Why is dccd reporting database broken?

On Thu, Jan 19, 2006 at 09:28:28PM -0700, Vernon Schryver wrote:
> 
> If your dcc2 has 2 or 3 GBytes while dcc1 has 4 GBytes or more,
> the version you are using should try to have those sizes.
> By far the best solution is to install the current version (perhaps

I upgraded dcc1 this morning.  During the day, I see that dbclean ran
a few times to repair dcc_db.  It had been doing this before, as well.
The database is now down to about 1.6 GB.

Later, I noticed that an e-mail message from one of our internal
servers had been rejected for bulkiness.  This shouldn't have happened
because the server's IP address is listed in a CIDR block in the
server whitelist.  What could have gone wrong?  I'm bracing myself
for lots of complaints tomorrow is this problem persists.

We have two class B networks that need to be whitelisted.  They're in
a file called localnets.wh that's included by the whitelist file, as
512 /24 subnets.  I thought I could work around this problem by
including it in the whiteclnt file, but then I got:

	dccm[15639]: [ID 702911 mail.error] too many CIDR blocks in line 513 of localnets.wh included from whiteclnt

The limit seems to be 64.  I hope that limit doesn't apply to the
server whitelist?  Can I whitelist those two networks using larger
CIDR blocks?

-- 
-Gary Mills-    -Unix Support-    -U of M Academic Computing and Networking-