From aew.conserver@planix.com Mon Oct 15 04:57:14 2001 Received: from whome.planix.com (whome.planix.com [204.29.161.33]) by underdog.stansell.org (8.11.6/8.11.6) with ESMTP id f9FBvDO16349 for ; Mon, 15 Oct 2001 04:57:13 -0700 (PDT) Received: from woffi.planix.com([204.29.161.34]) (859 bytes) by whome.planix.com via sendmail with P:esmtp/R:inet_hosts/T:smtp (sender: ) id for ; Mon, 15 Oct 2001 07:57:11 -0400 (EDT) (Smail-3.2.0.112 2001-Feb-5 #2 built 2001-Jun-19) Date: Mon, 15 Oct 2001 07:57:11 -0400 (EDT) From: Andreas Wrede Reply-To: aew.conserver@planix.com To: users@conserver.com Subject: nailing-up serial connections Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: I have a couple of serial consoles connected to a conserver 7.1.1. When one of the attached servers reboots conserver drops the console connection, but it does not re-aquire it until a 'console' issues a ^Eco. Is there any way to nail-up serial console connections so that the connections is not dropped in spite of CD going low? -- - aew From setzer@epoch.placemark.com Mon Oct 15 08:52:45 2001 Received: from epoch.placemark.com (epoch.placemark.com [12.37.36.167]) by underdog.stansell.org (8.11.6/8.11.6) with SMTP id f9FFqjO18158 for ; Mon, 15 Oct 2001 08:52:45 -0700 (PDT) Received: (qmail 351 invoked by uid 1033); 15 Oct 2001 15:52:05 -0000 Date: Mon, 15 Oct 2001 10:52:05 -0500 From: Kelly Setzer To: users@conserver.com Subject: 2 stop bits Message-ID: <20011015105205.D28076@placemark.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: I'm looking at installing the conserver package, but I cannot find any hint that it supports devices that require 2 stop bits. Is there a solution? thanks, Kelly -- "Try not. Do or do not. There is no try." --Yoda From bryan@stansell.org Mon Oct 15 11:43:43 2001 Received: (from bryan@localhost) by underdog.stansell.org (8.11.6/8.11.6) id f9FIhhC19762 for users@conserver.com; Mon, 15 Oct 2001 11:43:43 -0700 (PDT) Date: Mon, 15 Oct 2001 11:43:43 -0700 From: Bryan Stansell To: users@conserver.com Subject: Re: nailing-up serial connections Message-ID: <20011015114343.S5739@underdog.stansell.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from aew.conserver@planix.com on Mon, Oct 15, 2001 at 07:57:11AM -0400 Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: Hi Andreas, Conserver used to re-open downed lines automatically. I recently changed it so that it wouldn't do that. The reason? In most cases it caused conserver to respawn processes and run in a fairly tight loop, causing the load to rise, etc. It all depended on the situation, but it usually wasn't very pretty. I'd like the code to try and reopen the console once and then force it down if it fails, but that'll have to be in the future. Because of the change, you and others are now seeing another problem of treating console lines like modems instead of hardwired terminals (carrier loss drops the line). 7.1.2 (which should be out in a day or two - depends on how fast I package it up) will have the following code change... In conserver/consent.c you'll see a line: termp.c_cflag = CREAD; change that to: termp.c_cflag = CREAD | CLOCAL; I'm looking at changing the non-termios based code as well, but since I don't have a machine to test it, I'm fearful of putting in the code. It would be something like: if (-1 == ioctl(pCE->fdtty, TIOCLBIS, LNOHANG) { Error("ioctl3: %d: %s", pCE->fdtty, strerror(errno)); return -1; } and put somewhere near the TIOCSLTC code in TtyDev(). I'm also not sure how many people out there have non-termios based systems still - it could all be dead code these days and should just be removed. Bryan On Mon, Oct 15, 2001 at 07:57:11AM -0400, Andreas Wrede wrote: > I have a couple of serial consoles connected to a conserver 7.1.1. > When one of the attached servers reboots conserver drops the console > connection, but it does not re-aquire it until a 'console' issues a > ^Eco. Is there any way to nail-up serial console connections so that > the connections is not dropped in spite of CD going low? > > -- > - aew From bryan@stansell.org Mon Oct 15 11:50:21 2001 Received: (from bryan@localhost) by underdog.stansell.org (8.11.6/8.11.6) id f9FIoLX19929 for users@conserver.com; Mon, 15 Oct 2001 11:50:21 -0700 (PDT) Date: Mon, 15 Oct 2001 11:50:21 -0700 From: Bryan Stansell To: users@conserver.com Subject: Re: 2 stop bits Message-ID: <20011015115021.T5739@underdog.stansell.org> References: <20011015105205.D28076@placemark.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20011015105205.D28076@placemark.com>; from setzer@placemark.com on Mon, Oct 15, 2001 at 10:52:05AM -0500 Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: The code doesn't currently support it. It really should. I'll put this on the TODO list and hopefully it will get in the next release (not 7.1.2 since I'm packaging that up, but the one after). If you'd like to hardcode that on for all consoles, you can add the CSTOPB constant to the parity[] definitions in conserver/consent.c. This would change from: {'e', PARENB | CS7, 0} , /* even */ {'m', PARENB | CS7 | PARODD | PAREXT, 0} , /* mark */ {'o', PARENB | CS7 | PARODD, 0} , /* odd */ {'p', CS8, 0} , /* pass 8 bits, no parity */ {'s', PARENB | CS7 | PAREXT, 0} , /* space */ to {'e', PARENB | CS7 | CSTOPB, 0} , /* even */ {'m', PARENB | CS7 | PARODD | PAREXT | CSTOPB, 0} , /* mark */ {'o', PARENB | CS7 | PARODD | CSTOPB, 0} , /* odd */ {'p', CS8 | CSTOPB, 0} , /* pass 8 bits, no parity */ {'s', PARENB | CS7 | PAREXT | CSTOPB, 0} , /* space */ At least, I believe that should work. Let me know if it doesn't. Ideally, you'd be able to flag that in the configuration file on a per-console basis. The syntax for specifying those kinds of options isn't there (it would be nice to be able to specify other things as well), so there's some thought that needs to be put into this. I'll make sure it gets on the TODO list so it's at least tracked. Bryan On Mon, Oct 15, 2001 at 10:52:05AM -0500, Kelly Setzer wrote: > I'm looking at installing the conserver package, but I cannot find any > hint that it supports devices that require 2 stop bits. > > Is there a solution? > > thanks, > Kelly > -- > "Try not. Do or do not. There is no try." --Yoda > > _______________________________________________ > users mailing list > users@conserver.com > https://www.conserver.com/mailman/listinfo/users From bryan@stansell.org Mon Oct 15 23:26:05 2001 Received: (from bryan@localhost) by underdog.stansell.org (8.11.6/8.11.6) id f9G6Q5o09163; Mon, 15 Oct 2001 23:26:05 -0700 (PDT) Date: Mon, 15 Oct 2001 23:26:05 -0700 From: Bryan Stansell To: announce@conserver.com, users@conserver.com Subject: conserver 7.1.2 is available Message-ID: <20011015232605.Y5739@underdog.stansell.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: The big news in this release is the addition of tcp_wrappers support and fixing the line-based timestamp code. There are other good things worth upgrading for as well - here's the complete list ... version 7.1.2 (Oct 15, 2001): - fixed line-based timestamp code - reported by Benn Oshrin - tcp_wrappers support (--with-libwrap) - CLOCAL bit set for local ports - patch by Egan Ford - timestamp added to 'lost carrier' error - suggested by Todd Stansell - Alternate break sequence for Solaris 8 is available as ^ecl2 escape sequence - patch by William Charles The following suggested by Trevor Fiatal - Widened username field of 'console -w' output - Added server hostname to password entry prompt - AC_CHECK_LIB replaced with AC_SEARCH_LIBS in configure.in so that irrelavent (and sometimes incompatible) libraries aren't linked in Bryan Stansell From bryan@stansell.org Tue Oct 16 21:35:48 2001 Received: (from bryan@localhost) by underdog.stansell.org (8.11.6/8.11.6) id f9H4Zmq01710; Tue, 16 Oct 2001 21:35:48 -0700 (PDT) Date: Tue, 16 Oct 2001 21:35:48 -0700 From: Bryan Stansell To: announce@conserver.com, users@conserver.com Subject: conserver 7.1.3 is available Message-ID: <20011016213548.F5739@underdog.stansell.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: Yep folks, I've done it again. I put a really stupid bug into the code and didn't realize it. If you want to be able to send a real break to a terminal server based console, don't use 7.1.2 - get 7.1.3. Sorry about this folks. If got a copy of 7.1.2, just erase it. version 7.1.3 (Oct 16, 2001): - NetBSD 1.5 termios bug/compatibility - patch by Andreas Wrede - Missing quotes broke ^ecl1 code - reported by William Charles - Alternate break code didn't exist for local ports Bryan Stansell From jh@axis.com Thu Oct 25 12:07:52 2001 Received: from krynn.axis.se (krynn.axis.se [193.13.178.10]) by underdog.stansell.org (8.12.1/8.12.1) with ESMTP id f9PJ7nxM019529 for ; Thu, 25 Oct 2001 12:07:51 -0700 (PDT) Received: from milamber.axis.se (jh-vtun-home.axis.se [10.2.8.10]) by krynn.axis.se (8.12.1/8.12.1/Debian -1-woody1) with ESMTP id f9PJ7e7Z027451 for ; Thu, 25 Oct 2001 21:07:41 +0200 Received: from axis.com (localhost [127.0.0.1]) by milamber.axis.se (8.12.0/8.12.0/Debian -2) with ESMTP id f9PJ7U1L014651 for ; Thu, 25 Oct 2001 21:07:30 +0200 To: users@conserver.com From: =?iso-8859-1?Q?J=F6rgen_H=E4gg?= Subject: running conserver as non-root? MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-ID: <14647.1004036847.1@axis.com> Date: Thu, 25 Oct 2001 21:07:29 +0200 Message-ID: <14650.1004036849@axis.com> Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: Would there be any problem running conserver as a normal user instead of root? (Except for using a higher port number than 782 of course.) Something like having a 'conserver'-user belonging to the same group as the serial ports. Any other disadvantages? From bryan@stansell.org Thu Oct 25 12:20:37 2001 Received: from underdog.stansell.org (localhost [127.0.0.1]) by underdog.stansell.org (8.12.1/8.12.1) with ESMTP id f9PJKbxM019812 for ; Thu, 25 Oct 2001 12:20:37 -0700 (PDT) Received: (from bryan@localhost) by underdog.stansell.org (8.12.1/8.12.1/Submit) id f9PJKb6B019811 for users@conserver.com; Thu, 25 Oct 2001 12:20:37 -0700 (PDT) Date: Thu, 25 Oct 2001 12:20:37 -0700 From: Bryan Stansell To: users@conserver.com Subject: Re: running conserver as non-root? Message-ID: <20011025122037.K5739@underdog.stansell.org> References: <14650.1004036849@axis.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.2.5i In-Reply-To: <14650.1004036849@axis.com>; from jorgen.hagg@axis.com on Thu, Oct 25, 2001 at 09:07:29PM +0200 Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: As long as you can open the console ports (be it local serial or TCP connections or whatever) there should be no problem (I do a lot of my testing by using my own account and not root). The only real reason for conserver to run as root is for the low-numbered port. If you've been running conserver as root and change to a non-root account (or generically want to change from one account to the other), you'll need to modify permissions on logfiles and such, but that's about it. Other disadvantages? Just make sure you have a good startup script that folks use religiously so things don't start as root (even for BSD systems). If it creates a new logfile as root and then you re-run as a non-root user you get into a small mess (not major, but annoying). Other than that, I can't think of any. Bryan On Thu, Oct 25, 2001 at 09:07:29PM +0200, Jörgen Hägg wrote: > > Would there be any problem running conserver as a normal user > instead of root? > (Except for using a higher port number than 782 of course.) > > Something like having a 'conserver'-user belonging to the > same group as the serial ports. > > Any other disadvantages? > _______________________________________________ > users mailing list > users@conserver.com > https://www.conserver.com/mailman/listinfo/users From bryan@stansell.org Thu Oct 25 14:55:47 2001 Received: from underdog.stansell.org (localhost [127.0.0.1]) by underdog.stansell.org (8.12.1/8.12.1) with ESMTP id f9PLtlxM021160 for ; Thu, 25 Oct 2001 14:55:47 -0700 (PDT) Received: (from bryan@localhost) by underdog.stansell.org (8.12.1/8.12.1/Submit) id f9PLtlu1021159 for users@conserver.com; Thu, 25 Oct 2001 14:55:47 -0700 (PDT) Date: Thu, 25 Oct 2001 14:55:47 -0700 From: Bryan Stansell To: users@conserver.com Subject: Re: running conserver as non-root? Message-ID: <20011025145547.A21100@underdog.stansell.org> References: <14650.1004036849@axis.com> <20011025122037.K5739@underdog.stansell.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20011025122037.K5739@underdog.stansell.org>; from bryan@conserver.com on Thu, Oct 25, 2001 at 12:20:37PM -0700 Sender: users-admin@conserver.com Errors-To: users-admin@conserver.com X-BeenThere: users@conserver.com X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Post: List-Subscribe: , List-Id: Conserver Users List-Unsubscribe: , List-Archive: Ooops...as a friend just reminded me, the other potential issue is reading shadow password files. If your system uses them and you use '*passwd*' as the password entry in the conserver.passwd file, user authentication will fail (since only root can read the shadow password entries). To get around this, just put the encrypted password in the conserver.passwd file. Some would say doing that decreases your security (effectively no longer using shadow password files), but if you're using the console client on another host, you have bigger security concerns 'cause the passwords are going over the wire in the clear. So, take it all with a grain of salt. Bryan