Re: Unidentified subject!
Hi, On Mon, Jun 18, 2001 at 06:43:36PM +0200, Alexander Skwar wrote: So sprach [EMAIL PROTECTED] am Mon, Jun 18, 2001 at 06:29:47PM +0200: Well the price of a Solaris installation is _exactly_ the same a the price for a Linux installation! Ah, okay. This also includes the numerous Solaris boards and huge pile of software available for Solaris? Where can I find, let's say, KDE II or Evolution 1.0.3 for Solaris? Does it compile just as easily there as it does on Linux? FYI, from the SUN website: - Sun is a charter member of the GNOME Foundation and an active member of the GNOME community. Sun is committed to delivering a high-quality GNOME user environment for the Solaris Operating Environment, and to providing dedicated resources for the ongoing development and enhancement of GNOME. Availability Sun will offer GNOME 2.0 for the Solaris Operating Environment. Development of the 2.0 release is still in the early stages. The following is a preliminary release schedule: * Q1 CY 2001: Public early access * Q2 CY 2001: Beta early access * Q3 CY 2001: General availability - AFAIK you can already download Gnome 1.4 at the moment. So there may may be no KDE, but there's Gnome and it will be the standard Solaris dektop in the future. For further information, look at : http://www.sun.com/software/gnome/siteindex.html Also you can download Solaris 8 free of charge and at least parts of it are open source now. Cheers, Dirk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Urheberrecht in English
Hi all, I just grabbed in a dictionary and have an idea Urheberschaft is Authorship, so how about using Authorship-right. Would this be understood the right way? Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
problems with cdrecord 1.9 - RedHat 6.2 - HP 9100
Hi burners. I have Red Hat 6.2 on a HP server and I had attached an IDE HP 9100 CD-Writer. Actually I have read lots of documentation, FAQ's, lists, and I cannot burn anything. This is what I've already know: - cdrecord only works with SCSI commands. - for IDE drives I need to load ATAPI, or IDE-SCSI emulation. - RedHat 6.2 supports this module, so I have put append="hdb=ide-scsi" on my lilo.conf and run lilo - This is my modules.conf file: alias scsi_hostadapter megaraid alias scsi_hostadapter1 ncr53c8xx alias eth0 eepro100 alias eth1 eepro100 alias parport_lowlevel parport_pc alias scsi_hostadapter2 ncr53c8xx alias scsi_hostadapter3 ncr53c8xx options ide-cd ignore=hdc pre-install sg modprobe ide-scsi - I have restarted my server. - I have run cdrecord -scanbus and this is the result: Cdrecord 1.9 (i686-pc-linux-gnu) Copyright (C) 1995-2000 Jörg SchillingLinux sg driver version: 2.1.36cdrecord: Warning Linux Bus mapping botch.cdrecord: Warning Linux Bus mapping botch.Using libscg version 'schily-0.1'scsibus0: 0,0,0 0) '' '@F@h8@Ð@0' '' unknown device type 0x12 0,1,0 1) * 0,2,0 2) * 0,3,0 3) * 0,4,0 4) * 0,5,0 5) 'HP ' 'SAFTE; U160/M BP' '1021' Processor 0,6,0 6) * 0,7,0 7) * - Then I have run cdrecord -v -debug dev=0,5,0 -data /logs/trading/cdr1.img and this is the output from the system: File: '/logs/trading/cdr1.img' tracksize: 639850496 secsize: 2048 tracktype: 1 = CD-ROM sectype: 1 = CD-ROM mode 1 dbtype: CD-ROM mode 1 flags 1080dev: 0,5,0 speed: -1 fs: -1Cdrecord 1.9 (i686-pc-linux-gnu) Copyright (C) 1995-2000 Jörg SchillingTOC Type: 1 = CD-ROMfs: 4194304 buflen: 4198400cdrecord: shared memory segment allocated: 769cdrecord: shared memory segment attached: 4010e000buf: 4010E000 bufend: 4050F000, buflen: 4198400buf: 4010E000 bufend: 4050F000, buflen: 4198400 (align 0)scsidev: '0,5,0'scsibus: 0 target: 5 lun: 0Linux sg driver version: 2.1.36l1: 0x2405 l2: 0x311Bus: 0 Target: 5 Lun: 0 Chan: 0 Ino: 36Using libscg version 'schily-0.1'Target (0,5,0): DMA max 129024 old max: 64512SCSI buffer size: 64512Target (0,5,0): DMA max 129024 old max: 64512scsi_getbuf: 64512 bytesatapi: 0DMA addr: 0x size: 0 - using copy bufferDMA addr: 0x size: 0 - using copy bufferDMA addr: 0x08071E38 size: 36 - using copy buffercdrecord: Cannot do inquiry for CD/DVD-Recorder.DMA addr: 0x size: 0 - using copy buffercdrecord: The unit seems to be hung and needs power cycling. And that's all folks. Maybe I have forget something, so any helpwould be appreciated. Regards.
Re: cdrecord weakness
Matthias Schniedermeyer [EMAIL PROTECTED] wrote: cdrecord has a problem with SCSI-(RAID)-Controllers that don't allocate a ne scsibus for each Channels. Yesterday i got a 5 Channel FAST-SCSI RAID-Controller (ICP-Vortex) I have 4 CD-Rs and every one has a dedicated channel on the new Controller. First i had configured every CD-R to ID-0. But then cdrecord could only see the first one. Now i configured every drive with diffrent IDs (cdrecord -scanbus attached). But that's not the(tm) solution. The answer is that some RAID controllers present the PCI interface with each RAID group shown as a single drive (with a virtual device number and LUN 0) regardless of how many drives and/or SCSI busses are used by the RAID group. If this is the case the o/s can not tell the actual physical connect layout, and probably shouldn't care. AFAIK other OSses use Adapter,Channel,ID,Lun Adresses too. Why does cdrecord only use scsibus,ID,Lun (Hmmm. Do ANY cases exist where the LUNs are needed? (i only know 1 case, SCSI2SCSI-Bridges which map the ID from the slave-Channels to LUNs of the ID on the Main-Channel. But i think those bridges are unusual) Sure, if you buy a CD tower config, all the drives have the same device number and use LUNs. This allows you to connect WAY TOO MANY to a single controller. Now my RAID controllers, which IBM ships with Netfinity systems, clearly states that it should not be used with CD-ROM devices, so I would take their word for it. I assume your controller has CD capability, but I'm not surprised that the information returned doesn't match the actual setup, and as long as the controller has a no RAID setting it doesn't matter if the CDs look as if they are on a single bus. I believe that there is no problem in cdrecord or Linux in this case, it just the way it works. Also note that if you use devfs under the current (2.4) kernel series, the detail is much more obvious, although you still only see what the controller chooses to show you. Undoubtedly someone will claim that this is a Linux bug and that some other o/s calls the phychic hotline and gets the actual config ;-) -- cdrecord -scanbus -- scsibus0: 0,0,0 0) * 0,1,0 1) * 0,2,0 2) * 0,3,0 3) * 0,4,0 4) * 0,5,0 5) * 0,6,0 6) 'IBM ' 'DDYS-T18350N' 'S93E' Disk 0,7,0 7) * scsibus1: 1,0,0 100) * 1,1,0 101) * 1,2,0 102) * 1,3,0 103) * cdrecord: Warning: controller returns wrong size for CD capabilities page. 1,4,0 104) 'PIONEER ' 'DVD-ROM DVD-303 ' '1.10' Removable CD-ROM 1,5,0 105) * 1,6,0 106) * 1,7,0 107) * scsibus3: 3,0,0 300) 'TEAC' 'CD-R58S ' '1.0H' Removable CD-ROM 3,1,0 301) 'TEAC' 'CD-R58S ' '1.0N' Removable CD-ROM 3,2,0 302) 'PIONEER ' 'DVD-ROM DVD-304 ' '1.03' Removable CD-ROM 3,3,0 303) 'PIONEER ' 'DVD-ROM DVD-304 ' '1.03' Removable CD-ROM 3,4,0 304) 'TEAC' 'CD-R58S ' '1.0K' Removable CD-ROM 3,5,0 305) * 3,6,0 306) 'TEAC' 'CD-R58S ' '1.0P' Removable CD-ROM 3,7,0 307) * -- end -- -- /proc/scsi/scsi -- Attached devices: Host: scsi0 Channel: 00 Id: 06 Lun: 00 Vendor: IBM Model: DDYS-T18350N Rev: S93E Type: Direct-AccessANSI SCSI revision: 03 Host: scsi1 Channel: 00 Id: 04 Lun: 00 Vendor: PIONEER Model: DVD-ROM DVD-303 Rev: 1.10 Type: CD-ROM ANSI SCSI revision: 02 Host: scsi3 Channel: 01 Id: 02 Lun: 00 Vendor: PIONEER Model: DVD-ROM DVD-304 Rev: 1.03 Type: CD-ROM ANSI SCSI revision: 02 Host: scsi3 Channel: 01 Id: 03 Lun: 00 Vendor: PIONEER Model: DVD-ROM DVD-304 Rev: 1.03 Type: CD-ROM ANSI SCSI revision: 02 Host: scsi3 Channel: 02 Id: 00 Lun: 00 Vendor: TEAC Model: CD-R58S Rev: 1.0H Type: CD-ROM ANSI SCSI revision: 02 Host: scsi3 Channel: 03 Id: 01 Lun: 00 Vendor: TEAC Model: CD-R58S Rev: 1.0N Type: CD-ROM ANSI SCSI revision: 02 Host: scsi3 Channel: 04 Id: 06 Lun: 00 Vendor: TEAC Model: CD-R58S Rev: 1.0P Type: CD-ROM ANSI SCSI revision: 02 Host: scsi3 Channel: 05 Id: 04 Lun: 00 Vendor: TEAC Model: CD-R58S Rev: 1.0K Type: CD-ROM ANSI SCSI revision: 02 -- end -- -- -bill davidsen ([EMAIL PROTECTED]) The secret to procrastination is to put things off until the last possible moment - but no longer -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: cdrecord weakness Part II
Matthias Schniedermeyer [EMAIL PROTECTED] wrote: Now i tried to record something with my new RAID-Controller but i had no luck at all. # cdrecord -eject -dao speed=8 dev=3,0,0 fs=30m -v IMAGE Cdrecord 1.10 (i686-pc-linux-gnu) Copyright (C) 1995-2001 Jörg Schilling TOC Type: 1 = CD-ROM scsidev: '3,0,0' scsibus: 3 target: 0 lun: 0 Linux sg driver version: 3.1.17 cdrecord: Warning Linux Bus mapping botch. Using libscg version 'schily-0.5' atapi: 0 Device type: Removable CD-ROM Version: 2 Response Format: 2 Capabilities : RELADR SYNC LINKED Vendor_info: 'TEAC' Identifikation : 'CD-R58S ' Revision : '1.0H' Device seems to be: Generic CD-ROM. cdrecord: Sorry, no CD/DVD-Recorder or unsupported CD/DVD-Recorder found on this target. Seems pretty clear, the device identifies itself as a CD-ROM, read only. I assume that the device is a recorder? -- -bill davidsen ([EMAIL PROTECTED]) The secret to procrastination is to put things off until the last possible moment - but no longer -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mkisofs Joliet fs 64 bytes unique names restriction
Suspect this is due to the problem with generating unique Joliet file names in case the first 64 bytes in the names of original files are equal. Is there any way out? Either don't use the Joliet extension, or rename the files to less than 64 characters before running mkisofs. I don't believe that's needed. The names should be *unique* in 64 characters, but I think they can safely be longer (255?). The Joliet Specification states: The File or Directory Identifiers may be up to 128 bytes (64 unicode characters) in length. I have never tested this limit, and I'm not sure if this should include any version numbers. I have no idea why Microsoft have this limit - I _think_ the ISO9660 directory structures will allow a filename to be up to about 200 characters (100 Unicode characters). I guess the only why to test this is to hack joliet.c and allow filenames to be longer and see if the resulting CDs can be read on MS systems ... James Pearson -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Write error and scsi bus reset
I've downloaded the cdrecord1.11a04, the compilation in Linux system is fine, but on the FreeBSD 4.2 I got the following message when I type make: -- ./RULES/rules1.top, line 79: Could not find ./RULES/mk-.XxZzy-.XxZzy-make.id ./RULES/rules1.top, line 118: Could not find .XxZzy-/.XxZzy-/Defaults ./RULES/rules.top, line 39: Could not find ./RULES/-freebsd-.XxZzy-.rul make: fatal errors encountered -- cannot continue -- I've set some variables but I see that isn't the right way to solve that. The default FreeBSD system needs any special variables to compil cdrecord? I had not found anything related with this special behaviou in README.FreeBSD file. [EMAIL PROTECTED] wrote: From: Gleydson Mazioli da Silva [EMAIL PROTECTED] Sometimes the cdrecord hang while/after writing the first track (in DAO mode) and my SCSI bus is reseted: Had anyone some clue to solve that? [ cdrecord output ]-- Track1 Start: '33' (33) pregap1: 183 Cdrecord 1.9 (i386-unknown-freebsd4.2) Copyright (C) 1995-2000 Jörg Schilling Please don't use outdated versions when you have problems. http://www.fokus.gmd.de/research/cc/glone/employees/joerg.schilling/private/problems.html Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- --- Gleydson Mazioli da Silva [EMAIL PROTECTED] [EMAIL PROTECTED] Biquini: Pedaço de pano cercado por mulher por todos os lados. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Write error and scsi bus reset
From [EMAIL PROTECTED] Tue Jun 19 19:13:03 2001 I've downloaded the cdrecord1.11a04, the compilation in Linux system is fine, but on the FreeBSD 4.2 I got the following message when I type make: -- ./RULES/rules1.top, line 79: Could not find ./RULES/mk-.XxZzy-.XxZzy-make.id ./RULES/rules1.top, line 118: Could not find .XxZzy-/.XxZzy-/Defaults ./RULES/rules.top, line 39: Could not find ./RULES/-freebsd-.XxZzy-.rul make: fatal errors encountered -- cannot continue -- You are using a broken make program Read README.compile. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cdrtools-1.11a04 ready
[EMAIL PROTECTED] notified us: I hope that if I give you some candy, I will get some test results... NEW features of cdrtools-1.11a04: I won't quote these, many, many! Okay, with the various new raw modes work (or fail) in -dummy mode, so I can run this against my various Sony, Plextor and Philips drives? I would rather be able to check that it looks as if it works before I do a live test. I want to try pulling the toc and audio stuff again, raw mode on Philips, etc, but a bit of caution is desirable. -- -bill davidsen ([EMAIL PROTECTED]) The secret to procrastination is to put things off until the last possible moment - but no longer -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cdrtools-1.11a04 ready
From: [EMAIL PROTECTED] (Bill Davidsen) Okay, with the various new raw modes work (or fail) in -dummy mode, so I can run this against my various Sony, Plextor and Philips drives? I would rather be able to check that it looks as if it works before I do a live test. I want to try pulling the toc and audio stuff again, raw mode on Philips, etc, but a bit of caution is desirable. Well, if you like -dummy in RAW mode to work, ask the SCSI standard kommittee to change the standard for your needs Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
[EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] (Bill Davidsen) I'm sure there's some limit to characters, but I haven't hit it yet. 64k is the limiting factor when you specify lines - the default. as blocks are easier to type I use blocks when I whant to go back further and if I do this on Linux, it does not work! I see this: newssvr15:root tail -7 /var/log/debug | wc 7 1255826 12390095 newssvr15:root uname -a; uptime Linux newssvr15.news.prodigy.com 2.4.5-ac5 #4 SMP Wed Jun 6 12:37:23 EDT 2001 i686 unknown 1:28pm up 11 days, 22:18, 5 users, load average: 0.99, 1.10, 1.26 newssvr15:root So you are not running UNIX tail No I'm running Linux tail, as it comes with Slackware, Redhat, mandrake, SuSE, etc. Did you deliberately install UNIX tail on your Linux system? And if so why doesn't the -NNNb option work to your satisfaction? I can't seem to find this behaviour on any Slackware, SuSE or Redhat system even on systems last modified in 1968. This looks as if it works There was not UNIX in 1968. yes, typo, that's 1998. But you knew that, since it's in the build date of my original post: Linux newsutil1 2.1.131 #4 SMP Thu Dec 10 10:59:04 EST 1998 i686 unknown fine. You're not running obsolete software are you? Or your own custom version of 'tail'? You are using a own custom version of tail, I use the UNIX tail command. No, I'm running whatever tail comes with all the distributions I use from choice or necessity. That's why I tried several machines of various flavors, to see if the tail I used was something recent. I really think you have some damaged version, the -b is flagged as tail: b: invalid suffix character in obsolescent option (means becoming obsolete, I think it is). Both Solaris and AIX also accept byte counts larger than 64k, so there's no obvious reason to use b other than habit. No, I don't know or claim POSIX says it's obsolete, but I haven't seen anyone actually use it in at least the last eight years, any I'm sure I haven't used it in longer than that. Anyway, I think there's no problem with 64k in standard Linux distributions. I don't know what you're running, but it's not a Linux problem in popular distributions. -- -bill davidsen ([EMAIL PROTECTED]) The secret to procrastination is to put things off until the last possible moment - but no longer -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
So sprach [EMAIL PROTECTED] am Tue, Jun 19, 2001 at 07:54:02PM +0200: There was not UNIX in 1968. Ah, so you never did a typo? You are using a own custom version of tail, I use the UNIX tail command. Fine, so what? Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.digitalprojects.com | http://www.iso-top.de iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 4 hours 2 minutes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cdrtools-1.11a04 ready
On 19 Jun, [EMAIL PROTECTED] wrote: Well, if you like -dummy in RAW mode to work, ask the SCSI standard kommittee to change the standard for your needs= Indeed the MMC standard says that the Test Write bit is only valid for TAO and DAO, but not for Raw writing. On the other hand, I have ignored this sentence and in practice test writes work very well even with raw writing. If you really want to be safe then a warning that the drive might ignore the -dummy option might be appropriate, but I know no drive that really ignores it. Do you? Bye, Patrick Ohly -- [EMAIL PROTECTED] [EMAIL PROTECTED] (MakeCD related mails) http://home.pages.de/~Ohly/ http://makecd.core.de/ (MakeCD home page) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Cdrtools-1.11a04 ready
From: [EMAIL PROTECTED] (Bill Davidsen) From: [EMAIL PROTECTED] (Bill Davidsen) Okay, with the various new raw modes work (or fail) in -dummy mode, so I can run this against my various Sony, Plextor and Philips drives? I would rather be able to check that it looks as if it works before I do a live test. I want to try pulling the toc and audio stuff again, raw mode on Philips, etc, but a bit of caution is desirable. Well, if you like -dummy in RAW mode to work, ask the SCSI standard kommittee to change the standard for your needs Still trying to be popular, aren't you? I ask a polite yes/no question, and you provide a long hostile answer... I'm ready to try the test results you asked people to provide and you take the opportunity to say something nasty. The reason you are not getting the feedback you want might be because people don't like to be insulted, belittled, or otherwise treated less than politely. If you are sending an aggressive answer to my request on testing you should epect an aggressive reply I simply did reply in the same tone as your mail. If you like a polite answer simply send your question in the manner you expect. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
So sprach Joerg Schilling am Wed, Jun 20, 2001 at 01:04:46AM +0200: What should this be? Okay, thanks, this just shows that all your objections can be disregarded. dialog is the toll used in the Linux kernel config when you run make menuconfig (although lxdialog is a specialized version). dialog is used to easily create, well, dialog windows from within shell scripts. Further information is at http://www.advancedresearch.org/dialog/ It is not part of UNIX. That's not what I asked. I asked if it's pre-installed, not if it's part of some standard., Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.digitalprojects.com | http://www.iso-top.de iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 13 hours 24 minutes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
From [EMAIL PROTECTED] Wed Jun 20 06:46:10 2001 So sprach Joerg Schilling am Wed, Jun 20, 2001 at 01:04:46AM +0200: What should this be? [out of scope message deleted] dialog is the toll used in the Linux kernel config when you run make menuconfig (although lxdialog is a specialized version). dialog is used to easily create, well, dialog windows from within shell scripts. Further information is at http://www.advancedresearch.org/dialog/ It is not part of UNIX. That's not what I asked. I asked if it's pre-installed, not if it's part of some standard., so it is an internal part of the Linux kernel (even a hidden part) and nokody needs to konw about it unless he is going to change the Linux config menues. I see no relation to out discussion here -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Unidentified subject!
From: [EMAIL PROTECTED] (Bill Davidsen) [EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] (Bill Davidsen) Mike A. Harris [EMAIL PROTECTED] wrote: /usr/bin/which is an ELF executable on Red Hat Linux, but I agree that it definitely is not an application that can be expected to be portable. One of the reasons bash added 'type' as a builtin. Type is a shell builtin at least since 1982. I didn't know bash had been around that long, but why is the date relevant to the choice to include the feature when designing bash? This is funny you write a statement that is close to 'bash invented type' and dont understand when I tell you that type is a stone age command. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: AW: Unidentified subject!
From: Hermes, Stefan [EMAIL PROTECTED] I don't think bash has been around at 1982, what J=F6rg refers to is /bin/sh on at least Solaris 1 (SunOS), Sinix and ATT Unix (being the only Systems that old I have access to) in 1982 there only was sh and csh. In 1984 we got ksh and my bsh which where the first shells with cursur editable interactive history. IIRC bash started in 1992. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
X-Envelope-Sender: [EMAIL PROTECTED]
Mike A. Harris wrote:
/usr/bin/which is an ELF executable on Red Hat Linux, but I agree
that it definitely is not an application that can be expected to
be portable.
Which in Debian GNU/Linux it's a simple shell script that contains:
#!/bin/bash
unalias -a
unset -- $@ /dev/null
enable -n -- $@ /dev/null
type -p $@
---
So it will not do what people who know where which comes from expect.
The real which is (same file is on *BSD):
/*--*/
#! /usr/bin/csh -f
#
# Copyright(c) 1997, by Sun Microsystems, Inc.
# All rights reserved.
#
#ident @(#)which.csh 1.4 97/04/23 SMI
#
# which : tells you which program you get
#
# Set prompt so .cshrc will think we're interactive and set aliases.
# Save and restore path to prevent .cshrc from messing it up.
set _which_saved_path_ = ( $path )
set prompt =
if ( -r ~/.cshrc -f ~/.cshrc ) source ~/.cshrc
set path = ( $_which_saved_path_ )
unset prompt _which_saved_path_
set noglob
foreach arg ( $argv )
set alius = `alias $arg`
switch ( $#alius )
case 0 :
breaksw
case 1 :
set arg = $alius[1]
breaksw
default :
echo ${arg}:aliased to $alius
continue
endsw
unset found
if ( $arg:h != $arg:t ) then# head != tail, don't search
if ( -e $arg ) then # just do simple lookup
echo $arg
else
echo $arg not found
endif
continue
else
foreach i ( $path )
if ( -x $i/$arg ! -d $i/$arg ) then
echo $i/$arg
set found
break
endif
end
endif
if ( ! $?found ) then
echo no $arg in $path
endif
end
/*--*/Jörg
EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin
[EMAIL PROTECTED] (uni) If you don't have iso-8859-1
[EMAIL PROTECTED] (work) chars I am Jorg Schilling
URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
s/Which/which/ Gleydson Mazioli da Silva wrote: Mike A. Harris wrote: /usr/bin/which is an ELF executable on Red Hat Linux, but I agree that it definitely is not an application that can be expected to be portable. Which in Debian GNU/Linux it's a simple shell script that contains: #!/bin/bash unalias -a unset -- $@ /dev/null enable -n -- $@ /dev/null type -p $@ --- --- Gleydson Mazioli da Silva [EMAIL PROTECTED] [EMAIL PROTECTED] Não há quem não cometa erros e grandes homens cometem grandes erros. --Paulo Francis -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- --- Gleydson Mazioli da Silva [EMAIL PROTECTED] [EMAIL PROTECTED] Não há quem não cometa erros e grandes homens cometem grandes erros. --Paulo Francis -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Unidentified subject!
[EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] (Bill Davidsen) [EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] (Bill Davidsen) Mike A. Harris [EMAIL PROTECTED] wrote: /usr/bin/which is an ELF executable on Red Hat Linux, but I agree that it definitely is not an application that can be expected to be portable. One of the reasons bash added 'type' as a builtin. Type is a shell builtin at least since 1982. I didn't know bash had been around that long, but why is the date relevant to the choice to include the feature when designing bash? This is funny you write a statement that is close to 'bash invented type' and dont understand when I tell you that type is a stone age command. I still don't see why the date matters. When the design was done, some existing features like 'type' were included, some were left out, and some were slightly changed or enhanced. I don't see anywhere that I made any statement at all about invention, only that the feature was chosen for inclusion. We were talking about shells and 'which,' not sure what you're talking about. For some reason you seem to think that the date is very important, but I think it's also wrong, bash isn't that old. -- -bill davidsen ([EMAIL PROTECTED]) The secret to procrastination is to put things off until the last possible moment - but no longer -me [EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] (Bill Davidsen) [EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] (Bill Davidsen) Mike A. Harris [EMAIL PROTECTED] wrote: /usr/bin/which is an ELF executable on Red Hat Linux, but I agree that it definitely is not an application that can be expected to be portable. One of the reasons bash added 'type' as a builtin. Type is a shell builtin at least since 1982. I didn't know bash had been around that long, but why is the date relevant to the choice to include the feature when designing bash? This is funny you write a statement that is close to 'bash invented type' and dont understand when I tell you that type is a stone age command. I still don't see why the date matters. When the design was done, some existing features like 'type' were included, some were left out, and some were slightly changed or enhanced. I don't see anywhere that I made any statement at all about invention, only that the feature was chosen for inclusion. We were talking about shells and 'which,' not sure what you're talking about. For some reason you seem to think that the date is very important, but I think it's also wrong, bash isn't that old. -- -bill davidsen ([EMAIL PROTECTED]) The secret to procrastination is to put things off until the last possible moment - but no longer -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
[EMAIL PROTECTED] wrote: From: [EMAIL PROTECTED] (Bill Davidsen) [EMAIL PROTECTED] wrote: All implementations I know use 512 byte. You need it when you like to go back more then 64kB. I didn't realize the standard required that, on Linux or AIX you just use the c option to get what you would like: deathstar:davidsen zcat /tmp/lp2/LocalPosts.199912.gz | tail -15c | wc -c 15 I'm sure there's some limit to characters, but I haven't hit it yet. 64k is the limiting factor when you specify lines - the default. as blocks are easier to type I use blocks when I whant to go back further and if I do this on Linux, it does not work! I see this: newssvr15:root tail -7 /var/log/debug | wc 7 1255826 12390095 newssvr15:root uname -a; uptime Linux newssvr15.news.prodigy.com 2.4.5-ac5 #4 SMP Wed Jun 6 12:37:23 EDT 2001 i686 unknown 1:28pm up 11 days, 22:18, 5 users, load average: 0.99, 1.10, 1.26 newssvr15:root And on the oldest systems I could find easily: gatekeeper:davidsen tail -7 mail/inn-workers | wc 7 388964 3279144 gatekeeper:davidsen uptime; uname -a 1:37pm up 73 days, 16:12, 7 users, load average: 0.25, 0.07, 0.02 Linux gatekeeper 2.1.106 #11 Sun Dec 6 14:08:00 EST 1998 i686 unknown newsutil1:davidsen tail -75000 /home/davidsen/nsmail/Trash | wc 75000 112995 4554404 newsutil1:davidsen uname -a; uptime Linux newsutil1 2.1.131 #4 SMP Thu Dec 10 10:59:04 EST 1998 i686 unknown 1:41pm up 124 days, 21:46, 5 users, load average: 0.13, 0.03, 0.01 I can't seem to find this behaviour on any Slackware, SuSE or Redhat system even on systems last modified in 1968. This looks as if it works fine. You're not running obsolete software are you? Or your own custom version of 'tail'? -- -bill davidsen ([EMAIL PROTECTED]) The secret to procrastination is to put things off until the last possible moment - but no longer -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
So sprach Walter Hofmann am Tue, Jun 19, 2001 at 11:13:56PM +0200: My mktemp binary actually uses mkstemp(3) instead of mktemp(3). mkstemp is supposed to be a secure version of mktemp. Uhm, I don't know for sure, but I would suppose so as well. But simply according to the names I said the most safe answer without looking at the source - but you're most certainly right. It does. Your method is insecure, that's the difference. But it's not standard. Uhuh... Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.digitalprojects.com | http://www.iso-top.de iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 6 hours 15 minutes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
So sprach Walter Hofmann am Tue, Jun 19, 2001 at 11:42:15PM +0200: I know what I'm going to choose. Yep, so do I. But still Jörg's point is somewhat valid. If there's no mktemp on the system, a portable way has to be chosen. Even if the portable way is FAR less capable. It all boils down to how portable the thing is supposed to be. BTW: Is (c)dialog preinstalled on (most) Unices? If not, then it needs to be installed to make the CDcontrol scripts run anyway. And while installing one package, some other packages might be installed right away as well. Like less, mktemp (? don't know - is there something like this for legacy systems?), bash Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.digitalprojects.com | http://www.iso-top.de iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 6 hours 30 minutes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
So sprach Mike A. Harris am Mon, Jun 18, 2001 at 08:02:38AM -0400: define it in PAGER so there should be no problem. All releases of Red Hat Linux I have at my disposal seem to define PAGER properly anyways. Okay, I just checked a default install of RedHat 7.1 where less is also avaiable, and there $PAGER isn't set. I really don't know about your RedHats, but it is unset at 6.2 and 7.1. Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.digitalprojects.com | http://www.iso-top.de iso-top.de - Die günstige Art an Linux Distributionen zu kommen Uptime: 6 hours 48 minutes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
From [EMAIL PROTECTED] Tue Jun 19 23:25:06 2001 [EMAIL PROTECTED] schrieb am Sonntag, den 17. Juni 2001: Well UNIX _allows_ you to be creative: use /tmp/xx.$$.someting This is a gaping security hole. Especially given that cdrecord is often used as root or setuid root. If you prove me that you gained root privilleges using a recent cdrecord in suid mode, I'll give you a bottle of Champain. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
From [EMAIL PROTECTED] Tue Jun 19 23:25:33 2001 [EMAIL PROTECTED] schrieb am Sonntag, den 17. Juni 2001: If you are going to write a security related application, this note may be useful. Unfortunately your thoughts are based on a wrong assumption: There is only one reason to hide the name of a tempfile from other people: You are going to write a security relevant appliaction where people could gain something from attaking the /tmp files. You see, it only applies to suid or sgid applications. So having a script which, when run by a user, can cause the user's mail spool file to be overwritten is not a security problem? If youremove the old file first, you may be close to 100% sure that there is no such problem. Note that many UNIX programs create /tmp/ files and some of them make it easy to know the names in advance. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
From [EMAIL PROTECTED] Tue Jun 19 23:50:08 2001 So sprach Walter Hofmann am Tue, Jun 19, 2001 at 11:42:15PM +0200: I know what I'm going to choose. Yep, so do I. But still J=F6rg's point is somewhat valid. If there's no mktemp on the system, a portable way has to be chosen. Even if the portable way is FAR less capable. It all boils down to how portable the thing is supposed to be. BTW: Is (c)dialog preinstalled on (most) Unices? If not, then it needs to be installed to make the CDcontrol scripts run anyway. And while installing What should this be? It is not part of UNIX. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am Jorg Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
On Tue, 19 Jun 2001, Alexander Skwar wrote: So sprach Walter Hofmann am Tue, Jun 19, 2001 at 11:42:15PM +0200: I know what I'm going to choose. Yep, so do I. But still Jörg's point is somewhat valid. If there's no mktemp on the system, a portable way has to be chosen. Even if the portable way is FAR less capable. It all boils down to how portable the thing is supposed to be. Putting the temp file in the user's home directory being the obvious choice. Walter -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
Joerg Schilling schrieb am Mittwoch, den 20. Juni 2001: You are going to write a security relevant appliaction where people could gain something from attaking the /tmp files. You see, it only applies to suid or sgid applications. So having a script which, when run by a user, can cause the user's mail spool file to be overwritten is not a security problem? If youremove the old file first, you may be close to 100% sure that there is no such problem. Note that many UNIX programs create /tmp/ files and some of them make it easy to know the names in advance. The chance that this happens by accident is, indeed, very small. I was thinking about an attacker who deliberately and repeatedly creates links from files /tmp.123 to /var/spool/mail/username and also creates some additional load to make the window large enough. This seems very feasible for an attacker. Walter -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: CDcontrol software released (
Joerg Schilling schrieb am Mittwoch, den 20. Juni 2001: From [EMAIL PROTECTED] Tue Jun 19 23:25:06 2001 [EMAIL PROTECTED] schrieb am Sonntag, den 17. Juni 2001: Well UNIX _allows_ you to be creative: use /tmp/xx.$$.someting This is a gaping security hole. Especially given that cdrecord is often used as root or setuid root. If you prove me that you gained root privilleges using a recent cdrecord in suid mode, I'll give you a bottle of Champain. Note that I was not claiming that cdrecord has a security hole, but some script. Walter -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
