TACACS (tac-plus) help??
Greetings, Is anyone on the list using the tac-plus package for TACACS+ services? If so, can I ask you to send me a copy of your tacacs.conf file? Please sanitize it before sending to me (passwords, etc.), but I'm looking for guidance on the format of the file. I'm having a hard time figuring out how to do privilege levels and groups at this point. Thanks for your help! pt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
RE: OT Aliasing Multiple Addresses in Mutt
He was referring to MUAs, not MTAs. Sheesh. Pete Templin [EMAIL PROTECTED] -Original Message- From: Walt Mankowski [mailto:[EMAIL PROTECTED]] Sent: Tuesday, January 14, 2003 8:17 PM To: debian-user Subject: Re: OT Aliasing Multiple Addresses in Mutt On Mon, Jan 13, 2003 at 07:09:58PM -0800, Paul Johnson wrote: What were you expecting? The only way you can suppress each recipient from knowing who else is recieving something is using BCC... That's not true. Mailing list managers do it (I don't know the addresses of everyone subscribed to debian-user). You can also supress the names of the recipients by adding an entry to /etc/aliases (or whatever your MTA uses). Walt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
debian 2.1 - 2.2 upgrade woes
I did a live upgrade (production server, company doesn't have funds for a development or transition server) on a box from Debian 2.1 to Debian 2.2, and reminded myself why I don't like to upgrade boxes. I've got most of the stuff fixed (finally), but I get these errors in dselect: When trying to [A]ccess, Connecting to ftp.debian.org... Can't call method quit on an undefined value at /usr/lib/dpkg/methods/ftp/setup line 148, STDIN chunk 10. query/setup script returned error exit status 29. Press enter to continue. When trying to [I]nstall, Couldn't open /var/lib/dpkg/methods/ftp/vars : No such file or directory Try to relaunch the « Access » step in dselect, thanks. installation script returned error exit status 2. Press enter to continue. Any suggestions on how to get my dselect happy again? Please reply directly, as I'm not on debian-user at this time. Thanks, Pete -- Peter J. Templin, Jr., CCNP, CCDP Networking Consultant On-Line Internet Services - URDirect.net A division of Global On-Line Computers 5606 Randolph Blvd [EMAIL PROTECTED] San Antonio, TX 78233 (210)477-7603
Bungled bash
I've got a Debian 1.3 box with a mucked up bash - I get Segmentation fault if I try to run it or any scripts referencing it. Could somebody per chance mail me a working bash, and/or tell me some quick fixes to get it working. I tried reinstalling the .deb, but of course that doesn't work. I tried recompiling from the source, but that doesn't work. TIA, Pete -- Peter J. Templin, Jr. Systems and Networks Administrator JD-WEB Computer Sales and Service 429 Market St. [EMAIL PROTECTED] Lewisburg, PA 17837 (570)523-6800
Re: change the ip address without rebooting
On Sun, 12 Sep 1999, Robert Maynard Rhyu wrote: On Mon, 13 Sep 1999, Shao Zhang wrote: How do I change the ip address of my machine without rebooting? I did an ifconfig eth0 xxx.xxx.xxx.xx netmask 255.255.255.0 broadcast xxx.xxx.xxx.255 But it does not seem to be enough. After the above command, I cannot telnet to anywhere unless I do a reboot... Can someone please help me?? You must take down the interface prior to assigning it a new IP address: Be sure to edit /etc/init.d/network and /etc/hosts to reflect your changes if you intend to keep your new IP address. When did this become a requirement? I dynamically changed the address of several boxes live. I just had ifconfig eth0 old address parameters ifconfig eth0:1 new address parameters route blah route blah:1 in operation and ifconfig eth0 new address parameters ifconfig eth0:1 old address parameters route blah route blah:1 in my /etc/init.d/network. I just ran /etc/init.d/network and it changed fine. Even wilder was the time I was coming in from a telnet session and ran /etc/init.d/network from a remote computer. Not only did the address swap successfully, I didn't even lose my telnet session! Pete -- Peter J. Templin, Jr., CCNA Systems and Networks Administrator On-Line Internet Services - URDirect.net A division of Global On-Line Computers 2414 Babcock Rd. Suite 106 [EMAIL PROTECTED] San Antonio, TX 78229 (210)692-9911
Software to provide free email service, anybody?
Anyone here know of any software packages which simplify the process of offering a free email service like Hotmail? It doesn't have to be _free_, but affordable is important. Operability on Debian Linux is essential. Thanks, Pete -- Peter J. Templin, Jr. Systems and Networks Administrator On-Line Internet Services - URDirect.net A division of Global On-Line Computers 2414 Babcock Rd. Suite 106 [EMAIL PROTECTED] San Antonio, TX 78229 (210)692-9911
Re: Amanda questions
On Wed, 23 Jun 1999, Tommi Virtanen wrote: On Tue, Jun 22, 1999 at 03:28:03PM -0400, Pete Templin wrote: A few questions about amanda, for anyone who might be familiar with it: ERROR: charlie: [access as backup not allowed from [EMAIL PROTECTED] ERROR: alpha: [access as backup not allowed from [EMAIL PROTECTED] ERROR: bravo: [access as backup not allowed from [EMAIL PROTECTED] echo charlie.3threes.net backup ~backup/.amandahosts Next idea? That's already there. Pete -- Peter J. Templin, Jr. Systems and Networks Administrator Jlink Internet Services 1000 S. Market St. [EMAIL PROTECTED] Bloomsburg, PA 17815(717)389-6400
Re: samba/network neighborhood question
On Thu, 22 Apr 1999, Ben Frame wrote: I just got Samba installed and it seems to be working fine. But it doesn't always show up in my Network Neighborhood under Win95. Both the Debian machine and Win95 machine are on the same subnet and both are in a workgroup called linux. I've made several changes to my smb.conf file, and consequently restarted the debian machine a few times. Sometimes it shows up in Network Neighborhood, and sometimes it doesn't. Sometimes it will show up later, but not immediately. So my question is, what makes it show up (or not) in Network Neighborhood? I can see the Debian machine if I do a search for it with Win95 (start/find/computer/name). And for now I've just placed a shortcut to it on my desktop to keep from having to search every time. In a mixed NT/Linux environment, I found that using a WINS server (and configuring it into smb.conf) makes my Linux servers show up immediately for all TCP/IP-ready Windoze boxen. Of course, if (when - it was NT) the WINS box crashes and is rebooted, the Linux servers aren't visible until samba is restarted. I'd bet in a non-NT environment, merely enabling wins support in smb.conf and configuring said WINS server into Winders should at least help the visibility of the server. YMMV. Pete -- Peter J. Templin, Jr. Systems and Networks Administrator Jlink Internet Services 1000 S. Market St. [EMAIL PROTECTED] Bloomsburg, PA 17815(717)389-6400
Re: Resizing ext2 filesystems
On Thu, 15 Apr 1999, Max wrote: I have a problem in that I'm quickly running out of space in /usr but I have tons of space left in /home. Here's what df shows: Filesystem 1k-blocks Used Available Use% Mounted on /dev/sda1 497667 35357436608 7% / /dev/sda5 497667385312 86653 82% /var /dev/sda6 2478138 2271514 78508 97% /usr /dev/sda7 4616953643170 3734818 15% /home My suggestion? Too late to do it now, but DON'T PARTITION so much. Make three partitions (IMHO primary partitions for all three) as follows: /dev/sda1 500MB / /dev/sda2 120MB swap /dev/sda3 rest /usr Create /usr/home and symlink /home to /usr/home. Create /usr/var and symlink /var there. Now, your poor overworked disk head will have less partition-wide hunting to do to go from ~/yourfiles to /var/log/yourlogs, and the only space crisis you'll run into is filling the whole disk. If you can do that, than, well, you've got yourself a legitimate problem. From O'Reilly's System Perfomance Tuning (http://www.ora.com) by Mike Loukides (I think - please don't shoot me if I'm wrong), if at all possible, don't use more than one partition per disk. On a single disk system, I prefer to bend the rules just a little bit: separating the root filesystem makes sense to me, and a swap partition is important. Pete -- Peter J. Templin, Jr. Systems and Networks Administrator Jlink Internet Services 1000 S. Market St. [EMAIL PROTECTED] Bloomsburg, PA 17815(717)389-6400
Hashed mailspool on procmail and qpopper
Does anybody have good patches for procmail and qpopper (from the hamm vintage) that implements a hashed mail spool (aka /var/spool/mail/u/s/user), preferably such that they use the hashed version if it exists, and the non-hashed if it doesn't? I'm not a good enough C programmer to even attempt this on my own. FWIW, I have managed to get procmail to use _only_ the hashed mailspool, but haven't been able to get qpopper to compile (missing a mailock.h file as referenced in pop_dropcopy.c). Any help is appreciated, Pete -- Peter J. Templin, Jr. Systems and Networks Administrator Jlink Internet Services 1000 S. Market St. [EMAIL PROTECTED] Bloomsburg, PA 17815(717)389-6400
Re: Dell Poweredge 1300 can't boot kernel 2.0.36
On Thu, 15 Apr 1999, Ossama Othman wrote: On 15 Apr, Pete Templin wrote: I'm trying to set up a Poweredge 1300 (with integrated Adaptec 7890 Ultra2 LVD SCSI controller) on a custom kernel v2.0.36 (added support for the AIC-7xxx, of course), and I get the following error messages: I've documented how I installed Debian on my Dell PowerEdge Server 6300. The instructions are at: http://www.debian.org/~ossama/DELL6300 As I expected the moment I read your email, it worked like a champ. For some reason, dinstall dumped me back to the beginning of dinstall every time I tried to load the OS Kernel and Modules from CD (/dev/hdc, not even a SCSI CD-ROM), but I'm quite fluent with the seven floppy method. Sad thing is Dell called me back two hours after I received your email to indicate that they do not support Linux. Funny how they're selling machines with Linux (Red Hat, of course) and they claim to support the Linux concept, yet they can't even troubleshoot a kernel problem (which theoretically should be distribution-independent). Although I can't stand the filesystem disorganization of RedHat, I almost want to get my hands on a RedHat CD and try it on my server. I really hope RedHat _can't_ boot it without a customized boot disk or something, just to prove that they have a problem to solve. Anyway, thanks for the help. Still a 100% satisfied Debian customer... Pete -- Peter J. Templin, Jr. Systems and Networks Administrator Jlink Internet Services 1000 S. Market St. [EMAIL PROTECTED] Bloomsburg, PA 17815(717)389-6400
Dell Poweredge 1300 can't boot kernel 2.0.36
I'm trying to set up a Poweredge 1300 (with integrated Adaptec 7890 Ultra2 LVD SCSI controller) on a custom kernel v2.0.36 (added support for the AIC-7xxx, of course), and I get the following error messages: (scsi0) Adaptec AIC-7890/1 Ultra2 SCSI host adapter found at PCI 11/0 (scsi0) Wide Channel, SCSI ID=7, 32/255 SCBs (scsi0) Downloading sequencer code... 407 instructions downloaded scsi1: SCSI bus busy, waiting up to five seconds scsi1: bus busy, attempting abort If I disable the integrated SCSI controller in the bios, I get only lines 4 5, with s/scsi1/scsi0/ as seen above. Any hints on getting it up and running? I really don't want to have to put RedHat (Dell's buddy in the Linux game) on this box... Pete -- Peter J. Templin, Jr. Systems and Networks Administrator Jlink Internet Services 1000 S. Market St. [EMAIL PROTECTED] Bloomsburg, PA 17815(717)389-6400
Static routing through dynamic interface
Good evening, I'm having some stressful difficulties with some routing issues. I've got a modem server with 16 modems on a Cyclades serial port box. I'm trying to statically route six Ip addresses (204.186.230.8/255.255.255.248) down one of those modems. I can do so, but anytime the modem hangs up, the associated ppp# interface goes down, and the static route goes away. I need to glue this route in place, even if it ends up forcing me to discard packets in the modem server. I've been unsuccessful running RIP between the client and the server, even with statically configured hosts in /etc/gateways. I seem to be semi-successful using RIP on my Cisco to redistribute static routes, but that's definitely asking for trouble, and a pain to manage (in a way). Any help or guidance will be greatly appreciated. The gateway of choice is 204.186.230.177 for any who are curious. Pete -- Peter J. Templin, Jr. Systems and Networks Administrator JD-WEB Computer Sales and Service 429 Market St. [EMAIL PROTECTED] Lewisburg, PA 17837 (717)523-6800 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
samba error: device does not exist on the network
Good morning, I'm experiencing some difficulties with a samba server I've got, and frankly I'm perplexed. I'm trying to use user security on a server that was providing server security, and I'm getting the above message (The specified device does not exist on the network). It worked fine as a server-security samba server, and I believe it worked fine as a user-security samba server, but doesn't any more. The only thing which I can identify as having changed recently is additional virtual interfaces on the primary ethernet card. Any suggestions on where to look? I've attached my config file for reference Pete -- Peter J. Templin, Jr. Systems and Networks Administrator JD-WEB Computer Sales and Service 429 Market St. [EMAIL PROTECTED] Lewisburg, PA 17837 (717)523-6800 [global] printing = bsd printcap name = /etc/printcap load printers = no guest account = nobody workgroup = jdweb netbios aliases = members socket options = TCP_NODELAY IPTOS_LOWDELAY interfaces = 204.186.230.249/255.255.255.128 domain master = no local master = no preferred master = no os level = 2 max log size = 1024 security = user wins server = wins.jdweb.com username map = /etc/smb.users [homes] comment = Home Directories browseable = no read only = no create mode = 0755 directory mask = 0775 [printers] comment = All Printers browseable = no path = /tmp printable = yes public = no writable = no create mode = 0700
linux diald - linux mgetty
Folks, I'm trying to configure a dial-on-demand Linux box to dial into a Linux dialin server with a Cyclades multiport card and the standard mgetty AutoPPP stuff. Since it's AutoPPP and hence PAP authentication, I need to disable the script. I've done so, but I don't seem to be getting authenticated on the remote site. Can anyone shed some light on how to configure the dialup client appropriately? TIA, Pete -- Peter J. Templin, Jr. Systems and Networks Administrator JD-WEB Computer Sales and Service 429 Market St. [EMAIL PROTECTED] Lewisburg, PA 17837 (717)523-6800 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Poor interaction between chown and quota
Hi: In the process of debugging a variety of quota problems visible in procmail-based mail delivery and qpopper-based mail pickup, I believe I've identified an interaction problem between chown and quota. Essentially, if a program running as superuser creates/appends/edits a file which would put the user over soft quota, and then chown's it to the appropriate user, the user is considered over soft quota with NO grace period. I've already found this to be problematic in the following situations: 1) I use sendmail with procmail as the local delivery agent. If an incoming message would put the user over soft quota, the message is returned undeliverable with the message overquota. IMHO, this shouldn't happen unless the user would exceed hard quota or has expired his/her grace period. The tricky part is I use a perl script run out of cron to notify people if they're over quota...unfortunately I notify via email. 2) I use qpopper to offer email to our customers. Quotas are enforced (10M soft, 21M hard) on the /var partition, which therefore limits both the user's mailbox and the corresponding copy which qpopper creates beneath /var/spool/pop . Since the user's mailbox can't exceed 10M (see problem 1, above), the user can't (easily) exceed 20M of disk usage while qpopper has the /var/spool/pop/username.pop file in use, so the user can't exceed their hard quota. Some of our users are unable to POP their mail because of the following error: Apr 1 13:15:59 webhost in.qpopper[8981]: [EMAIL PROTECTED]: -ERR Unable to copy mail spool file, quota exceeded (122) Obviously, something isn't right. Are there any patches/fixes that I can apply? Should I file a bug, and if so, against which package? If anyone wants further information, please email. TIA! tcsh# dpkg -s qpopper Package: qpopper Status: install ok installed Priority: extra Section: mail Installed-Size: 110 Maintainer: Miquel van Smoorenburg [EMAIL PROTECTED] Version: 2.2-4 tcsh# dpkg -s procmail Package: procmail Status: install ok installed Priority: standard Section: mail Installed-Size: 209 Maintainer: Santiago Vila [EMAIL PROTECTED] Version: 3.10.4-2 tcsh# dpkg -s quota Package: quota Status: install ok installed Priority: optional Section: admin Installed-Size: 118 Maintainer: Heiko Schlittermann [EMAIL PROTECTED] Version: 1.55-8 tcsh# dpkg -s fileutils Package: fileutils Essential: yes Status: install ok installed Priority: required Section: base Installed-Size: 1011 Maintainer: Galen Hazelwood [EMAIL PROTECTED] Version: 3.16-2 Pete -- Peter J. Templin, Jr. Systems and Networks Administrator JD-WEB Computer Sales and Service 429 Market St. [EMAIL PROTECTED] Lewisburg, PA 17837 (717)523-6800 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Poor interaction between chown and quota
On Thu, 2 Apr 1998, Nathan E Norman wrote: On Thu, 2 Apr 1998, Pete Templin wrote: : : Hi: : : In the process of debugging a variety of quota problems visible in : procmail-based mail delivery and qpopper-based mail pickup, I believe I've : identified an interaction problem between chown and quota. Essentially, : if a program running as superuser creates/appends/edits a file which would : put the user over soft quota, and then chown's it to the appropriate user, : the user is considered over soft quota with NO grace period. I've already : found this to be problematic in the following situations: : : 1) I use sendmail with procmail as the local delivery agent. If an : incoming message would put the user over soft quota, the message is : returned undeliverable with the message overquota. IMHO, this shouldn't : happen unless the user would exceed hard quota or has expired his/her : grace period. The tricky part is I use a perl script run out of cron to : notify people if they're over quota...unfortunately I notify via email. Hmm, I hadn't noticed this (yet). I will test and try to confirm. : 2) I use qpopper to offer email to our customers. Quotas are enforced : (10M soft, 21M hard) on the /var partition, which therefore limits both : the user's mailbox and the corresponding copy which qpopper creates : beneath /var/spool/pop . Since the user's mailbox can't exceed 10M (see : problem 1, above), the user can't (easily) exceed 20M of disk usage while : qpopper has the /var/spool/pop/username.pop file in use, so the user can't : exceed their hard quota. Some of our users are unable to POP their mail : because of the following error: : : Apr 1 13:15:59 webhost in.qpopper[8981]: [EMAIL PROTECTED]: -ERR Unable : to copy mail spool file, quota exceeded (122) Since quotas are filesystem based, you must plan your filesystem layout accordingly. The only easy way around this is to either insure that /var/spool/mail and /var/spool/pop are seperate filesystems, or grab the qpopper sources and make popper put its tempfiles somewhere other than /var/spool/pop ... preferably a seperate filesystem. I don't think you can blame the second problem on qpopper. This would solve problem 2, but not problem 1. I have a wonderful workaround for problem 2: NFS mount /var/spool/mail on another server, install qpopper, and change the IP address of pop.jdweb.com. However, this only works if management store staff (which are the same two people) used the correct hostnames in everyone's setup. :) I figure the problem is somewhere between the chown system call and quota, but I'm not sure. Looking forward to your ideas and suggestions, or directions towards what to file a bugreport on... Pete -- Peter J. Templin, Jr. Systems and Networks Administrator JD-WEB Computer Sales and Service 429 Market St. [EMAIL PROTECTED] Lewisburg, PA 17837 (717)523-6800 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Procmail as delivery agent + quotas
Hello there! Here at JD-WEB I'm using procmail as the local delivery agent within sendmail 8.8.5 on Debian 1.3.x with quotas enabled and in effect. I've found that procmail won't deliver (a.k.a. will bounce) a mail if it would put the user over soft quota. Is it me, or is this an odd behavior? It seems to me that it should choke and bounce if the email would send the user over hard quota or if the user is out of grace period. Unfortunately, I'm running daily overquota messages via email, which stand to fail if the user's inbox is above or close to soft quota. Any hints, tips, tricks, Fine manuals, or other forms of advice are all openly welcomed. TIA, Pete -- Peter J. Templin, Jr. Systems and Networks Administrator JD-WEB Computer Sales and Service 429 Market St. [EMAIL PROTECTED] Lewisburg, PA 17837 (717)523-6800 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
questions about pppd (fwd)
(A forwarded message, due to mailer problems...) -- please forward this to the mailing list for me, thanks Fuz we have gotten the pppd daemon to start correctly using pon and poff scripts but we note that now that we ppp support compiled in verses in modules the ppp.o module always tries to load itself and fail. I sent mail from my slackware box to the debian box, (they are both connected to different ISPs and event tho owl.asarian.org is in my domain, its got ISP2's IP address. owl does have a static connection and is correctly defined in my name server and my ISP2's in-addr.arpa domain for IP address. so there is something to transfer. this shows that CSLIP is being called even tho I intentionally left it and SLIP out of the kernel gen. unless the make config that make-dpkg does changes the options from what I selected in make config myself. BTW is there a way to get makedpkg NOT to do a second make config after I've done one with the option I want? it then attempts load ppp.o module which should not match the new gen. then it figures out the ppp code is compiled in and uses it ;)? I'd prefer it not bother loading an out of date set of modules, then figuring out the PPP code was already compiled in and using it, I think it should look for the code in the kernel 1st. the connection goes normally, email sent from onbe ISP to the other arrives as expected. when I went to disconnect it reported Jan 8 16:34:51 owl pppd[209]: disconnect script failed obviously it successfully disconnected and the modems light went out. what did I do wrong?? console excerpt follows... thanks in advance for any thoughts. oh BTW I did commented out ALL the modules in /etc/modules as all the things I need are compiled into the kernel except for sound support and thats handled by OSS's soundon command. we can telnet into the box when its connected. we always get this line too... Jan 8 16:33:06 owl pppd[209]: Cannot determine ethernet address for proxy ARP we would like to be able to setup the 10base2 connection between the boxes so we can have the debian box do the serving for the slackware box, until we can get it upgraded. since we are unsure how long it will be before we actually get the promised /29 subnet, we figured we could do IP-MASQ on the slackware box so clients using our servers would 'think' the servers were on the slackware box and it IP-forward to the debian box. both systems (slackware 3.0/kernel 1.2.13(with MASQ patch) and debian 1.3.1/kernel 2.0.29) recognise the NE2k clones during bootup properly. Fuzzy Sys Admin, ASARian.org Jan 8 15:34:40 owl login[185]: ROOT LOGIN on `tty3' Jan 8 16:32:42 owl kernel: CSLIP: code copyright 1989 Regents of the University of California Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol kill_fasync_Rf99c4230 Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol register_netdev_R24876de2 Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol netif_rx_R578c179c Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol dev_kfree_skb_R44b132be Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol tty_register_ldisc_Rbc2419d2 Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol n_tty_ioctl_Ra07e8b74 Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol dev_alloc_skb_R8587f622 Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol unregister_netdev_Rc260075b Jan 8 16:32:42 owl insmod: /lib/modules/2.0.29/net/ppp.o: unresolved symbol dev_close_R391db8f6 Jan 8 16:32:42 owl kernel: registered device ppp0 Jan 8 16:32:42 owl pppd[209]: pppd 2.2.0 started by root, uid 0 Jan 8 16:32:43 owl chat[210]: abort on (NO CARRIER) Jan 8 16:32:43 owl chat[210]: abort on (BUSY) Jan 8 16:32:43 owl chat[210]: abort on (VOICE) Jan 8 16:32:43 owl chat[210]: abort on (ERROR) Jan 8 16:32:43 owl chat[210]: abort on (NO DIALTONE) Jan 8 16:32:43 owl chat[210]: abort on (NO DIAL TONE) Jan 8 16:32:43 owl chat[210]: abort on (NO ANSWER) Jan 8 16:32:43 owl chat[210]: send (ATZ^M) Jan 8 16:32:43 owl chat[210]: expect (OK) Jan 8 16:32:43 owl chat[210]: ATZ^M^M Jan 8 16:32:43 owl chat[210]: OK -- got it Jan 8 16:32:43 owl chat[210]: send (ATE1M1V1X4\A3L3S0=0D2C1Q0^M) Jan 8 16:32:44 owl chat[210]: expect (OK) Jan 8 16:32:44 owl chat[210]: ^M Jan 8 16:32:44 owl chat[210]: ATE1M1V1X4A3L3S0=0D2C1Q0^M^M Jan 8 16:32:44 owl chat[210]: OK -- got it Jan 8 16:32:44 owl chat[210]: send (ATM1DT4061544^M) Jan 8 16:32:44 owl chat[210]: expect (CONNECT) Jan 8 16:32:44 owl chat[210]: ^M Jan 8 16:33:02 owl chat[210]: ATM1DT4061544^M^M Jan 8 16:33:02 owl chat[210]: CONNECT -- got it Jan 8 16:33:02 owl chat[210]:
Re: bounced mails
On Tue, 11 Nov 1997, Shaleh wrote: I get e-mail from various people's server about mail bouncing. This appears to be a problem with the debian list-serv. FYI: The bounceograms are, in fact, from various peoples' servers, and in particular, ones that don't follow standardized email RFCs. As Debian listmaster, I receive an average of 500 bounceograms per day (aren't filters great?). Since those bounceograms keeping arriving and nothing has changed with respect to the Debian lists, I'm well convinced that the Debian mailing lists are working properly. Now, a request for the masses: In the future, please READ the automatic footer on the bottom of the various Debian mailing lists. If there is trouble with the mailing lists, email me at my day job. Since it is a day job, I shouldn't rightfully be reading debian-user all day long, and directing the message to me will help speed results (I read through debian-user on my own time, which is sporadic, so I'm not as likely to notice a message in the heaps of new mail there). Please DO NOT send messages like this through the list(s)...chances are, it won't do any good for 99% of the folks who get the message (or at least 99% can't fix it). Thanks, Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: 06032921167-0001@T-Online.De
On Tue, 16 Sep 1997, Bruce Perens wrote: Me too. Write to [EMAIL PROTECTED] or [EMAIL PROTECTED] (Pete Templin). By the way, only a broken mail system would bounce the message to you. There's stuff in the RFCs about using the envelope from for errors that this mail delivery agent must be ignoring. This is one of the victims of malicious subscriptions to our lists, who happens to have seven or eight email addresses, all of which forwarding to a single account. To our continued benefit, the final recipient email account typically ends up overquota, and the destination mail system just happens to be non-RFC-standard. Yay. I've noticed a distinct pattern to the malicious subscriptions - 27 of Debian's mailing lists are always involved (the same 27). I've invoked some minor, transparent changes to one of the lists in an attempt to track down problems. Some evidence has been collected already, but not enough to point any fingers. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: adding users after installing NIS
On Tue, 28 Oct 1997, R Chris Ross wrote:
I have just set up NIS on several of my machines and I am having a
little trouble adding a user. If I use adduser the group, shadow and
passwd files in /etc are all updated properly but adduser doesn't
complete properly properly. It terminates before asking for any
information on the user with a message that says that the user
doesn't exist and NIS knows nothing about the user. The user is
being added to the machine that is the master server.
Is there a different way to add users after NIS is installed?
I recently encountered similar difficulties. I understand that it's a bug
in either adduser or useradd.
The fix which was proposed to me and seems to be working well (although
expensive) is this:
tcsh diff /usr/local/sbin/adduser /usr/sbin/adduser
618d617
system(cd /var/yp; make);
Here's the context in my fixed adduser:
sub systemcall {
my $c = join(' ', @_);
print $c\n if $debugging;
system(@_) cleanup($0: `$c' returned error code $?. Aborting.\n);
system(cd /var/yp; make);
}
It's updating the maps each step of the way. Apparently the errant
program is adding the new user beneath the plus entry in the password
file, which is then not read by the NIS-modified libraries or whatever.
This hack causes the NIS tools to see the new entry properly.
Pete
--
Peter J. Templin, Jr. Client Services Analyst
Computer Communication Services tel: (717) 524-1590
Bucknell University [EMAIL PROTECTED]
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble?
e-mail to [EMAIL PROTECTED] .
malicious people subscribing others
Hello, It has come to my attention that individuals are using the open architecture of the debian mailing lists to subscribe other individuals. The most recent incident which has come to my attention involves a victimized recipient who has several email addresses, some of which forward to other mailboxes. This person's name appears on numerous (about 27) of the debian mailing lists; this list of lists is identical to the last time the person was maliciously added to the lists. I ask that anyone involved in this activity cease immediately. Anyone who has any information about said actions is asked to contact me by any means desired. I don't think the Debian project wants to increase its membership (nor its hate list) by actively subscribing unwishing people to its mailing lists, nor do I think that the Debian project wants extra load to be placed on its donated resources. I thank you for your cooperation with this matter. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
email to pager software???
Hello, I'm working on a project to handle email-to-pager service. I found a software tool to do so, but the C program segfaults. Can anyone suggest a software package, and/or volunteer to help me debug the problem? Thanks, Pete -- Peter J. Templin, Jr. Network Administrator JD-WEB Computer Sales and Service 429 Market St. (717) 523-6800 Lewisburg, PA 17837 [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: DHCP Server
On Thu, 11 Sep 1997, Jim Pick wrote: There might be a way to configure Win95 (via the registry) to require a valid password when logging in at the console. But I'm not sure how. I would never use a Windows 95 machine as a public access terminal. I've seen somebody build a super-secure public access Win 3.1 machine however (involving lots of bizarre contortions). This is possible with a Windows 95 Policy, but the only good way to enforce it is to use an NT server domain and place the newly created policy file config.pol in the NETLOGON directory. You'll need to use user-level security on the W95 machines, and you'll need to require (in the policy) a valid user logon (and you can dictate which NT domain shall be used, too). It's quite a neat feature, although it too can be slightly subverted. I used this in a computer store earlier this year. We set up a policy which took away a lot of functionality for the demo logon, gave the demo logon a blank password, and required a validated logon to use the computer. This restricted what demo users could do, especially browse our network and play/add DOS games. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Sendmail and domain name
On Mon, 8 Sep 1997, Nathan E Norman wrote: On Mon, 8 Sep 1997, David Morris wrote: : I would take a look at the follwing lines in sendmail.cf: : : # my official domain name : # ... define this only if sendmail cannot automatically determine your : domain : #Dj$w.Foo.COM : : Ah, so that is the $j variable it told me about. No wonder a 'grep $j *' : showed nothing. OK, so I tried to change it (and reboot) but when I : changed it to: : : Dj$gently : : or to : : Dj$gently.myispdomain.net : : Neither one solved the problem. According to the (big) sendmail book from O'Reilly, you would change this to Dj$w.myispdomain.net It would seem that you could also put Djgently.myispdomain.net Haven't tried these myself, just read it out of the book, so can't guarantee it :/ However, from your description, 'hostname' returns gently, and that confuses sendmail. If this is the case, use the first example I gave. Another option, depending on what IP address(es) you're using, is to set up a DNS server (bind) on your host (or somewhere on your network, if you're so lucky) which can provide the FQDN of your host. I did that with a diald'ed box (it was already supposed to be a primary nameserver for the domain, but I made it a secondary for the reverse domain of both the network card and the modem's IP addresses) and solved the boottime delay and error messages. Holler if you'd like help with this fix. It would probably make other things faster/easier for you, also. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: WantWEB/Linux/IP Masquerading
On Tue, 9 Sep 1997, Mike wrote: Nils Rennebarth wrote: No, it's just vice versa. Upload speed with 56k Modem (from you to your provider) is 56k. From your provider to you it's still 33.6 maximum, Bullsh*t. Go read something on the subject. Upload is 33, download is 56. Otherwise it would be impossible to sell 56k modems, because no one does more uploading than downloading. Easy now. Not everyone can be as informed as you. I work with a local computer store, which currently has its Internet presence through a dedicated modem. Your download is their upload, so there's a concrete example of more upload than download. On a different angle, what about the FCC rules which essentially limit 56k modems to 51 or 53k? I know practically nothing about it, except that I read it somewhere in Network World, I believe. Back when I was a student and took a class on Communication and Information Systems, we derived that the theoretical maximum bandwidth of a phone line was really about 34kbps, so it doesn't surprise me that there are all of these restrictions on 56k modems. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Who knows stuff about WAN cards?
Hi there!! I'm doing some network consulting for a computer store, and we'd like to connect the store LAN to the Internet through a 56k or T1 connection. If possible, we'd like to save the expense of the router and perhaps the CSU/DSU if possible. I think I've seen some adds in network magazines for WAN cards. Are any of you using them? If so, can you tell me about driver compatibility? Does it replace the CSU/DSU, or did you still have to rent/buy one? What sort of interface options did you have to select from (I've heard that different telco boxes have different interface types)? Thanks for your help, Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: heard all the who-haha?
On Wed, 13 Aug 1997, Donovan Baarda wrote: I'd be interested in knowing if the ping takes a long time or fails. I do think that the DNS would be cached from the ping speeding up DNS for the telnet, but the ping should then experience all the problems. If not, why is resolver responce times different for different applications? Is it possible that reverse lookups are slow, forward lookups fast, and a cached forward lookup entry can be used for a reverse lookup? Is the ping somehow causing a forward lookup first, but the telnet causing a reverse lookup first? I was experiencing slow boot times on a ppp-connected machine (sendmail would pause, waiting for my *misconfigured* diald to *not* dial in). A fix which works well for me was to reconfigure that machine's named to be a secondary server for each of the reverse domains it lived on (you need to specify a backup file location for each domain, which should be standard practice for most folks anyway, for this to work on a dialup). By doing so, most relatively local reverse lookups are incredibly quick. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Passwd: encrypted pw entry
On Wed, 20 Aug 1997, Matthew Tebbens wrote: How can I disable an /etc/passwd entry. Isn't there something I can place in the encrypted-pw section of /etc/passwd to disable the account ? Inserting an asterisk in front of a user's encoded password will disable the account without destroying the ability to restore that account later (by removing the password). I believe an asterisk is not a valid character in the character set used by the password encrypting algorhythm, and it will also shift all of the characters into a different position. Of the thirteen characters, the first two are the salt used to make things random, and the other eleven store the encoded password. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Sendmail broken (Debian 1.3)
On Sun, 27 Jul 1997, D. W. Wieboldt wrote: Good suggestion. I think it is indeed timing out for want of a good lookup. Have host name in /etc/hosts but that doesn't help. Now howto hack sendmail into submission! Does anybody know the simple fix to the .cf file to make it run? Thanks all! One beneficial fix is to install the bind package (or find a convenient, well-connected-to-your-LAN-if-you have-one name server) and set it up to be primary or secondary for the reverse mapping of your IP address(es). I set up a dial-on-demand router for a computer store, and sendmail startup took forever until I configured named to be a secondary for the ISP's modems' subnet x.x.x.in-addr.arpa and the store subnet x.x.x.in-addr.arpa . Bingo! No more delays (at least for that reason...now the delay is the squid boot sequence). Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: bind - no binfmt-0 / too many open files.
On Tue, 29 Jul 1997, ninjaz wrote: Jul 29 01:52:15 www modprobe: can't locate module binfmt-0 Jul 29 01:52:15 www modprobe: can't locate module binfmt-0 Jul 29 01:52:28 www named[20732]: starting. named 4.9.5-REL Mon Apr 28 20:39:58 MET DST 1997 [EMAIL PROTECTED]:/debian/bind/bind-4.9.5/named Jul 29 01:52:28 www named[20732]: fcntl(dfd, F_DUPFD, 20): Too many open files Jul 29 01:52:28 www last message repeated 14 times Also: # lsof | wc -l 852 I was suffering from the file limit when too many sendmails were delivering queued mail. I just accidentally deleted kernel-source-2.0.27, so I can't easily find the hack, but I think I found a similar place to hack it in 2.0.30: In /usr/src/kernel-source-2.0.30/include/linux/ , here's my changed fs.h (excerpted): /* And dynamically-tunable limits and defaults: */ extern int max_inodes, nr_inodes; extern int max_files, nr_files; #define NR_INODE 98304 /* this should be bigger than NR_FILE */ #define NR_FILE 32768 /* this can well be larger on a larger system */ This was 3072 and 1024, respectively. Kernel memory usage hasn't changed anything drastically, and I've certainly done away with those pesky errors. This may or may not have anything to do with what you're experiencing, though. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Unsubscribing
On Sun, 3 Aug 1997, Marcus Brinkmann wrote: You are right, this is really high traffic here! Please read the last three lines of your mail, or the last three lines of my mail :-) They say, that you can get off the list via [EMAIL PROTECTED] ^^^ PS: This is true for every debian-* list. Just send a mail to [EMAIL PROTECTED] to get removed from the list. One IMPORTANT note for readers of the DIGEST form of the lists: Contrary to the information at the bottom of the individual messages, unsubscription from the digest form of the lists is at [EMAIL PROTECTED] . Many, many people seem to get caught by this loophole, and some can get quite loud with their frustrations. I can't seem to come up with an easy fix for this situation, unfortunately. PPS: If you have further trouble, you can mail to a real human being, Pete Templin. He will help you if he can. His email address is [EMAIL PROTECTED] True, true. That's me. If you do have questions or requests, it's often helpful to tell me the following information: 1) Your email address. If you have multiple addresses, this can help me tremendously, particularly if other addresses forward to a central address. 2) What list(s) you're on, or more specifically which list(s) you would like to be removed from. On a side note, some people get the Debian mailing lists by way of intermediate lists. If you are one of those people, I will have little ability to help you. It's often best to save the auto-subscription messages from any mailing list, especially in the above situation. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: speed of X
On Mon, 4 Aug 1997, Paul Wade wrote: Linux will use a swap partition of up to 128 meg. You can add swap files if you need more. I haven't heard anything about slowdowns. Maybe you're thinking about windows swap usage and performance? Somebody correct me if I'm wrong. I know that Linux (or at least Debian, but this sure seems like a Linux issue) can use multiple swap partitions (I think up to 8, perhaps even 16, being up to almost 128MB each). I had two 120M swap partitions at one time, but removed the second due to IDE performance problems (I had a cron-scheduled process that would heavily access the slave drive while swapping to the master drive, and due to my configuration, this was happening on BOTH IDE controllers at the same time. Yuk!!!). I think you can have up to 16 swap files, and I think swap files can be up to 16MB each, but I'm not sure. I was unable to create a swap file in a IDE-based Multiple Drives (md) RAID-0 array, but YMMV. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Swap Space
On Thu, 24 Jul 1997, Shaleh wrote: I agree that the 3x swap rule makes little sense on high end systems. A guru friend of mine explained it as if the kernel needs to swap out your memory it might have to swap the whole thing so you should plan for that. He believes you should use 1.5x your RAM on high end sytems ( = 32 MB). I think I will simply go with a 65MB partition to keep it even. I had posted about the Enlightemment WM -- is no one using it? I have seen very few responses. Any other nice WM's out there? My only (minimal) experience is with fvwm a year and a half ago on Slackware 2.0. I just use the limit of 128MB per swap partition as a guiding rule. Since it's actually slightly less than 128MB, I use 120MB, and use more swap partitions if I actually need it (not that I have, but I did have a second area on templinux just in case). Back when templinux was only 32MB of RAM, I had experienced 57MB of swap usage. Normal operations put me quite close to the 32MB barrier, and simultaneous action on my backups of the debian mailing lists and two separate instances of mirror running (I was migrating to a new disk drive at the time) put me deep into swap. The machine actually ran well until it had to swap, but recovered quickly once the big tasks settled down. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
routing question
Hi there, I'm in the process of setting up a 486 sx/25 as a dialup router (with one modem and one network card). Unfortunately, the ISP can't seem to get the external routing right yet, so my testing is being held up. The question is this: I've compiled a lean, mean kernel with the appropriate IP forwarding enabled (no firewalling or masquerading is being used). Will it route by default, or do I need to add a specific package or other external software? Here's the output of route: Kernel IP routing table Destination Gateway Genmask Flags Metric RefUse Iface cs10.mil.ptd.ne *255.255.255.255 UH0 02 ppp0 cs10.mil.ptd.ne *255.255.255.255 UH1 00 sl0 204.186.230.0 *255.255.255.0 U 0 0 35 eth0 127.0.0.0 *255.0.0.0 U 0 0 18 lo default *0.0.0.0 U 0 0 87 ppp0 default *0.0.0.0 U 1 0 15 sl0 The modem is 204.186.27.145 (cs10-01.mil.ptd.net). Our IP addresses (not yet completely routed, but will be routed through the modem) are: 204.186.230.1, 204.186.230.2, 204.186.230.3 . The first address is given to the network card in the Linux dialup router, and the second address is assigned to an NT server on the network, so once the external router to the ISP recognizes the route, pinging 204.186.230.2 is a good test. _Any_ advice would certainly be helpful! Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Dead list?
On Thu, 5 Jun 1997, Brian N. Borg wrote: Has this list died? I've received 0 messages all day and little more in the previous 2 days. BTW: Mail is still steadily pouring in here. Mail does seem to be flowing, although I am a bit cautious due to some strange hang-ups at the primary listserver. Something along the lines of sending a message to (say) [EMAIL PROTECTED] with list subscribed in the body, and it will send back a message saying debian-user, debian-announce, ... Or perhaps a web interface to it (syncing the lists between list and web server would be interesting). Syncing the lists between two listservers is interesting, too... I've thought about how to implement such a tool. It's not easy to do so since the two listservers are running different mail transport agents and the primary is using a tricky hack to handle mail for lists.debian.org. If the server is running Majordomo, send which as the text of an e-mail message to majordomo@server in question. Unfortunately, the Debian lists are run using Smartlist, which doesn't offer the which function. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Debian-talk defunct?
On Thu, 5 Jun 1997, Max HYRE wrote: I just got a message returned from an attempt to send to the debian-talk list. (Relevant? extracts below) Is the list still operational, or did the lack of traffic lead to its demise? There hasn't been a debian-talk at least since I took over management of the lists. If there is sufficient reason to do so, I'll gladly recreate the list. - --- Start of forwarded message --- Date: Thu, 5 Jun 1997 11:53:41 +1000 From: Mail Delivery Subsystem [EMAIL PROTECTED] - The following addresses had permanent fatal errors - [EMAIL PROTECTED] (expanded from: [EMAIL PROTECTED]) - Transcript of session follows - ... while talking to mail.vv.com.au.: RCPT To:[EMAIL PROTECTED] 550 [EMAIL PROTECTED]... User unknown 550 [EMAIL PROTECTED] User unknown These errors suggest that perhaps the message was not addressed to @lists.debian.org or that vv.com.au may have other configuration oddments. Any ideas? Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Apache server-side includes questions
On Mon, 2 Jun 1997, Tom Lees wrote: Somewhere along the line, my tricks (hacks?) for doing server-side inclusion of standardized (and separately changeable) headers and footers on web pages broke. I admit to not knowing the _right_ way to do it, having stolen lots of bits and pieces along the way and learning originally on the ncsa server on my school's DEC box. Here's what I'm currently doing: !--#exec cmd=./header-- is in the page. I had an .htaccess file that looked like this: AddType text/x-server-parsed-html .html Use AddHandler server-parser .html instead, this will fix everything. What's the Right Way(tm) to include the output of a perl script into a web page? As you were doing it. Or, use a CGI, and point it at a target frame :) PS. the perl script you had there can be done without using perl: !--#include blah -- instead. Blah can !--#echo var=LAST_MODIFIED --. Much quicker, less resource-hungry. It's still showing the #includes as comments. Any ideas? http://templinux.bucknell.edu/~templin Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Apache server-side includes questions
On Mon, 2 Jun 1997, Tom Lees wrote: Ooops, that should be !--#include virtual=blah --. Does that work? Or file=blah (not sure if virtual is a 1.2 extension). Otherwise, try adding XBitHack Full to .htaccess, and chmod +x the .html files. It's still showing the #includes as comments. Any ideas? http://templinux.bucknell.edu/~templin If I view the document source, the !--stuff-- shows up in italics, as if it's still a comment (and hasn't been parsed). If I add the XBitHack Full to my .htaccess, I get the following error in my error log: [Mon Jun 2 15:39:40 1997] access to /home/templin/public_html/.htaccess failed for templinux.bucknell.edu, reason: Invalid command XBitHack If it's of any benefit, it's the apache from frozen. tcsh dpkg -s apache Package: apache Status: install ok installed Priority: optional Section: web Installed-Size: 781 Maintainer: Johnie Ingram [EMAIL PROTECTED] Version: 1.1.3-6 Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Temporary list outage should now be restored.
Debian users and developers, For at least 12 hours, the primary Debian list server has been accepting mail destined for the debian mailing lists but not passing it along to the intended recipients. The server was, however, still running. It has been rebooted and appears to be delivering mail to a long list of hosts at the very moment. Unfortunately, since it was still running and therefore still visible to hosts across the Internet, mail to the debian lists did not automatically fall back to the secondary list server unless individual Internet routes made the primary server invisible (as occasionally does occur). Now that the server has been restored, earlier postings will be processed and subscribe/unsubscribe requests will be handled. The machine is running under a high load average at the moment (18/20/14), so you can expect some delay for perhaps an hour or more. We'll investigate the cause of the problem as best we can and attempt to prevent it from reoccuring. Meanwhile, we thank you for your patience as queued mail is delivered. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: drop dead?
On Fri, 30 May 1997 [EMAIL PROTECTED] wrote: I've been experiencing sudden loss of subscription to this list. No warnings, nothing to indicate that the list might have a problem with delivering mail. One day I will have new mail from the list, the next day, nothing... and nothing afterwards. Has anyone else had this experience as well? The mailing list server logs indicate that you were unsubscribed for excessive bounces twice since the lists moved to debian.novare.net and templinux.bucknell.edu. Smartlist removes addresses if it gets a certain number of bounce messages back. Those bounce messages are usually the result of misconfigured mail servers, faulty network links, etc. If it was having trouble delivering list mail, it stands to reason that it would have trouble delivering the unsubscription message. I've recently increased the threshold for unsubscription on debian-user *only*. With the most traffic of the debian lists, it's also the most likely to suffer bounce removals. I get copies of the 200-500 bounce messages a day, so I'm fairly certain that *in general* the list server isn't making things up. If anyone finds themselves getting unsubscribed regularly, please contact me and I'll dig through the logs for further information. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Apache server-side includes questions
Hello,
Somewhere along the line, my tricks (hacks?) for doing server-side
inclusion of standardized (and separately changeable) headers and footers
on web pages broke. I admit to not knowing the _right_ way to do it,
having stolen lots of bits and pieces along the way and learning
originally on the ncsa server on my school's DEC box.
Here's what I'm currently doing:
!--#exec cmd=./header--
is in the page.
I had an .htaccess file that looked like this:
AddType text/x-server-parsed-html .html
Options All
I've basically NOT touched the srm.conf file of my frozen version of
apache. I have uncommented two lines, but recommented them for testing
purposes.
If I attempt to view the page on Netscape under Debian, I am prompted to
save the file to my home directory.
What's the Right Way(tm) to include the output of a perl script into a web
page?
Here's an example script, for those who are curious:
#!/usr/local/bin/perl
print EOM;
P
Last modified $ENV{'LAST_MODIFIED'}.
P
HR
ADDRESS
Peter J. Templin, Jr. / Bucknell University / [EMAIL PROTECTED]
/ADDRESS
P
/body
/html
EOM
It happens to work fine at http://www.bucknell.edu/~templin (the DEC box).
Thanks in advance for any/all help!
Pete
--
Peter J. Templin, Jr. Client Services Analyst
Computer Communication Services tel: (717) 524-1590
Bucknell University [EMAIL PROTECTED]
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble?
e-mail to [EMAIL PROTECTED] .
Re: failure notice
On Mon, 26 May 1997, Alexander Koch wrote:
Quoting [EMAIL PROTECTED] ([EMAIL PROTECTED]):
Hi. This is the qmail-send program at debian.novare.net.
I'm afraid I wasn't able to deliver your message to the following addresses.
This is a permanent error; I've given up. Sorry it didn't work out.
[EMAIL PROTECTED]:
Your message was addressed incorrectly. Here is a list of all of the
valid addresses in the lists.debian.org domain:
debian-admintool-REQUEST: request server for mailing list.
WHY THE FSCK DO I HAVE TO WRITE REQUEST IN capital LETTERS?
This is rather unusual. really unusual.
I was doing it the way it should be and it got bounced.
Please, Peter (?), simply add an alias from -request to -REQUEST or fell
guilty or whatever, IMO this is bad list policy.
As Ray Dassen pointed out, the mailing list that you had attempted to
subscribe to is debian-admintool, not debian-admintools.
In comment to your statements about capital letters, I reprint information
found in the Internet Engineering Task Force's RFC 822, Standard for ARPA
Internet Text Messages, section 3.4.7:
3.4.7. CASE INDEPENDENCE
Except as noted, alphabetic strings may be represented in any
combination of upper and lower case. The only syntactic units
which requires preservation of case information are:
- text
- qtext
- dtext
- ctext
- quoted-pair
- local-part, except Postmaster
When matching any other syntactic unit, case is to be ignored.
For example, the field-names From, FROM, from, and even
FroM are semantically equal and should all be treated ident-
ically.
When generating these units, any mix of upper and lower case
alphabetic characters may be used. The case shown in this
specification is suggested for message-creating processes.
The message which you received as an autoreply is merely a script, written
to clearly delineate the difference between the mailing list and the
administrative address for that mailing list. A fair number of messages
regularly come through on the incorrect address. Smartlist can catch some
of them and point them in the right direction, but not all. For the
curious, here's the script:
#! /bin/sh
PATH=/bin:/usr/bin
cat EOF
Your message was addressed incorrectly. Here is a list of all of the
valid addresses in the lists.debian.org domain:
listmaster: A human being, not a machine, your last resort.
EOF
cd /var/list
for i in */dist; do
name=${i%%/dist}
echo $name: Mailing list.
echo $name-REQUEST: request server for mailing list.
done
exit 100
Pete
--
Pete Templin, Debian List Administrator [EMAIL PROTECTED]
--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble?
e-mail to [EMAIL PROTECTED] .
Re: May 13 instalation disk's problem (vfat)
On Mon, 19 May 1997, Eugene Sevinian wrote: P.S. Just for curiosity ... HOw many people are participating in this mailing list? The list has 854 subscribers currently, and I've seen it slightly above 1000 at times. There are another 191 on the -digest form of the list. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Xwindows running finally!! what's next?
On Fri, 16 May 1997, Tan Wee Yeh wrote: twm [background processes snipped] xterm -title Measun10 -geometry 80x40+30+200 -ls Just a little comment (for discussion). Normally, I will prefer to put twm (or whatever windows manager) as the last process (let the others be background). This causes the window manager, rather than other processes, to be the anchor process, so that X kicks the user off when the windows manager quits. I wonder if this is a good practice. I've gotten bitten enough times (having just set up a friend's machine and not fully tested it yet) with the window manager as the final process but didn't yet have my mouse working. I'd often have to forcibly kill X, which I tend not to like to do. If I'm using my normal fvwm, I can (once I remember the default keys)drive my cursor by keyboard and exit an xterm, but there isn't much option for killing the window manager by keyboard. For the curious, here's a (slightly trimmed) version of my .xinitrc: #! /bin/sh # # first things first. # # xhost - allows x clients onto this server. xhost coral reef templinux # Backgrounded jobs process in parallel. # # xclock - self explanatory xclock -bg black -fg red -hd white -update 1 -geometry 96x96+3+809 # xbiff - visual new mail. xbiff -update 3 -volume 0 -bg black -fg blue -geometry 96x96+110+809 # xsysinfo - kernel activity/usage indicator xsysinfo -geometry 206x100+0-0 # fvwm - window manager fvwm xearth -proj merc # xautolock...automatically xlocks when unused for ten minutes xautolock -locker xlock -mode maze # xcon - symlink to xterm (allows for different behavior in .fvwmrc xcon -geometry 80x6+216-0 -T daemon.log -n daemon.log -e tail -f /var/log/daemon.log # NON BACKGROUNDED: # this script (and therefore xinit) finishes when # this process ends. # xcon - symlink to xterm (allows for different behavior in .fvwmrc xcon -C -geometry 80x6+711-0 ---end of included script--- Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Lyx: Where is Xforms? (fwd)
Messages to the Debian mailing lists will NOT make it to the list recipients if they are sent as root. Here's one that was diverted to [EMAIL PROTECTED] for that reason, and should have gone to the list. -- Forwarded message -- Date: 14 May 1997 20:27:16 - From: Victor Torrico [EMAIL PROTECTED] To: Pedro I. Sanchez [EMAIL PROTECTED] Cc: Paul McDermott [EMAIL PROTECTED], Debian-user debian-user@lists.debian.org Subject: Re: Lyx: Where is Xforms? Pedro I. Sanchez wrote: Well, it seems that the current Lyx package is still looking for Xforms, not Xforms0. As I said in my original message, I already installed Xforms0 but Lyx insists in having Xforms. Given that Xforms0 is already installed, is there a way to force dpkg to configure Lyx? (I tried --force-configure-any but it didn't work). Paul McDermott wrote: yes lyx depends on xforms0 On Wed, 14 May 1997, Pedro I. Sanchez wrote: Hello, I'm trying to install LyX version 0.10.7-2 but it fails because the Xforms package is not installed. However the only package which seems to be available is xforms0 (which I installed succesfully). So, what is Xforms0? Is it supposed to replace Xforms? Is Lyx supposed to depend on Xforms0? Thank you, -- Pedro I. Sanchez Product Manager CTI Datacom Inc. 514.683.6363 x31 -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] . -- Pedro I. Sanchez Product Manager CTI Datacom Inc. 514.683.6363 x31 You can install both packages to make Lyx and other programs work. First purge both packages. Then install xforms and rename the library to something else. Then purge xforms which leaves the library under the new name. Then install xforms0. Then rename the xforms library back to it's original name. You now have both libraries and everything works fine. I did this and Lyx works OK as well as programs which rely on xforms0. Cheers, Victor -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Problems adding swap files
Hello! I'm having some difficulty creating special swapfiles to be able to test swapping to/from an md device (raid0). I've followed the manpage hints as indicated below, and think that it's complaining about the individual file(s). Can anyone guide me in the right direction? tcsh# df /server/ Filesystem 1024-blocks Used Available Capacity Mounted on /dev/md0 5970540 3908064 1753225 69% /server tcsh# cd /server/ tcsh# mkdir swap tcsh# cd swap tcsh# dd if=/dev/zero of=swapfile bs=1024 count=8192 8192+0 records in 8192+0 records out tcsh# mkswap swapfile 8192 Setting up swapspace, size = 8384512 bytes tcsh# sync tcsh# swapon swapfile swapon: swapfile: Invalid argument tcsh# Any ideas? Thanks, Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: debian mirrors
On Sat, 10 May 1997, Bob Nielsen wrote: What are some good up-to-date mirrors of ftp.debian.org? I am continuously having problems connecting because of the 100 anonymous user limit. I have tried some mirrors, like ftp.infomagic.com, but not everything is there. I've got a twice-a-day mirror of master.debian.org on templinux.bucknell.edu. It's not yet registered as an official Debian mirror because we're soon changing our internal IP address structure and perhaps also our Internet provider. With those changes on the horizon, I don't want to provide a flaky mirror. Some extras about my site: I still have buzz (Debian 1.1) on there, so anyone who's looking for some legacy packages should be able to find them. I also keep most everything that disappears from master.debian.org in /pub/debian/deletes/former_path_to_package/package So if you find that you have a problem with a new package, chances are you'll be able to find the old one on my site. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
inquiry about exploder mailing lists of Debian lists
As list manager, I often assist people in the process of unsubscribing from Debian lists. Occasionally, I end up with a person who is receiving Debian mail from a mailing list that is _on_ a Debian mailing list. Many of these lists serve excellent purposes (they're based in a locality that has slow external connectivity, etc), but can make it difficult for a subscriber to realize that the normal method of unsubscribing doesn't work. To assist with managing this dilemna, I'd like to ask anyone who currently operates/manages/oversees any of those lists to contact me with the following information: Your name/email What debian list(s) you explode What address receives the mail from the first-tier Debian lists Approximately how many subscribers are on your list(s) I'm not going to ask you to stop what you're doing. I'm merely hoping to get a sense of how many there are, and attempt to keep track of who I can contact to assist with problems. Thanks for your help, Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: Posting restrictions
On Mon, 5 May 1997, Robert D. Hilliard wrote: When I first subscribed to debian-* lists, I was required to agree to a number of anti-spam restrictions before being allowed to post to the list. Violation of this agreement would result in being removed from the list of authorized posters, and perhaps other penalties. The debian mailing lists were moved and recreated sometime during February, and I volunteered to handle list management duties in early March. As a result of those transitions (and at the advice of Bruce Perens, who preceded me in my present role), there no longer exists a spam filter. I'd be willing to consider this if there is suitable request from the crowd. There was an obnoxious posting this morning from some creature offering guanteed credit, and I have seen a few similar messages from time to time, and there is the evangelist who posts a long sermon about once a month. Is the authorized posting list still being used since the list server was moved? If so, are these offenders being purged from the list? Was that on the -user list? I didn't catch it. I've been on and off of the user list, due to a 1.5 week vacation/training session and the job pressures that result. Now that life is settling back down, I'm planning on keeping up with -user regularly. If anyone is offended, bothered, or in any way disturbed by a post, please feel free to contact me. I'd be happy to handle the process of pestering the poster and the persons responsible for allowing the post to happen. I do purge obnoxious posters if they are on the list (I did so last week, after getting no response to a simple question). If there are other actions you'd like me to take as Minister of Information (aka mailing list manager), please contact me. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
kernel build (fwd)
Some messages get rejected by Smartlist (our list handler) because they appear to be coming from a daemon (root in this case). Here's a message from Ralph Winslow. -- Forwarded message -- Date: 3 May 1997 17:34:55 - From: Ralph Winslow [EMAIL PROTECTED] To: Debian Users debian-user@lists.debian.org Subject: kernel build Since upgrading many packages in the frozen distribution, I'm unable to re-build a new kernel (I need to add modules support for the OSS sound package). When I cd to /usr/src/linux (kernel-source-2.0.27) and try make mrproper (or make menuconfig or make config or make all), I get make: *** No rule to make target 'mrproper'. Stop. I tried un - re -installing make as well as the kernel-source package for 2.0.27 but no joy. Could this relate to having inadvertantly started an install of 2.0.29 at some point in the past? Some traces of 2.0.29 remain, BTW. Any help on this would be deeply appreciated. -- - Ralph Winslow [EMAIL PROTECTED] Someday soon I really MUST find a way to piss away a LOT of bandwidth on this .sig -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
hot-change disk arrays...can I do it on Linux?
Disk array question: Our computer center here has several NT servers with RAID5 disk arrays (4x4GB disks, so 12GB useable). I'm curious about some implementation questions regarding RAID arrays. I'll try to separate my questions, so that informed people can set me straight on each of the issues. Hot-swap drives: this is a functionality of the drives, right? I'd have to have disk drives that were manufactured to stand up to that, correct? Assuming that I have a hot-swap disk in my machine and a spare disk on my shelf, and that I was not using RAID or anything special on that disk, would I need a special controller or customizations to the OS to use it (i.e. be able to take out the failed drive and stick in the new (formatted) drive)? Is hot-swap only for RAID arrays? Hardware RAID: Hardware RAID is independent of the operating system, right? Aside from software to control/tweak the array parameters, I need only to have driver support for the controller, right? Hot-growth arrays: Our NT servers have the ability to add a disk to the RAID array live on the fly. All the administrator has to do is tell the controller to add the new disk to the array and BINGO! bigger array. Is that a functionality of the controller? What other parts of the computing system need to be modified to support that (i.e. can I do that with my Debian systems)? Thanks for the help. I'm soon embarking on a development project from the ground up for a local computer store and want to plan things right from the beginning. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Sendmail performance tuning question
I'm running into some errors trying to use some of the more advanced performance tuning features of sendmail 8.8.5. If you're responsible for a debian system running sendmail and you're using any of these features: Single Threaded Delivery - the most important one to me Host Status Directory Deferred delivery mode please contact me. Thanks! Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: bi (Please stop it)
On Wed, 16 Apr 1997, Leslie Mikesell wrote: The issue relevant to this group is: what editor should someone expect to find on a system's boot/rescue disk? That someone presumably being a person with enough unix experience to recover from the usual problems that can make your machine fail to boot. The last thing you need at that point (especially if this is a server for many people) is a surprise from the editor or to have to learn a new one. So why is the issue that _seems_ to be relevant to the group (or at least the posters within) the minimization of the number of keystrokes or the level of injury supposedly inflicted by its interface? Besides, wouldn't a discussion of an appropiate boot/rescue disk editor be better suited for the developer's list? It would seem to me that they are the ones responsible for developing the actual boot disks. I agree with Joey's original message: let's let the editor debate rest a bit, folks, or give it focus and a new thread name. Thanks, Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
Re: [HELP] root damaged
On 14 Apr 1997, Linh Dang wrote:
Yesterday, I booted up my buzz (Debian-1.1) box and surprise ! I in root
account with right away !
An `ls -l' on the root dir showed my that `/sbin/' is now a huge _FILE_
I can't read floppy, cdroms ... because I can't load modules since insmod and
co. are in `/sbin/' !!! :-((( (My Debian-1.1 CD is lost somewhere :-) ...
Anyone knows a ftp which still carries 1.1 Please
templinux.bucknell.edu has /debian/buzz{,-fixed,-updates}
Pete
--
Peter J. Templin, Jr. Client Services Analyst
Computer Communication Services tel: (717) 524-1590
Bucknell University [EMAIL PROTECTED]
Re: deity mailing list
On Mon, 14 Apr 1997, Martin Schulze wrote: On Apr 14, Ryan Shaw wrote the poster also mentioned a mailing list to discuss to new program and its development. however, upon browsing www.debian.org i couldn't find any mention of the new list. could someone point me in the right direction and/or perhaps validate the claims made by the poster in c.o.l.m? send a message to [EMAIL PROTECTED] and you'll receive list of valid adresses - containing deite I suppose. That mailing list exists for the members of the deity team. Other debian users are welcome to post to it, but it is not open for subscription. We should be hearing about progress on the project from Brian White, who is leading that project. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED]
Re: UNSUBSCRIB ME RIGHT !@#$%^ NOW
Debian Users, I removed Nathan from the list, at his albeit loud and rude request. I've asked him to contact me if he had any further complaints. I'd like to ask that we as a group not flame him, to prevent further postings in a similar manner. Hopefully, he'll quietly settle. Thanks, Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED]
RE: bi
On Sat, 12 Apr 1997, Rick wrote: Sorry. I was watching these vi msgs go back and forth and had to jump in and make it worse. I think everyone should use whatever they want to. I agree 100% about emacs. I have better things to do than to memorize all that crap. That's why I use the GUI version, menu's. Agreed. Let's chill this thread a bit, folks. Everyone, go back to your computers and use an editor, any editor. :) Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Re: No nfs, no boot
On Mon, 7 Apr 1997, Steve Hsieh wrote: Now the obvious solution is to fix up the other machine, but the thing which worried me most was the fact that the attempt to nfs mount didn't timeout. Which means that my system is entirely dependent on the other system in order to boot. I am guessing that something is probably trying to access a file on the NFS mounted drive. You could try using the soft option when mounting it. From the nfs man page: soft If an NFS file operation has a major time- out then report an I/O error to the calling program. The default is to continue retry- ing NFS file operations indefinitely. Also from nfs(5): bg If the first NFS mount attempt times out, continue trying the mount in the back ground. The default is to not to back ground the mount on timeout but fail. This is almost necessary if two NFS servers are clients of each other. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
request for help debugging my dump problem
I've been having a problem with dump hanging at 100% cpu hog during my nightly backups and when run interactively. I've strace'd the dump command that I use (strace dump 0udf 325000 /dev/rft0 / ! ~root/dump.out) and copied it to my public ftp area. Would anyone be willing to look at it and tell me if I've done the right stracing, and perhaps what the problem might be? The output is available at: templinux.bucknell.edu/pub/dump.out Thanks, Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
xbigfoot message seen on debian-user
Debian Users, I recently tracked down what I believe to be a problematic email address which was on several of the debian mailing lists. I think that the address would send mail back to anyone/thing which sent it mail, causing a mailing loop. Worse yet, when I unsubscribed the address, it sent mail back to the list address of some of the lists it was on. Unless otherwise notified, there is no reason to fear that you've been removed from our mailing lists. Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] On Thu, 3 Apr 1997, Petr Barta wrote: I don't know if this mail is sent to me accidentally or it is a spam, but I did not send this message. It is send to me, either it has different To: field from my address ([EMAIL PROTECTED]). I'm writing this just in case I should be removed from this list. 32760 [EMAIL PROTECTED] [EMAIL PROTECTED] You have been removed from the list. If this wasn't your intention or you are having problems getting yourself unsubscribed, reply to this mail now (quoting it entirely (for diagnostic purposes), and of course adding any comments you see fit).
Dump hangs on my system
My nightly backup script seems to be hanging in the middle of its first dump: DUMP: Date of this level 0 dump: Wed Apr 2 00:11:29 1997 DUMP: Date of last level 0 dump: the epoch DUMP: Dumping /dev/hda2 (/) to /dev/rft0 DUMP: mapping (Pass I) [regular files] DUMP: mapping (Pass II) [directories] DUMP: estimated 14710 tape blocks on 0.01 tape(s). DUMP: dumping (Pass III) [directories] ps listing: root 722 92.1 0.6 1016 400 p3 R 00:11 464:07 /usr/local/sbin/dump Any ideas, suggestions on how to fix? There have been no recent changes to the system, except for the removal of some development libraries via dselect. Thanks, Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Re: Secret debian lists? (was: Debian Book list)
On Mon, 31 Mar 1997, J.P.D. Kooij wrote: Still, while the list does exist, it doesn't appear on the list of lists that comes with the subscription failure notification. Which raises the question: what happened to debian-admintool (for rantings and whinings about dselect)? It was mentioned some time ago, but it doesn't appear on the list of debian-lists either? That problem is now fixed. Funny thing is: the only action necessary to fix it was chmod o+r /var/list/listname Go figure. :) Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED]
Re: improving pine speed when using remote smtp server
On Tue, 1 Apr 1997, Frank Swasey wrote: On Tue, 1 Apr 1997, Douglas L Stewart wrote: What I'd _like_ to do is to set up smail on my laptop where it'll forward the mail to the mailhost. This isn't really a problem, and there seems to be an option to do this when the smail package isn't configured. I don't know if you can do it with SMAIL, but there is a mini-howto on how to do this with SENDMAIL -- The title is something about sendmail and queueing. I set it up once long ago and haven't touched it since (so I've long since forgotten the name of the howto I used), but it walks you through customizing the sendmail.cf file so the mail is queued and then only delivered when sendmail runs the queue. I set up my system so that sendmail does not run the queue when I'm not connected to the Internet. When I connect to the Internet, sendmail is stopped and then started with the necessary parameters to run the queue (-q10) every 10 minutes. --for sendmail, not smail-- You can either edit your /etc/init.d/sendmail to read start-stop-daemon --start --verbose --exec /usr/sbin/sendmail --\ -bd -odq -q1m or change your /etc/mail/sendmail.cf entry for: # default delivery mode O DeliveryMode=background - change this to be queue and/or change your scripts to use sendmail -odq recipient. Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Mailing list information
Debian Users, I'd like to provide some interim information which some may find helpful to explain some of the extra mail which has come through, some of the reasons why people who unsubscribed are thinking that they are still subscribed, and why others haven't seen their posts coming through the lists. The official hostname for the lists is lists.debian.org. Please use it for all of your postings and subscription/unsubscription requests (remember, send subscribe or unsubscribe to the REQUEST form of the list, as debian-listname[EMAIL PROTECTED]). At one point, master.debian.org (also known as debian.org by IP address) was acting as lists.debian.org for list delivery. There are some legacy mailing lists on that machine from approximately one month ago. I have requested that the folks who created those lists redirect any incoming posting to the appropriate address. In the interim, you may see email from [EMAIL PROTECTED] Hopefully those lists will soon either point to the correct list addresses or contain up to date subscriber information. If you think that you are subscribed to those legacy lists, you do not need to worry...we will take care of the situation in the near future. The mailing list processor, Smartlist, does not accept submissions from usernames which appear to be system users (i.e. root, admin, daemon, bin) for either posting or subscription/unsubscription. It is not recommended that anyone utilize root or other system accounts for ordinary mail and news purposes. Please utilize your normal accounts for the purposes of the Debian mailing lists (among other things) If you have any questions about the above information or the lists in general, please send email to [EMAIL PROTECTED] and I or someone else will provide you with answers. Thanks, Pete -- Pete Templin, Debian List Administrator [EMAIL PROTECTED]
md autostart?
Hi there, I just put the md package to use, and I have some semi-important questions... 1) Do I need to do anything to ensure that my /dev/md0 has been started if I expect to use it in /etc/fstab? If so, where do I put that? My kernel has md compiled in (not a module). 2) Can I also assume that it'll get unmounted cleanly and properly? Thanks for your help, Pete -- Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Possible problems with lists.debian.org
Debianers, It's possible that there are problems with debian.novare.net, aka lists.debian.org. It's also possible that I'm just losing it, or that my employer's internet route to that site is out to lunch. I've contacted someone at novare.net, and I'll pass word along as I find out more. -- Pete Templin, Debian List Administrator [EMAIL PROTECTED] Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = BD 9D 90 1C 8D 6D CA 21 D7 0F 2D C6 29 93 A6 1E
lists.debian.org seems unreachable
Debianers, Debian.novare.net (aka lists.debian.org) seems to be unavailable at the moment. I will check with my contacts to find out more information, but for the time being we cannot expect normal list mail to be flowing. -- Pete Templin [EMAIL PROTECTED] (717) 966-9656 Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = BD 9D 90 1C 8D 6D CA 21 D7 0F 2D C6 29 93 A6 1E
Re: Keep getting removed from list
On Tue, 18 Mar 1997, Tim O'Brien wrote: About twice a day, I get removed from debian-users because of bounced messages. Is this happening to anyone else? I am getting I think that's what's happening to me too... And I have no idea why, but I'd like to find a way to fix it. Any ideas? Smartlist, the mailing list handler which handles the debian-* mailing lists, automatically removes a subscriber if mail to that address bounces more than a configured number of times. Is it possible that your mail reception point isn't accessible around the clock? If anyone experiences this, please contact me and I will attempt to track it and/or fix it. I'm new to managing these lists, so please bear with me. Thanks, Pete -- Pete Templin [EMAIL PROTECTED] (717) 966-9656 Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = BD 9D 90 1C 8D 6D CA 21 D7 0F 2D C6 29 93 A6 1E
Re: HOw much space do I need to mirror stable and unstable ?
On Sun, 16 Mar 1997, Stan Brown wrote: HOw much space do I need to mirror stable and unstable ? That's a tough question, only because Debian is on the edge of a transition between stable releases. Here's what I can tell you: stable - rex-fixed (you'll need the following three dirs I think) rex is 488M rex-fixed is 5M rex-updates is 116M development - unstable unstable - bo (soon to be frozen I think.) bo is 620M *future unstable* - hamm hamm is 3K (nothing really there yet) A full mirror of ftp.debian.org/debian is about 1.5GB (there's a lot of other stuff on the site). -- Pete Templin [EMAIL PROTECTED] (717) 966-9656 Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = BD 9D 90 1C 8D 6D CA 21 D7 0F 2D C6 29 93 A6 1E
Re: Ideal partition sizes.
On Sun, 16 Mar 1997, Paul van Berlo wrote: to dedicate this machine to Linux now (1,2gb). What would be the ideal partition sizes to split up the hdd for Linux? It'll be used for I have two machines. I dislike extended/logical partitions. I like performance tuning. My machines have at least two drives in them. All that aside, here's the scoop: Machine 1 (Daily X work, text processing, ftp mirrors): hda (1.2 WD): hda1: 300M spare (more on this later) hda2: 40M / hda3: 120M swap hda4: rest /usr (757M, there's no real need for that much) hdc (1.2 WD) (I'm going out of order for a reason) hdc1: 500M /local (gets /home, /usr/local, /var/spool/mail, see below) hdc2: 250M /tmp hdc3: 120M swap hdc4: 350M /var (no need for that much, but 100M helps when installing) hdb (3.1 WD) hdb1: (all) /server (gets my anon ftp area and *if I add it* news). Machine 2 (Was going to be a mailing list server): hda (850M Quantum) hda1: 200M spare hda2: 40M / hda3: 120M swap hda4: rest /usr (445M, there's no real need for that much) hdc (850M Quantum) hdc1: 300M /local (see above) hdc2: 100M /tmp hdc3: 120M swap hdc4: 300M /var (I wanted this big for large email logfiles) For the most part, this scheme works exceptionally well. My spare partition holds a complete (and completely separate) linux install (single partition, just like you had). If I should ever trash my main install, I can boot into the spareland and run a tape backup of whatever's left - very handy to have around. My /local preserves important files whenever I decide to do a complete scrub and reinstall. And with both spare and /local as the first partitions on each disk, it's very easy to resize my partitions (assuming I'm reinstalling) without destroying my safety net and/or my personal files. Both of my swap areas are directly between critical system areas (/ and /usr, /tmp and /var), making swap an easy trip from any typical program, logfile access, or tempfile. The biggest performance hit comes when sendmail is delivering to my mailbox (which is in /local) and is logging to /var at the other end of the disk, but some recent sendmail tunings have really helped that. To summarize, here's a df from each of my systems: Machine 1: Filesystem 1024-blocks Used Available Capacity Mounted on /dev/hda2 39039 1196525058 32% / /dev/hda4 757151 244581 473460 34% /usr /dev/hdc4 349343 18561 312739 6% /var /dev/hdc2 247871 43 235027 0% /tmp /dev/hdc1 495714 23008 447105 5% /local /dev/hdb12990073 2305554 529894 81% /server Machine 2: Filesystem 1024-blocks Used Available Capacity Mounted on /dev/hda2 40007 1127226669 30% / /dev/hda4 445949 157541 265376 37% /usr /dev/hdc4 287850 16326 256656 6% /var /dev/hdc2 99061 1393933 0% /tmp /dev/hdc1 2976034080 278153 1% /local Hope this helps, --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Cron root@templin-wks backup save (fwd)
I'm running into some sort of filesystem corruption when I dump one of the filesystems on my spare machine. Anyone got any ideas? The machine runs real well, and came up A-OK after a reboot the other morning. Thanks in advance, -- Pete Templin [EMAIL PROTECTED] (717) 966-9656 Finger [EMAIL PROTECTED] for PGP public key. PGP fingerprint = BD 9D 90 1C 8D 6D CA 21 D7 0F 2D C6 29 93 A6 1E -- Forwarded message -- Date: Tue, 11 Mar 1997 23:03:53 -0500 From: Cron Daemon [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Cron [EMAIL PROTECTED] backup save DUMP: Date of this level 0 dump: Tue Mar 11 23:03:53 1997 DUMP: Date of last level 0 dump: the epoch DUMP: Dumping /dev/hda2 (/) to /dev/rft0 DUMP: mapping (Pass I) [regular files] DUMP: mapping (Pass II) [directories] DUMP: estimated 12869 tape blocks on 0.01 tape(s). DUMP: dumping (Pass III) [directories] /dev/hda2: EXT2 directory corrupted while converting directory #24 mt: /dev/nrft0: Device or resource busy DUMP: Date of this level 1 dump: Tue Mar 11 23:03:54 1997 DUMP: Date of last level 0 dump: Mon Mar 3 23:08:40 1997 DUMP: Dumping /dev/hda4 (/usr) to /dev/rft0 DUMP: mapping (Pass I) [regular files] DUMP: mapping (Pass II) [directories] DUMP: estimated 333 tape blocks on 0.00 tape(s). DUMP: dumping (Pass III) [directories] DUMP: master/slave protocol botched. DUMP: The ENTIRE dump is aborted. mt: /dev/nrft0: Device or resource busy DUMP: Date of this level 1 dump: Tue Mar 11 23:04:12 1997 DUMP: Date of last level 0 dump: Tue Mar 11 16:56:52 1997 DUMP: Dumping /dev/hdc4 (/var) to /dev/rft0 DUMP: mapping (Pass I) [regular files] DUMP: mapping (Pass II) [directories] DUMP: estimated 9054 tape blocks on 0.01 tape(s). DUMP: dumping (Pass III) [directories] DUMP: dumping (Pass IV) [regular files] DUMP: DUMP: 9055 tape blocks on 1 volumes(s) DUMP: level 1 dump on Tue Mar 11 23:04:12 1997 DUMP: Closing /dev/rft0 DUMP: DUMP IS DONE mt: /dev/nrft0: I/O error DUMP: Date of this level 1 dump: Tue Mar 11 23:07:59 1997 DUMP: Date of last level 0 dump: Mon Feb 3 23:52:42 1997 DUMP: Dumping /dev/hdc1 (/local) to /dev/rft0 DUMP: mapping (Pass I) [regular files] DUMP: mapping (Pass II) [directories] DUMP: estimated 4933 tape blocks on 0.00 tape(s). DUMP: dumping (Pass III) [directories] DUMP: End of tape detected DUMP: Closing /dev/rft0 DUMP: Change Volumes: Mount volume #2 DUMP: fopen on /dev/tty fails: No such device or address DUMP: The ENTIRE dump is aborted.
problem with mirror after transition to 1.2.8
Hello, I'm having trouble with mirror, and I think the problems have arisen since I upgraded from 1.2.7 to 1.2.8. Here's some example errors: package=debian ftp.debian.org:/debian/ - /server/ftp/pub/debian/ main:/usr/bin/mirror:2205 Caught a SIGSEGV shutting down at /usr/bin/mirror line 3601. ls -l on a certain file gives: ?x 0 root daemon 16842757 Jul 14 1970 www-search_1.007-1.diff.gz and that file can't be rm'ed, chmod'ed, or written over. Any ideas or fixes? --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Re: pine producing gratuitous folder locks
On Fri, 7 Mar 1997, David C. Winters wrote: Today, I started getting running into a problem with Pine 3.94--it began telling me that every folder I tried accessing was locked. This persisted even after I rebooted the machine, so I upgraded to Pine 3.95q out of the 3.95L-7 .deb package. This resulted in the same thing--if I try to open a folder, it's locked. After looking through /usr/doc/pine/tech-notes, I checked for whatever.lock files corresponding to the various folders, as well as a general lockfile in either /var/lock or /tmp without finding anything promising. Has anyone else run into this with this package? Are the permissions on the /tmp directory correct? (1755, which looks like drwxrwxrwt in ls -l /) This is often my first clue that /tmp has the wrong perms. Then again, it could be some other problem... --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Re: turning off computer
Your best bet is to use the sudo program or some such. You can establish groups of users and allow them to run predefined programs as root by feeding sudo their OWN password. For sanity sake, set your EDITOR environment variable to the editor of your liking and then use the visudo command to edit the file which defines the permissions - visudo will check for syntax errors, very important for a tool which can allow some powerful permissions. Once configured, a permitted user merely types sudo reboot and feeds it his/her password. Voila! --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] On Wed, 5 Mar 1997, I Brake for Moths wrote: Is there a way for a user without root priveleges to cleanly unmount the root file system and shutdown the computer? I've been running 'init 0' as root before turning off the box, but I don't want to have to give out the root password to my family (it's bad enough that I know it!) just so they can turn the thing off when I'm not around. We'd rather not leave it on when it is not in use. Thanks, Rikki *** If you have already paid your bill, please disregard this notice. ***
Re: Package configuration philosophy
On Tue, 25 Feb 1997, [iso-8859-1] Nicolás Lichtmaier wrote: On Wed, 26 Feb 1997, Craig Sanders wrote: # and $ are standard/expected prompts. if you want something different, customise it yourself. He's right Debian should provide a nicer default for the prompt. Many people take this things into account when deciding which distribution they like best. If someone is going to evaluate an entire distribution on a prompt (even if there are other factors), I'm not going to be upset if they don't choose Debian. Perhaps we could/should point people towards how to change it, but I see NO reason to depart from the standard. If we did decide to provide a fancy prompt, I'd bet the ensuing chatter about what it should be would create more traffic than the regular noise about dselect's supposedly impossible interface. I like my own prompt of `PS1=(\h-\u) [\t] \w\$ ` (looks like: (siva-cas) [12:14:32] ~$ ) but I wouldnt force everyone else to use it...it takes me 5 seconds on every new debian machine i build to edit ~/.bashrc as needed. [EMAIL PROTECTED]:/usr/doc/fvwm2$ _ newton:/usr/doc/fvwm2$ _ But, for those of you looking for custom prompts, here's mine (I use tcsh, mileage using bash fuel may vary): set prompt = -- %d %T on %m : pwd is %~ \ tcsh%# which produces the following two line prompt, always ready at position 6: -- Tue 21:17 on templinux : pwd is ~/files/personal tcsh --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: UPS questions
On Sat, 22 Feb 1997, Giuseppe Vacanti wrote: Is anybody using a UPS with debian who would care to email how they are faring? Is APC a good UPS to buy? Do I need a special cable? And how should I go about sizing the UPS (I mean, do I need a 200W UPS if the power supply of my box is 200W)? I'm using an APC Smart-UPS 1400 that is communicating only with my Win95 machine. Using the software, I've been able to determine the following usages: Pentium 133/32MB/512K with 2 or 3 HDs: 4.5% of capacity each (I have 3) Apple PowerMac 8100/100AV/40MB with 1 HD: 7% of capacity Apple 15 Multiscan monitor: 8% of capacity MAG 21 monitor: 11% of capacity Those Pentium numbers indicate that each is using about 40W. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Meta key: what's the right/best way?
Hi all, I've got a MS Natural Keyboard on my main system, and I'd like to have it configured such that Alt serves as the Meta key both in the virtual consoles and in X. I'd also like to have consistent behavior on any xterms which I then telnet to another machine, along with xterms/emacses launched on another machine pointed to my X display. I've looked at kbdconfig/loadkeys, I've looked at xf86config, and I can't make sense of the _right_ way to do it. I'd also like to know how to get backspace to be a backspace and delete to be a delete consistently in all of the things I do (xterms on localhost, xterms telnetted to another host, emacs locally, emacs in a xterm telnetted to another host, emacs remote onto my X display) - what's the right way to do it? Thanks in advance, --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
missing/non-functional identd service?
Hi there. I've been learning about PGP, and was in the process of taking a look at a few public keys, when I ran into a small stumbling block. You see, I was fingering [EMAIL PROTECTED] when I ran into a small obstacle: tcsh finger [EMAIL PROTECTED] [master.debian.org] Debian Linux 1.1 Copyright (C) 1993-1996 Debian Association, Inc. and others Your site has been rejected for some reason. This may be caused by a missing RFC 1413 identd on your site. Contact your and/or our system administrator. -- Tue 21:02 on templinux : pwd is ~ tcsh Since my system is also Debian (why else would I be sending this to deb user?), I decided to try it against my system from a machine which I knew suffers from a lack of identd: 20:48 on charcoal : ~ finger [EMAIL PROTECTED] [templinux.bucknell.edu] Your site has been rejected for some reason. This may be caused by a missing RFC 1413 identd on your site. Contact your and/or our system administrator. 20:54 on charcoal : ~ Is this a configuration error I need to fix, a package I haven't installed, or something else? It would seem best if Debian systems could at least finger Debian systems, right? Thanks in advance for the help, --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Two unrelated problems with cron
On Wed, 22 Jan 1997, Kevin McEnhill wrote: The other problem I have is in updating cron. I downloaded cron_3.0pl1-38.deb from ftp.debian.org and used 'dpkg --install'. Well dpkg crashes with the following message. bash# dpkg --pending --configure Setting up cron (3.0pl1-38) ... /usr/sbin/cron: can't lock /var/run/crond.pid, otherpid may be 5014: Try again # kill 5014 (resume install mode in dselect) Just did it last night on two machines, each of which seem to have successfully survived a reboot later in the night. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
kernel panics, crashes.
Hello all. Saturday night, while I was testing my tape backup procedure, I did a full restore into an unused partition. I tried an rm -r on that partition, and got a kernel panic, locking my system. Happened again later that night. Earlier today, while demonstrating the slowness and cpu usage of IDE, I was copying a directory to another place on the same filesystem. Locked up silly. Anyone have ideas as to what I'm doing wrong? --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Best Debian CD?
On 15 Jan 1997, Steve Dunham wrote: The guy at LSL says that he will make a Debian-only CD (with the whole thing on it) if we can come up with a single floppy installation system (he will include the single floppy too). (He also said something about helping to support the project too.) Hmmmseems to me that Windows NT is a three floppy system. Although three is certainly less than six, three is significantly more than one. My advice to those looking to install Debian via floppy: Use brand new floppies. Format then under DOS. If you get any bad sectors, you got a bad disk. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- This message was delayed because the list mail delivery agent was down.
Re: Two last problems...
On Fri, 17 Jan 1997, Daniel S. Barclay wrote: The first problem is that syslogd is keeping the load on my machine at 1.0 even if nothin else is happenning with the system. At the same time I get huge numbers of 'The last message repeated 123456 times' appearing in /var/log/messages' Is there a way to get syslogd to work correctly? This happens when one of the directories listed in /etc/syslog.conf doesn't exist. /var/log/news is usually the culprit. Workarounds are fine, but would someone _please_ report this as a bug so it can get fixed permanently? Are you allergic to bug reports? Actually, it probably isn't safe for a script to remove these lines, as they might break something else which references them. Syslogd.conf is a fantastic traffic cop, but it can't do its job unless the right entries remain in the file. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- This message was delayed because the list mail delivery agent was down.
Re: Making kernel using make install
On Fri, 17 Jan 1997, Victor Torrico wrote: When making a kernel 2.0.27 I do the following: make mrproper make config make dep make clean make zImage make modules make modules_install make install The make install is not documented in the /usr/src/linux directory as far as I know but when it is used it seems to put everything from the new kernel where it belongs properly in the /boot directory and lets you update lilo as well. I just tried doing this for the hell of it and it seems to work very well. Ah-hah! Finally, what seems to be a simple sequence of commands for building a new kernel. But what must I do to ensure that my old kernel will continue to work (with its modules), especially if lilo wants to complain that the new kernel is too large? I assume that certain files and directories ought to be backed up or renamed or something, but some pointers to safe kernel testing would be very helpful! --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
to help us lighten up...
There have been a few flame wars and other discussions going back and forth. Although many of the topics can certainly offer good criticisms when taken with a shake of salt, perhaps we need something besides the norm here. Let's not turn our wonderful list into a jokes-only list, but I just want to try something different for a change. So, here we go... If Operating Systems Were Beers... DOS Beer: Requires you to use your own can opener, and requires you to read the directions carefully before opening the can. Originally only came in an 8-oz. can, but now comes in a 16-oz. can. However, the can is divided into 8 compartments of 2 oz. each, which have to be accessed separately. Soon to be discontinued, although a lot of people are going to keep drinking it after it's no longer available. Mac Beer: At first, came only a 16-oz. can, but now comes in a 32-oz. can. Considered by many to be a light beer. All the cans look identical. When you take one from the fridge, it opens itself. The ingredients list is not on the can. If you call to ask about the ingredients, you are told that you don't need to know. A notice on the side reminds you to drag your empties to the trashcan. Windows 3.1 Beer: The world's most popular. Comes in a 16-oz. can that looks a lot like Mac Beer's. Requires that you already own a DOS Beer. Claims that it allows you to drink several DOS Beers simultaneously, but in reality you can only drink a few of them, very slowly, especially slowly if you are drinking the Windows Beer at the same time. Sometimes, for apparently no reason, a can of Windows Beer will explode when you open it. OS/2 Beer: Comes in a 32-oz can. Does allow you to drink several DOS Beers simultaneously. Allows you to drink Windows 3.1 Beer simultaneously too, but somewhat slower. Advertises that its cans won't explode when you open them, even if you shake them up. You never really see anyone drinking OS/2 Beer, but the manufacturer (International Beer Manufacturing) claims that 9 million six-packs have been sold. Windows 95 Beer: You can't buy it yet, but a lot of people have taste-tested it and claim it's wonderful. The can looks a lot like Mac Beer's can, but tastes more like Windows 3.1 Beer. It comes in 32-oz. cans, but when you look inside, the cans only have 16 oz. of beer in them. Most people will probably keep drinking Windows 3.1 Beer until their friends try Windows 95 Beer and say they like it. The ingredients list, when you look at the small print, has some of the same ingredients that come in DOS beer, even though the manufacturer claims that this is an entirely new brew. Windows NT Beer: Comes in 32-oz. cans, but you can only buy it by the truckload. This causes most people to have to go out and buy bigger refrigerators. The can looks just like Windows 3.1 Beer's, but the company promises to change the can to look just like Windows 95 Beer's - after Windows 95 beer starts shipping. Touted as an industrial strength beer, and suggested only for use in bars. Unix Beer: Comes in several different brands, in cans ranging from 8 oz. to 64 oz. Drinkers of Unix Beer display fierce brand loyalty, even though they claim that all the different brands taste almost identical. Sometimes the pop-tops break off when you try to open them, so you have to have your own can opener around for those occasions, in which case you either need a complete set of instructions, or a friend who has been drinking Unix Beer for several years. AmigaDOS Beer: The company has gone out of business, but their recipe has been picked up by some weird German company, so now this beer will be an import. This beer never really sold very well because the original manufacturer didn't understand marketing. Like Unix Beer, AmigaDOS Beer fans are an extremely loyal and loud group. It originally came in a 16-oz. can, but now comes in 32-oz. cans too. When this can was originally introduced, it appeared flashy and colorful, but the design hasn't changed much over the years, so it appears dated now. Critics of this beer claim that it is only meant for watching TV anyway. VMS Beer: Requires minimal user interaction, except for popping the top and sipping. However cans have been known on occasion to explode, or contain extremely un-beer-like contents. Best drunk in high pressure development environments. When you call the manufacturer for the list of ingredients, you're told that is proprietary and referred to an unknown listing in the manuals published by the FDA. Rumors are that this was once listed in the Physicians' Desk Reference as a tranquilizer, but no one can claim to have actually seen it. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED]
Re: Tip of the day (was :Re: Documentation)
On Wed, 15 Jan 1997, Jonas Bofjall wrote: On Wed, 15 Jan 1997, Paul Seelig wrote: What about a Tip of the day package in place of fortune I think it is a great idea. As long as it is an Optional package, I don't think anyone will complain. Shall we begin collecting Tips? Perhaps we can obtain the Tip of the Day database from Micro$oft Word. We all know there's tons of useful information in that databank. NOT! --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Sendmail redirects all *local* mail to domain MX ?????
On Mon, 13 Jan 1997, Perry Piplani wrote: I just installed debian in one of my boxes (from slakware) and noticed a problem. All local mail on the system and all mail from outside address to [EMAIL PROTECTED] is being re-addressed by sendmail to the main MX for may domain. I want mail specifically addressed to [EMAIL PROTECTED] to be delivered to thishost.mydom.com *not* mxhost.mydom.com. That's how it worked on slackware. A guess, and let me reiterate that it is just a guess, would be that sendmail on the local machine has one (or both) of these configuration errors: 1) It has been set up to forward all mail to a mailhub (Null client). 2) It doesn't recognize thishost as its host name. For #1, look for options in the sendmail.cf such as: DS (smart relay host) -or- DH (hub host) (to which all mail is sent) -or- DH (who gets all local email traffic). For #2, double check that /etc/hostname is the same as a reverse DNS lokup on the IP address of the machine. I'm a bit surprised that mx.mydom.com accepts your mail, as /etc/mail/sendmail.cw (in the debian implementation, was formerly in sendmail.cf as CW) must contain thishost.mydom.com. Unless of course that you are sending to [EMAIL PROTECTED] and wanting it to be delivered locally, in which case you need to add mydom.com to the /etc/mail/sendmail.cw. Let me know if any of this makes sense, let alone fixes the problem. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Sendmail redirects all *local* mail to domain MX ?????
On Tue, 14 Jan 1997, Richard G. Roberto wrote: On Mon, 13 Jan 1997, Pete Templin wrote: 2) It doesn't recognize thishost as its host name. For #2, double check that /etc/hostname is the same as a reverse DNS lokup on the IP address of the machine. Errr, what if somebody isn't running DNS??? (like me???) Won't defining Dwmyhost and Dj$w still work? What the heck is this .cw file for? Why is debian implementing an incompatable sendmail? Isn't sendmail complicated enough? I thought a project design goal was to have a *nix compatable system? Are other Linux distributions adopting this .cw file as well? Sorry. I guess I meant a reverse /etc/hosts or DNS lookup. The .cw file is implementing the Cw macro as a separate file, perhaps so that someone without root privs can maintain the mailserver. Cw is still in the .cf, so it is still _compatible_, at least with documented standards. From my mailhub templinux, I allow templin-wks: [snipped from sendmail.cf] Cwlocalhost # file containing names of hosts for which we receive email Fw/etc/mail/sendmail.cw [snipped from sendmail.cw] templin-wks.bucknell.edu I'm a bit surprised that mx.mydom.com accepts your mail, as /etc/mail/sendmail.cw (in the debian implementation, was formerly in sendmail.cf as CW) must contain thishost.mydom.com. Unless of course that I found no debian specific documentation for sendmail that mentioned a sendmail.cw file. Perhaps not, but sendmail has always (well, for a while, and documented in Eric Allman's bible) used at least the Cw macro (the sendmail.cw is at least in debian linux - I haven't even considered looking at another distribution since I saw the Debian light) to extend the definition of local addresses. Example: Mail.bucknell.edu handles local mail delivery for @bucknell.edu mail. As such, it needs to have bucknell.edu defined in Cw. If bucknell were to install a holding point to manage incoming mail (perhaps while a system upgrade took place, using DNS MX records), it would NOT handle bucknell.edu mail (as it doesn't have the users defined/mailboxes properly updated). Although I refuse to claim that I am a sendmail expert, I can probably field a few questions/provide a little advice on the above. Holler if I can help. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: how to find out who has fingered me?
On Mon, 13 Jan 1997, Lawrence Chim wrote: Rather than checking any log file, is it possible to find out who has fingered a user. I heard that I can create a .plan file to do it, but I don't know how to do it. Anyone knows how? Install cfinger (a debian package), a configurable finger server. It will create a .fingerlog in each user's home directory (if they are fingered), along with whatever central logging is normally performed by the program. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Dselect suggestions: mailing list created
Hello all. As much as I think that the many suggestions about dselect are beneficial, I'd like to see the debian-user list return to general usage hints, etc. As such, I've created a mailing list for dselect suggestions: [EMAIL PROTECTED] To subscribe or unsubscribe, send email to: [EMAIL PROTECTED] Please keep in mind that it is a manually-edited list until I can select and implement list processor software, so subscriptions may take a little time. The list will process mail to it at 30-minute intervals. If any one else wishes to provide this service (in a more standard fashion or at a more standard address), I would be happy to offer any assistance and transition aliases as long as necessary. Hopefully, people will use this opportunity to help minimize traffic on debian-user, as it's been rather busy lately (sometimes 2-4 hour delay to process mail, I believe). Thanks for your support. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: A proposal to improve dselect
On Sat, 11 Jan 1997, Jonas Bofjall wrote: No, this is wrong. A new user should not have to read long documents prior to installation. The configure scripts which runs directly after the installation should make reading docs unnecessary. I disagree. You should understand what you are doing. If you don't even want to know what is going and how you are to use it, what is the point of having it? Bragging to your friends? My totally-newbie friends were both given rex of my HD. They both called me after installation and asked how to get X started. Neither had configured X in any way. How are they supposed to know? The post-install configure script should take care of it. The installation process is not (necessarily) the place to learn about packages. Dselect in particular does not have room (unless you've got some incredible montior and a very long xterm) to give full installation, configuration, and operation instructions before or during installation. If it did, how would publishers like O'Reilly and Associates be in business? This is the real world. We as humans may have to read a bit and learn a bit to use a bit of our toys. Remember, it's JUST a computer. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Mail servers: share your experience with us
On Fri, 10 Jan 1997, Debian - Leander Berwers wrote: There are a number of mail server products like sendmail and smail. - What product are you using? Sendmail as made available from the debian developers. I've grown up as a sendmail junkie, and keep the O'Reilly book within arm's reach. I do need to buy the new one. - Are you satisfied with it? Does it do what you expect from it? Yup. Handles my unmoderated list very well. - How does it behave at higher loads? Performance losses? Brain deads? Minimal experience at high load, but then again, I used the sendmail bok to guide me when setting up my mailing list. My list would have to see a massive amount of email to cause the load to get crazy. It's set up so that mail to the list is delivered one at a time by a queue runner, not handled immediately. As such, I'd have to be delivering to all local mail boxes and have enough mail to last more than 30 minutes to push the load average above 1! So yes, I do like sendmail. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Disk partitioning - recommended sizes (fwd)
On Thu, 9 Jan 1997, Timothy Phan wrote: :Make the following symlinks: : :/tmp - /local/tmp (unless you might share this drive via NFS) :/home- /local/home :/usr/local -/local/usr :/var/spool -/local/spool (again, if using NFS, you should break this : down - talk to me individually) Do the above FS and symlinks comply with fsstnd (File system standard)? Just want to know. Thanks! Well, it falls under the category of site specific filesystems, and with the symlinks in place, everything that the fsstnd specifies will still exist where they belong (you can still do -- Thu 21:12 on Templinux : pwd is ~ tcsh ls -l /var/spool/mail/templin -rw-rw 1 templin mail 5483 Jan 9 21:06 /var/spool/mail/templin -- Thu 21:12 on Templinux : pwd is ~ tcsh without a problem, it's just that the file actually exists on a custom partition. My main idea in spec'ing the above was to cluster files that a user would want to retain from one install to another without fear of loss due to initialization of a system partition (notice I didn't say repartitioning!), and it also offers the ability to dynamically reallocate space as necessary: if you're running short on disk space, you can dump /usr/local/src to tape and scrap it, freeing up more space for home directories or monstrous inboxes. Strange way of looking at it, but I'd rather bunch /usr/local, /home, and /var/spool/mail in one disk (and guess on the right size) than have to guess three times. Given the WIDE variety of hardware out there and the fact that we can't go to the store and ask the Debian salesperson to assemble a system to our requirements (usually...), it'll be a guessing game already. Much as I enjoy reinstalling machines (sorry, in my line of work, you get numbed to machines that won't boot!), I'd rather not have to spin tape for all of my personal files just to repartition for them. Especially since I just found out that my backup script had lots of oopses in it!!! --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: using mirror with delete_excl
On Wed, 8 Jan 1997, Dirk Luetjens wrote: I tried to mirror the debian archive with the mirror command. I like to keep the local tree although in the debian hirarchie, so I included the line delete_excl=(/mnt/debian/local) to the config file. But the files in the local tree are still delete during the mirror process. the config file: package=debian site=ftp.inka.de remote_dir=/debian local_dir=/mnt/debian exclude_patt=((i-connect-fixes|ms-dos|msdos-i386|msdos-m68k|binary-alpha|binary-m68k|binary-sparc|source|project|local)($|/)) delete_excl=(/mnt/debian/local) # local_ignore=(/mnt/debian/local) do_deletes=true Here's my mirror config file, which successfully ignores all of the buzz (Debian 1.1) directories I have laying around: package=debian site=ftp.debian.org remote_dir=/debian/ local_dir=/server/ftp/pub/debian/ mail_to=templin delete_excl=buzz* Since I've now publicly mentioned that I carry buzz, I'll mention that my hostname is templinux.bucknell.edu. Since I'm hoping to work on some network changes, I'm not ready to announce it as a public mirror (sorry Bruce!), but it is usually available 24/7. For those who do explore my site, /home/ftp is a symlink to /server/ftp, so anonymous users will come in at /server/ftp. Hope this helps! --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Disk partitioning - recommended sizes
On Thu, 9 Jan 1997, Matt Kracht wrote: You might want something like the following: 50MB / 100MB /var 250MB /var/spool 250MB /tmp 500MB /usr 750MB /usr/local 100MB swap My $0.02: you're not going to win any _performance_ wars by making more partitions, particularly if you don't have many separate drives to put them on. A few things should factor into your decisions: / should be small, for best reliability. (YMMV, mine is /dev/hda1 19485 10257 8222 56% /) You'll only get this small if /usr, /var, and /home are anywhere else (even if they're all in one partition). Your apps (/usr) may want to write to their logfiles (/var) while they read your files (/home). If they're in different partitions (particularly if you also split /usr/local and /var/spool), your poor disk arm is going to earn its keep. My suggestion? 20M / 500M/usr 100M/var (the rest) /local (or whatever) Make the following symlinks: /tmp- /local/tmp (unless you might share this drive via NFS) /home - /local/home /usr/local - /local/usr /var/spool - /local/spool (again, if using NFS, you should break this down - talk to me individually) Here's my server: % df Filesystem 1024-blocks Used Available Capacity Mounted on /dev/hda1 19485 10257 8222 56% / /dev/hda2 223494 14690565048 69% /usr /dev/hdc3 198123 12324 175568 7% /var /dev/hdc4 2883543809 269652 1% /tmp /dev/hda3 5600608107 523024 2% /nfs /dev/hdb12990073 1559445 1276003 55% /server % I'm still learning myself. I rarely see /tmp in use at all. /nfs for me is /local as listed above, and /server is my 3.1G for my debian mirror and a few other services that I provide. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
nfsiods?
Hi all, I'm getting lambasted by our network administration for killing a subnet with NFS traffic. Before he and I discuss the future of my machines, I'm trying to clean up whatever mess I can, so my machines can hide away on his network management station. Which brings me to the question: What script starts the four nfsiod's that are on my stock, Debian 1.2 system (not upgraded from 1.1, but not necessarily 1.2.1)? How can I prevent them from even starting at boot time? A grep nfsiod * in /etc/init.d provides nothing. Ideas? Thanks, --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: A proposal to improve dselect
On Tue, 7 Jan 1997, Martin Konold wrote: Yes, a very good point. I am offering a host for a mailing list. We should first figure out how it should work and implement it afterwards. There is definetelly a need for a improved dselect. Actually why is the maintainer so silent? Perhaps you would be silent if discussions about your package were turning into some semi-serious bash N trash sessions. I'd like to offer my two cents about Debian and dselect: Most of us are brand new to Linux or are advancing up the UNIX ladder when we install Debian on a machine. Personal computers offer an ability to experiment that the departmental or enterprise server won't give us. With that experimentation comes a few oopses and a few lessons learned. With a true multitasking, multiuser system comes certain hurdles about the boot process and services (daemons). Keep in mind that we are all getting a generally fantastic product for the best price anyone could ask for. I've never been involved in the development of any of the DEC boxes which handle our campus net services, but I believe the standard sequence goes like this: get and compile gcc with the cc that came with the machine. get and compile emacs with gcc. get and compile tcsh, now that you can edit Makefiles with emacs. get and compile perl, now that you've got a shell you're familiar with. get and compile sendmail, so email can actually flow. Heaven forbid one of us gets a compilation error, and wait until it's time to build inn! Take your time with Linux. I openly admit that I had overly high expectations the day my first Pentium arrived. Now that I've finally acquired my second Pentium (http://www.bucknell.edu/~templin/pages/computer if you're curious), I let one run Linux 24/7, and try new packages on the other. Mistakes will happen. Dselect might lead you astray. But accept what the Debian project has given each of us, and send a few thanks to each and every person who has contributed their own time to simplify your life, to make it possible for you to experience UNIX with a minimum of effort on a variety of hardware. The project leader has managed to get a few emails onto the list while cleaning out from a devastating flood. That's what I call dedication. How about we all take a step or two back and peek at what is in front of us? There's a lot there. It may not be the best it can be yet, but it's quite fine in its current form, and a menu-driven is certainly a step up from the command-line origins of UNIX. That said, who is willing to coordinate efforts toward gathering suggestions for dselect, and what is the next step that we need to take? I also have a machine which I am willing to offer up towards mailing lists, disk space, web pages, or whatever. Let me know how I might help. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: OK to install across 2 HDs?
On Sun, 5 Jan 1997, Mark Blunier wrote: I would use hda and hdb. I don't have an eide controller so I don't have an hdc or hdd. Some might prefer to use hda and hdc, since both drives could be masters, but if you have an ide cdrom, I would use hda, and hdb, as I believe the cdrom can slow down the other side, ie cdrom on hdd slows down hdc. I've got two WD1.2s in my workstation (Win95/WinNT/DebianLinux). I had the two drives on primary master/slave, with CD as secondary slave. I noticed INCREDIBLE slowness in 95's ScanDisk when doing a thorough scan, so I fired up System Monitor and watched filesystem reads, bytes/sec. My primary master was showing 1.5MBytes/sec read performance, while my primary slave was showing 150KBytes/sec read performance. I (carefully) slid the second drive back to secondary master and moved the CD-ROM to secondary slave, and now both drives show nearly identical (i.e. 1.5MB) performance in ScanDisk. While we're on the subject of identical hard drives, can anyone tell me how to get both of my WD1.2s to show up with the same CHS? I can't seem to squeeze in a bootable (C1024) partition after the 800M in the beginning. Thanks, --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
Re: Any PPP-connection keeping utility ?
On Sun, 5 Jan 1997, Zenon Fortuna wrote: I would like to keep my PPP connection busy even during the time, when I don't use it. Otherwise my ISP will log-me-off after a couple of idle minutes. I had a script back when I ran Slackware which tidied up the ppp launching process. It would redial if I got dropped, email me at a remote site with my IP address, and launch the following (once): % ping -q 300 my.isp.name.server It's one ping every five minutes. Make sure you choose a site close to your dial-in point, as either my ISP or my university's ISP would lose their connection a lot, and I'd get dropped if I was only relying on telnets to remote sites. --Pete ___ Peter J. Templin, Jr. Client Services Analyst Computer Communication Services tel: (717) 524-1590 Bucknell University [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]
