-servers      128.239.29.9, 128.239.20.9;
> >         option domain-name              "wm.edu";
> >         deny                            bootp;
> >         default-lease-time              432000; # 5 days
> >         max-lease-time                  432000; # 5 days
> >
> >         pool {
> >             failover peer   "wm-dhcp-01-02";
> >             range           128.239.102.10  128.239.102.254;
> >         }
> >     }
> > }
> >
> > Am I missing something? Is there some reason why the first would
> > rebalance and the second not?
> >
> > Thanks,
> >
> > Norman Elton
> _______________________________________________
> dhcp-users mailing list
> dhcp-users at lists.isc.org
> https://lists.isc.org/mailman/listinfo/dhcp-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.isc.org/pipermail/dhcp-users/attachments/20141009/fa34ed57/attachment.html>


More information about the dhcp-users mailing list