Re: [SLE] 7.1 Won't Install, and Worse
On Mon, 19 Mar 2001, cll muzh said:
Low-level formats should never be necessary, IMHO. A complete format+check does the same sort of thing.
I question whether that's true in all cases. I have a couple of damaged IDE drives from my portables that won't format. If I had a low level format utility I might be able get around the problem. I think surface damage issues are sometimes best handled this way. I also think it depends on the drive design. I'm not sure that all drives are designed to do auto sector re-allocation when they encounter read / write errors. And if it's available, I don't see why I shouldn't use it - it's not black magic. ---------------------------------------------------------------------- John Karns jkarns@csd.net
John and Samy, The question of whether a low-level format is necessary or not is sort of moot for me, at the moment. I went back and did a separate low-level format and verify, no errors. I didn't really think there were. So then I inserted the 7.1 personal bootdisk in the floppy drive, it started up, went to Yast 2, after a couple of steps it told me that there was no hard drive (yes, the tekram card was in and the SCSI bios was installed), and dumped me to Yast1. I went through the autoprobe, put in the modules disk per instructions, it located the tekram SCSI card, and the Realtek NIC (but not the 3com 3C509). After that I selected installation, and there it hung. Just like it did before all of this low-level formatting, etc. Right now the 6.4/IDE system is up and running. I will go back and try to install 7.0 on the SCSI disk one more time, and then find some other use for the 7.1 package. This really looks like the end of the line for SuSE and me. Stan Koper ----- Original Message ----- From: John Karns To: cll muzh Cc: Samy Elashmawy ; SuSE Linux Sent: Sunday, March 18, 2001 1:00 PM Subject: Re: [SLE] 7.1 Won't Install, and Worse On Mon, 19 Mar 2001, cll muzh said:
Low-level formats should never be necessary, IMHO. A complete format+check does the same sort of thing.
I question whether that's true in all cases. I have a couple of damaged IDE drives from my portables that won't format. If I had a low level format utility I might be able get around the problem. I think surface damage issues are sometimes best handled this way. I also think it depends on the drive design. I'm not sure that all drives are designed to do auto sector re-allocation when they encounter read / write errors. And if it's available, I don't see why I shouldn't use it - it's not black magic. ---------------------------------------------------------------------- John Karns jkarns@csd.net -- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/support/faq
Are you using a terminator on the SCSI bus? The Seagate drive is 50 pin or 68? If it's an ultra SCSI device, I'd receommend using an active terminator. If you're not using any termination, that could well be the problem. Otherwise the drive or controller may have a problem. I've been using SCSI for a several years and never had any problems. This sounds like a possible termination issue. On Sun, 18 Mar 2001, Stan Koper said:
John and Samy,
The question of whether a low-level format is necessary or not is sort of moot for me, at the moment. I went back and did a separate low-level format and verify, no errors. I didn't really think there were.
So then I inserted the 7.1 personal bootdisk in the floppy drive, it started up, went to Yast 2, after a couple of steps it told me that there was no hard drive (yes, the tekram card was in and the SCSI bios was installed), and dumped me to Yast1. I went through the autoprobe, put in the modules disk per instructions, it located the tekram SCSI card, and the Realtek NIC (but not the 3com 3C509). After that I selected installation, and there it hung. Just like it did before all of this low-level formatting, etc.
Right now the 6.4/IDE system is up and running. I will go back and try to install 7.0 on the SCSI disk one more time, and then find some other use for the 7.1 package. This really looks like the end of the line for SuSE and me.
---------------------------------------------------------------------- John Karns jkarns@csd.net
John, The Seagate uses a 50 pin cable. It's the only device on the SCSI card. The Tekram card refers to some kind of automated terminator, but as for the drive, it came with the terminator enabled by default. I left it that way. News, tho. I did the low level format and verify, and reinstalled SuSE 7.0. Since then it's been doing some very weird stuff, but only off and on. For example, I'm getting httpd-conf errors at boot up, but not at every boot up. The warn file reports problems with PHP (3). So no surprise there. But, on some boot ups, when I go to log in, it rejects my log-in. What I mean is, I don't see the next line, "password:". As soon as I type in something and hit enter, I get "login incorrect". And sometimes that's followed by a "respawning too fast" message. And although the segmentation faults aren't as numerous as they were after the first install, they're still appearing from time-toi-time. But I just haven't yet had the time to sit down and figure out which specific problems are occuring, where, and what to do to fix them. But 7.1 still wouldn't install. Oh, and I'm getting "martian destination" messages on eth0. Finally, in previous installs, rc.config (I forget where--sorry) refers to my two NIC modules, the 3c509 and the rtl1839. Now that line includes the tekram card driver as well. I wonder if that's correct or necessary. A line for the rtl1839 card appears twice on boot up, but I'm not getting the "already loaded that module" message that I had before. Maybe I'll find some time this weekend to sit down and go through the logs, and do some research and see if I can figure out and fix some of these things. Or, I could re-install 7.0 for the third time... Thanks for your comments. Stan Koper ----- Original Message ----- From: John Karns To: Stan Koper Cc: cll muzh ; Samy Elashmawy ; SuSE Linux Sent: Monday, March 19, 2001 4:22 PM Subject: Re: [SLE] 7.1 Won't Install, and Worse Are you using a terminator on the SCSI bus? The Seagate drive is 50 pin or 68? If it's an ultra SCSI device, I'd receommend using an active terminator. If you're not using any termination, that could well be the problem. Otherwise the drive or controller may have a problem. I've been using SCSI for a several years and never had any problems. This sounds like a possible termination issue. On Sun, 18 Mar 2001, Stan Koper said:
John and Samy,
The question of whether a low-level format is necessary or not is sort of moot for me, at the moment. I went back and did a separate low-level format and verify, no errors. I didn't really think there were.
So then I inserted the 7.1 personal bootdisk in the floppy drive, it started up, went to Yast 2, after a couple of steps it told me that there was no hard drive (yes, the tekram card was in and the SCSI bios was installed), and dumped me to Yast1. I went through the autoprobe, put in the modules disk per instructions, it located the tekram SCSI card, and the Realtek NIC (but not the 3com 3C509). After that I selected installation, and there it hung. Just like it did before all of this low-level formatting, etc.
Right now the 6.4/IDE system is up and running. I will go back and try to install 7.0 on the SCSI disk one more time, and then find some other use for the 7.1 package. This really looks like the end of the line for SuSE and me.
---------------------------------------------------------------------- John Karns jkarns@csd.net -- To unsubscribe send e-mail to suse-linux-e-unsubscribe@suse.com For additional commands send e-mail to suse-linux-e-help@suse.com Also check the FAQ at http://www.suse.com/support/faq
Hi Stan, It maybe a memory related problem , especially since you are getting seg faults. Try running a proper memory check. Regards, Graham Smith -------------------------------- On 22-Mar-2001 Stan Koper wrote:
John,
The Seagate uses a 50 pin cable. It's the only device on the SCSI card. The Tekram card refers to some kind of automated terminator, but as for the drive, it came with the terminator enabled by default. I left it that way.
News, tho. I did the low level format and verify, and reinstalled SuSE 7.0. Since then it's been doing some very weird stuff, but only off and on. For example, I'm getting httpd-conf errors at boot up, but not at every boot up. The warn file reports problems with PHP (3). So no surprise there. But, on some boot ups, when I go to log in, it rejects my log-in. What I mean is, I don't see the next line, "password:". As soon as I type in something and hit enter, I get "login incorrect". And sometime s that's followed by a "respawning too fast" message.
And although the segmentation faults aren't as numerous as they were after the first install, they're still appearing from time-toi-time. But I just haven't yet had the time to sit down and figure out which specific problems are occuring, where, and what to do to fix them. But 7.1 still wouldn't install. Oh, and I'm getting "martian destination" messages on eth0.
Finally, in previous installs, rc.config (I forget where--sorry) refers to my two NIC modules, the 3c509 and the rtl1839. Now that line includes the tekram card driver as well. I wonder if that's correct or necessary. A line for the rtl1839 card appears twice on boot up, but I'm not getting the "already loaded that module" message that I had before.
Maybe I'll find some time this weekend to sit down and go through the logs, and do some research and see if I can figure out and fix some of these things. Or, I could re-install 7.0 for the third time...
Thanks for your comments.
Stan Koper
On Wed, 21 Mar 2001, Stan Koper said:
John,
The Seagate uses a 50 pin cable. It's the only device on the SCSI card. The Tekram card refers to some kind of automated terminator, but as for the drive, it came with the terminator enabled by default. I left it that way.
News, tho. I did the low level format and verify, and reinstalled SuSE 7.0. Since then it's been doing some very weird stuff, but only off and on. For example, I'm getting httpd-conf errors at boot up, but not at every boot up. The warn file reports problems with PHP (3). So no surprise there. But, on some boot ups, when I go to log in, it rejects my log-in. What I mean is, I don't see the next line, "password:". As soon as I type in something and hit enter, I get "login incorrect". And sometimes that's followed by a "respawning too fast" message.
That sounds like getty is having a problem. I've seen that in happen with mgetty (used for dial-up logins and handling a fax dialup line) if it's not passed the appropriate parameters in /etc/inittab. But SuSE doesn't install mgetty to handle VC logins, so I doubt that's it. Sounds like you have several issues there. The HD problem is almost certainly a hardware issue, possibly an IRQ conflict - check the IRQ of the controller, which is viewable and changeable within the Tekram BIOS. While running, 'lspci' should show the cfg info for the pci cards, or check /var/log/boot.msg for IRQ and i/o addresses of the devices. Try removing the ethernet cards until after completing the install and running for a while. Some SCSI facts to be aware of: Length of the SCSI cable is important. For a 50-pin cable, I believe the max length is something like 6 feet (however, a practical length is more like 2 - 3 ft). For 68 pin, I believe it's less. The rule of thumb is the shorter (and thicker) the better. Another important issue is the use of a terminator. SCSI specs call for termination at both ends of the bus. Most modern controllers implement auto termination on the card, which means that it will take of the issue on the card end of the bus. OTOH, if you should happen to connect both internal and external devices (thus placing the card in the middle of the bus), you would need a terminator on the ends of each cable, and the card would remain unterminated. All that said, in many instances you can get away without a terminator, especially when you have only a single device connected with a short cable. Also take a look at the device information for the SCSI devices to look for a clue. You could try re-installing, but without using Reiserfs to see if that would make a difference. The SCSI error you originally posted was definitely a read error, which would point to: bad termination, a problem with the cable, controller or HD. I would opt for a terminator on the cable and disable the one on the drive. Check the jumpers on the drive. You prolly shouldn't have terminator power enabled on the drive either, in case it is. If you don't have the jumper info for the drive, you should be able to find it on the Seagate Internet site. ---------------------------------------------------------------------- John Karns jkarns@csd.net
Hi! I recently updated to 7.1 (you might remember me complaining about yast2.) I have never had any problem at all getting my laserjet 6l to work with any other version of SuSE. However now if I configure it with yast2 it says it can't find my printer. It prints the "hello World" thing but still can't find it. When I configure it with yast (same way I have for years) it says: lpr: error - no default destination available. when I try to print. Any ideas? If I log in as root the command: cat file.txt > /dev/lp0 works ok (but is annoying.) Any ideas? Thanks, Matt. __________________________________________________ Do You Yahoo!? Get email at your own domain with Yahoo! Mail. http://personal.mail.yahoo.com/
* Matt Banta (tie_guy_matt@yahoo.com) [010322 17:20]: =>Hi! => =>I recently updated to 7.1 (you might remember me =>complaining about yast2.) I have never had any problem =>at all getting my laserjet 6l to work with any other =>version of SuSE. However now if I configure it with =>yast2 it says it can't find my printer. It prints the =>"hello World" thing but still can't find it. When I =>configure it with yast (same way I have for years) it =>says: lpr: error - no default destination available. =>when I try to print. Any ideas? If I log in as root =>the command: cat file.txt > /dev/lp0 works ok (but is =>annoying.) Any ideas? It sounds like the same problems I had when I tried to use CUPS...CUPS is crap right now. Use lprold or lprng..and use yast2 to configure it. YaST2 isn't designed for CUPS..KUPS is what you would have to use for it's config. -- Ben Rosenberg mailto:ben@whack.org ----- If two men agree on everything, you can be sure that only one of them is doing the thinking.
Hi Stan, * Stan Koper [Sun, 18 Mar 2001 20:21:39 -0500]: First of all, could you in the future please use a line length of about 75 chars? This makes reading and quoting of your mail much easier.
I went through the autoprobe, put in the modules disk per instructions it located the tekram SCSI card, and the Realtek NIC (but not the 3com 3C509).
Plain ISA cards can't be detected automatically, the drivers have to be inserted manually. I'd try this: - Use the DOS tool for the 3C509 to configure it at a fixed speed, i.e. either 10 mbit/sec or 100 mbit/sec. Note the IRQ and I/O port the card is configured for (if it isn't a PnP ISA card). - Boot from either floppy or CDROM. - at the boot: prompt enter manual - Manually load the modules for your SCSI controller and possibly the modules for your network cards, possibly passing parameters to the 3C509 card (for details on the parameters see the big manual). Loading the modules will give you a box with the output. For the SCSI driver you should see if it found your hard drive and for the network cards you should see if loading succeeded. -- Penguins to save the dinosaurs -- Handelsblatt on Linux for S/390
Philipp, Thanks. I'll do what I can. The 3c509 is just a 10M card, so no worries there. I just have to get used to the fact that I need set it up after the install. But of course I can't do an update via ftp, since it's the eth0 (world_dev) NIC. The Tekram SCSI card finds the SCSI drive and loads it automatically. It does it when the I turn the machine on, after the memory test and the bios screen appears. It runs through the LUN routine, then checks for SCSI devices, and finds the Seagate drive. Then the SuSE floppy boots. That's why it seems a little weird. On the other hand, the bios has provisions to boot from the CD-ROM, but even though I've set it that way, nothing in the drive (not any SuSE CD's anyway) will boot. Just one of life's little mysteries. One of these days, though, I will probably wind up installing 6.4, just to see if it works as well on the SCSI drive as it does on IDE. It's rock solid (knock on wood). Thanks, Stan Koper ----- Original Message ----- From: Philipp Thomas To: Stan Koper Cc: John Karns ; cll muzh ; Samy Elashmawy ; SuSE Linux Sent: Tuesday, March 20, 2001 6:43 PM Subject: Re: [SLE] 7.1 Won't Install, and Worse Hi Stan, * Stan Koper [Sun, 18 Mar 2001 20:21:39 -0500]: First of all, could you in the future please use a line length of about 75 chars? This makes reading and quoting of your mail much easier.
I went through the autoprobe, put in the modules disk per instructions it located the tekram SCSI card, and the Realtek NIC (but not the 3com 3C509).
Plain ISA cards can't be detected automatically, the drivers have to be inserted manually. I'd try this: - Use the DOS tool for the 3C509 to configure it at a fixed speed, i.e. either 10 mbit/sec or 100 mbit/sec. Note the IRQ and I/O port the card is configured for (if it isn't a PnP ISA card). - Boot from either floppy or CDROM. - at the boot: prompt enter manual - Manually load the modules for your SCSI controller and possibly the modules for your network cards, possibly passing parameters to the 3C509 card (for details on the parameters see the big manual). Loading the modules will give you a box with the output. For the SCSI driver you should see if it found your hard drive and for the network cards you should see if loading succeeded. -- Penguins to save the dinosaurs -- Handelsblatt on Linux for S/390
On Wed, 21 Mar 2001, Stan Koper said:
Then the SuSE floppy boots. That's why it seems a little weird. On the other hand, the bios has provisions to boot from the CD-ROM, but even though I've set it that way, nothing in the drive (not any SuSE CD's anyway) will boot. Just one of life's little mysteries.
One of these days, though, I will probably wind up installing 6.4, just to see if it works as well on the SCSI drive as it does on IDE. It's rock solid (knock on wood).
I believe this is dependent on your particular BIOS on the MB. Sometimes making the CDROM the master device and the HD the slave will help. Or what I've had to do is put the HD and the CD on separate IDE channels. ---------------------------------------------------------------------- John Karns jkarns@csd.net
participants (6)
-
Ben Rosenberg
-
Graham Smith
-
John Karns
-
Matt Banta
-
Philipp Thomas
-
Stan Koper