From mboxrd@z Thu Jan 1 00:00:00 1970 From: Lucio De Re To: 9fans@cse.psu.edu Subject: Re: [9fans] ndb/csquery: what is dns is not up? Message-ID: <20030423152542.O19261@cackle.proxima.alt.za> References: <20030423145118.J19261@cackle.proxima.alt.za> <0d7357fe22689e4f5ae741260582305f@plan9.bell-labs.com> <20030423151624.M19261@cackle.proxima.alt.za> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20030423151624.M19261@cackle.proxima.alt.za>; from Lucio De Re on Wed, Apr 23, 2003 at 03:16:25PM +0200 Date: Wed, 23 Apr 2003 15:25:43 +0200 Topicbox-Message-UUID: 96bfd79e-eacb-11e9-9e20-41e7f4b1d025 On Wed, Apr 23, 2003 at 03:16:25PM +0200, Lucio De Re wrote: > > Sadly, it seems to me that factotum attempts to connect directly to > sources.cs.bell-labs.com, according to tcpdump run on the remote > host. The diagnostic output from factotum indicates the expected > timeout caused by the various filters in the path. > Nopes, I stand corrected. This is what eventually pops out (I fear it's not the most recent version of factotum): 21: failure cs: can't translate address 21: write 141 in phase CNeedTreq yields phase CNeedTreq: failure cs: can't translate address 21: write 141 in phase CRelay yields phase CRelay: failure cs: can't translate address I don't know what address is involved, not what resolver service is being invoked. I also assumed that factotum itself did not communicate with the remote authenticator, but that it delegated all communication to the client or server! This does not seem to be the case here. > If I spawn a new factotum (it really needs an option to be removable, > by the way), I believe it fails to identify/resolve the auth server > associated with the auth domain. I have no idea quite how to take > care of that problem. Ideally, I'd create a /net/ndb entry, but > that's not possible with sshnet serving /net. > I don't want to do this right now, as spawning a new factotum tends to mess the namespace up enough for me to prefer to reboot. The alternative workstation I can use for testing is more or less a pristine installation and I don't know how much the absence of sensible configuration information actually cripples networking. ++L