tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: NetBSD 8.0 dead socket



On 11/08/2018 20:21, Roy Marples wrote:
On 11/08/2018 20:11, Emmanuel Dreyfus wrote:
On Sat, Aug 11, 2018 at 03:40:42PM +0100, Roy Marples wrote:
What is the state of the address as reported by ifconfig(8)?
If it's tentative, duplicated or detatched this is expected behaviour now.

It is reported as active.

Can you test rebuilding BIND with TUNE_LARGE set in src/external/bsd/bind/include/config.h? It should be unset by default setting it and will increase other parameters as well. I don't know if this is desirable for NetBSD BIND in base system and/or pkgsrc.

Upstream issue report and suggested patch:
https://gitlab.isc.org/isc-projects/bind9/issues/462

I've had a private email from someone more knowledgable about BIND that setting TUNE_LARGE is a bad idea.

christos@ has already committed this to -current here:
http://cvsweb.netbsd.org/bsdweb.cgi/src/external/bsd/bind/dist/lib/isc/unix/socket.c.diff?r1=1.22&r2=1.23&only_with_tag=MAIN&f=h

and for older bind here:
http://cvsweb.netbsd.org/bsdweb.cgi/src/external/mpl/bind/dist/lib/isc/unix/socket.c.diff?r1=1.2&r2=1.3&only_with_tag=MAIN&f=h

We can pull the latter up to -8 shortly.

Roy


Home | Main Index | Thread Index | Old Index