Date: Thu, 16 Oct 1997 21:53:37 -0000 (GMT) From: Chris Dillon <cdillon@tri-lakes.net> To: itojun@itojun.org Cc: stable@FreeBSD.ORG, ports@FreeBSD.ORG Subject: Re: Problem with Hylafax Message-ID: <XFMail.971016225733.cdillon@tri-lakes.net> In-Reply-To: <22272.877052840@coconut.itojun.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 17-Oct-97 itojun@itojun.org wrote: > > Hello, thanks for the detailed report, > >>I've had a problem with Hylafax for a little while, though I can't >remember >>for the life of me when it started (don't use it often enough). hfaxd >>throws a signal-11 at me every time it starts. I run 2.2-stable with >the >>last make world (cvsup right before) on October 9'th. My kernel >>configuration has remained the same before/after the problems (pretty >sure, >>since I haven't changed it in a while). >(snip) >>So.. I have no clue whats going on.. (novice C programmer to begin with, >>not to mention I have no clue about c++). > > It looks that your hostname (configured by /bin/hostname) does not > have correct entry in DNS database. gethostbyname() fails and > in InetFaxServer.c++ code touches incorrect memory region. > Is it the case? > I'll try submitting a report to the implementers of hylafax, so that > there'll be proper error handling. > >itojun This is correct, however, i don't have a DNS server setup on this machine, nor does my ISP have any clue in hell about DNS, (try and reverse my usual dialup address, 207.3.81.149, using any DNS server in the world... then use one of theirs, such as 207.3.81.5... I've bugged them about this for ages since it prevents me from connecting to sites who require reverse lookups to work) I DO have a hostname set with /bin/hostname (wyze.tri-lakes.net). This, however, was not a problem previously. Hylafax worked fine some time ago even with my 'bogus' hostname. Someone did suggest that bash was the culprit, and so I am reinstalling Hylafax (already removed it again.. might as well test a patch while I'm at it) with its dependancy for bash1 changed to bash2. Hmmm.. no luck with bash2 there. Still does a SIGSEGV.. Now to see if maybe I can fix that problem and supply a patch. :-) I could do: if (hp != NULL) memcpy(&in, hp->h_addr, sizeof (in)); a check just like the line before it, but that will probably cause something ELSE to fail.. :-( I'll look into it a bit deeper, but I notice you just notified the HylaFax team at SGI about it. --- Chris Dillon --- cdillon@tri-lakes.net --- Powered by FreeBSD, the best free OS on the planet ---- (http://www.freebsd.org)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.971016225733.cdillon>