On 4/12/2010 5:25 PM, Atro Tossavainen wrote:
> Jeffrey,
> 
>> actually it is because that server is reporting multiple addresses:
>>
>>   Server( 128.214.88.114 10.0.0.3 172.16.0.1 172.17.0.1 172.18.0.1 )
>>
>> several of which are lower than 128.214.58.174.   What are these other
>> interface addresses are do you expect them to be used for ubik
>> synchronization?
> 
> Private addresses for purposes other than AFS.
> 
> I believe I am using NetRestrict to avoid the servers from picking up
> these:
> 
> sun4x_58 # cat /usr/afs/local/NetRestrict
> M 10.255.255.255
> M 172.255.255.255
> M 192.168.255.255
> 
> I don't expect to see the RFC1918 addresses anywhere in connection
> with AFS.  The NetRestrict file is unaltered as of Feb 13, 2009 and
> has been essentially identical for years.
> 
> (I remember needing to raise a bug with IBM over wildcard support.
> PMR-73077... late 2004/early 2005.  Not that it says anything to anybody
> outside IBM, probably.  Merely using "255" wasn't enough, it needed
> adding "M" in front for wildcards to work and I think this was and is
> undocumented.)
> 
> I have not had any database issues for as long as the other sun4x_58
> host was the other database server.

The OpenAFS NetRestrict documentation does not mention the use of a
preceding 'M'.

  http://docs.openafs.org/Reference/5/NetRestrict.html

I suspect the change IBM implemented was never passed on to OpenAFS
and as far as I can tell from the source its presence will invalidate
the address and prevent it from being used as a filter.

Jeffrey Altman








Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply via email to