Date: Wed, 6 Oct 2004 20:30:01 -0500 From: Dan Nelson <dnelson@allantgroup.com> To: Sean McNeil <sean@mcneil.com> Cc: freebsd-current@freebsd.org Subject: Re: amd sitting on ldaps port Message-ID: <20041007013001.GH3848@dan.emsphone.com> In-Reply-To: <1097102594.1805.4.camel@server> References: <1097095438.1208.7.camel@server> <20041006205954.GB3848@dan.emsphone.com> <1097102594.1805.4.camel@server>
next in thread | previous in thread | raw e-mail | index | archive | help
In the last episode (Oct 06), Sean McNeil said: > On Wed, 2004-10-06 at 13:59, Dan Nelson wrote: > > In the last episode (Oct 06), Sean McNeil said: > > > Looking at /etc/services is states that 636 is for ldaps, but I see that > > > amd is using it: > > > > > > server# sockstat | grep 636 > > > root amd 468 5 tcp4 *:636 *:* > > > > That's just a random port rpcbind assigned to the "amd" rpc service. > > If you reboot I bet it'll bind to a different port. Run "rpcinfo -p > > localhost" to see all the local port numbers assigned to RPC clients. > > OK, but aren't there rules about rpc allowing assigned ports like that? Not as far as I know. I suppose bindresvport() could be changed to walk /etc/services and only use one of the 450 reserved ports not listed. Another alternative is to set the net.inet.ip.portrange.lowlast sysctl a little higher; 700 maybe. 600-1024 is the portrange that has been historically assigned as "local port numbers that root processes can use". -- Dan Nelson dnelson@allantgroup.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041007013001.GH3848>