balanced pool problem??

[project722]

Thanks Bob, as always, very insightful!

Cheers!

On Fri, Sep 14, 2018 at 10:15 AM Bob Harold <rharolde at umich.edu> wrote:

>
> On Thu, Sep 13, 2018 at 11:29 AM project722 <project722 at gmail.com> wrote:
>
>> Hey guys I just established a new failover relationship between 2 rhel 6
>> servers running dhcp 4.1.1. I created the relationship about 10 hours ago.
>> I'm doing some checking and I'm seeing the following in the logs from each
>> server:
>>
>> server A - (primary)
>> Sep 13 09:45:54 serverA dhcpd: balanced pool 55ab409d8790 <poolname>
>> total 884  free 151  backup 150  lts 0  max-misbal 45
>>
>> Server B - (secondary)
>> Sep 13 09:45:53 serverB dhcpd: balanced pool 5654c47474a0 <poolname>
>> total 884  free 157  backup 144  lts -6  max-misbal 45
>>
>> The ultimate plan is to decom server A and sync B with a new Primary once
>> I verify all lease data from A has synced to B. But, my concern here is the
>> discrepency I see in the "free" and "backup" numbers from each server. If
>> the pools are balanced, shouldn't these numbers be the same?
>>
>
> That looks normal to me.
> I am running:
> DHCP Server 4.3.6-P1
>
> And here are some sample log lines from the failover pair:
>
> Sep 14 10:34:31 adonis3 dhcpd: balancing pool 206cf30 35.0.120.0/23 total
> 470 free 157 backup 236 lts -39 max-own (+/-)39
> Sep 14 10:24:11 adonis4 dhcpd: balancing pool 1a88870 35.0.120.0/23 total
> 470 free 165 backup 238 lts 36 max-own (+/-)40
> Sep 14 10:34:31 adonis3 dhcpd: balancing pool 2097490 35.0.122.0/23 total
> 469 free 184 backup 156 lts 14 max-own (+/-)34
> Sep 14 10:24:11 adonis4 dhcpd: balancing pool 1ab2dd0 35.0.122.0/23 total
> 469 free 183 backup 166 lts -8 max-own (+/-)35
> Sep 14 10:34:31 adonis3 dhcpd: balancing pool 20c08d0 35.0.124.0/23 total
> 501 free 230 backup 231 lts 0 max-own (+/-)46
> Sep 14 10:24:11 adonis4 dhcpd: balancing pool 1adc210 35.0.124.0/23 total
> 501 free 228 backup 236 lts 4 max-own (+/-)46
>
> As long as they are balanced within "max-own", it should be ok.
>
> What you don't want to see is "requesting peer rebalance" lasting more
> than a few minutes at a time:
>
> Aug  8 07:59:09 adonis8 dhcpd: balanced pool 54334b0 35.0.118.0/25  total
> 107  free 83  backup 19  lts -32  max-misbal 15
> Aug  8 07:59:09 adonis8 dhcpd: balancing pool 54334b0 35.0.118.0/25
> total 107  free 83  backup 19  lts -32  max-own (+/-)10  (requesting peer
> rebalance!)
> Aug  8 07:59:09 adonis7 dhcpd: balanced pool 51e0d50 35.0.118.0/25  total
> 107  free 41  backup 61  lts -10  max-misbal 15
> Aug  8 07:59:09 adonis7 dhcpd: balancing pool 51e0d50 35.0.118.0/25
> total 107  free 41  backup 61  lts -10  max-own (+/-)10
>
> I have seen it stuck in that state, and one of the servers run out of
> IP's.  Had to remove failover and put it back, in order to clear the
> condition.  Have not figured out what causes it.
>
> --
> Bob Harold
>
> _______________________________________________
> 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/20180914/a78411b6/attachment.html>