dhcpd upgrade 3.0.6 -> 3.1.2p1
perl-list at network1.net
Thu Aug 27 16:58:34 UTC 2009
I'm not sure I can recreate the situation now. We have reverted back
and are not having the problem in the test lab. The test lab had only
a shortly pre-existing 3.0.6 lease database, however. The field units
had a lease database that has existed through several revisions from
3.0.1 -> 3.0.6. I assume that it may have had something to do with
something in the lease database.
On Aug 27, 2009, at 12:28 , David W. Hankins wrote:
> On Thu, Aug 27, 2009 at 08:49:55AM -0400, Darren wrote:
>> max-lease-ownership 1;
>> max-lease-misbalance 2;
>> min-balance 30;
>> max-balance 300;
> a more aggressive lease rebalance profile should reduce situations
> where one peer or the other exhaust their free leases...it shouldn't
> make it worse. in the worst case the '1' and '2' values cause the
> integer math to round down to zero and you get an even (or even/odd)
> split of the free leases enforced on every pool rebalance run, every
> 30-300 seconds.
> basically this is a tradeoff; finer-grained pool balance in exchange
> for greater cpu utilization.
> so i'm open to the possibility you found some kind of bug here, but
> i am thinking some other variable (even one outside of your control)
> changed at the same time...
> and yeah, they do need to be configured on both servers. i think the
> manual only states which ones explicitly need to be configured on the
> primary. at the time this feature was engineered, we were tracking
> the standards track failover draft, and didn't feel we should extend
> the protocol to carry our server-specific feature configurations.
> David W. Hankins "If you don't do it right the first time,
> Software Engineer you'll just have to do it again."
> Internet Systems Consortium, Inc. -- Jack T. Hankins
> dhcp-users mailing list
> dhcp-users at lists.isc.org
More information about the dhcp-users