Re: [expert] Bad hdparm settings in rc.sysinit
On Mon, 5 Jul 1999, Jean-Louis Debert wrote: Axalon wrote: Try this on that MVP3 board you should see vast improvements -- # Optimisation of Hard drive. if [ -x /sbin/hdparm ];then LIST_HD=$(grep '^hd.*' /var/log/dmesg|\ grep -ivE '(CD.*ROM|FLOPPY|TAPE|STATUS|DVD)'|cut -d: -f1|sort|uniq) for i in $LIST_HD;do action "Starting Hard Drive optimisations for $i" \ hdparm -d1 -u1 -X66 /dev/$i #^ I get another 5mb vs. useing 33 done fi -- No resets no coruption Model=Maxtor 90576D4, FwRev=WAS8283C, on an atrend ATC-5220 I'll give it a try, but I don't think the X66 will work: it is intended for ATA/66, right ? Well the MVP3 (VT82C586B southbridge) only supports UDMA/33, for ATA/66 I would need MVP4 (VT82C686 southbridge). At least this is what the VIA docs say. Besides, I'm pretty sure that the disk doesn't support it as well ... Thanks anyway. -- Jean-Louis Debert[EMAIL PROTECTED] 74 Annemasse France old Linux fan Right thats why i mentioned the 33 vs. 66, you mentioned the drive only supported the 33. This board has the VT82C586/B, but the drive will still be a limit. -- Bus 0, device 7, function 0: ISA bridge: VIA Technologies VT 82C586 Apollo ISA (rev 65). Medium devsel. Master Capable. No bursts. Bus 0, device 7, function 1: IDE interface: VIA Technologies VT 82C586 Apollo IDE (rev 6). Medium devsel. Fast back-to-back capable. Master Capable. Latency=64. I/O at 0xe000 [0xe001]. Bus 0, device 7, function 3: Host bridge: VIA Technologies VT 82C586B Apollo ACPI (rev 16). Medium devsel. Fast back-to-back capable. --
Re: [expert] Time jump / Anacrontab
Le "sam, 03 jui 1999", Axalon a écrit : / On sam, 03 jui 1999, Axalon wrote: ]On Fri, 2 Jul 1999, Stephen Carville wrote: ] -On Fri, 2 Jul 1999, Eric Simoëns wrote: ] ] - Hello ! ] - Time has a strange behaviour on my Linux Box since I installed Mandrake 6.0. ] - I'am in GMT+2, so I told this during the install. ] - Date says : ] - [root@mambo /opt]# date ; date -u ] - ven jui 2 08:04:57 CEST 1999 ] - ven jui 2 06:04:57 UTC 1999 ] - It's Ok ! ] - But a few hours later, the time jumps two hours ahead, both UTC and CEST. ] - (i.e. my kde clock shows "17:09", and one minute later "19:10". Time to go home ]! ;) ] - So I have to set the date back. Tryed to do so using date -s ; date -u -s ; ]linuxconf ] - with various GMT+2 settings (Europe/Paris, Posix/Europe/Paris, etc.), but still ]a ] -couple of ours later, time jumps two hours ahead. ] - ] - Never met this problem on previous installs. (Including Mandrake 5.3.) ] - Anyone ever heard about this time travel implemention ? ] - ] - Thanks in advance, ] - Eric ... Thank you for your answers. As a matter of fact, my computer clock *has* to be in local time because Linux has to share it from time to time with an other OS wich can't deal with UTC clocks. I checked that my conf is correct regarding this setting, and everything is ok : - "[ ] Hardware clock set to GMT" is unchecked in timeconfig - /etc/localtime - ../usr/share/zoneinfo/Europe/Paris So it seems there is a process that runs periodically to changes the hardware clock... But what could it be ? I haven't found anything like that in crontab. (It's empty.) Anacron is installed but doesn't seem to be launched during boot process. [BTW, It seems to be something wrong in /etc/anacrontab ? ... 7 10 cron.weekly run-parts /etc/cron.hourly 7 10 cron.weekly run-parts /etc/cron.weekly ... ] Still a mystery ! :( Does someone know of what could cause that time-jump ? Thanks, Eric
Re: [expert] Boot troubles
On Mon, Jul 05, 1999 at 03:49:32PM +, Dan C. Stefanescu wrote: I have trouble updating to 6.0. Here are the particulars: My computer is a Micron Powerdigm Xsu with dual CPUs(Pentium II 300MHz), Adaptec 2940 UW SCSI controller on which there is a 9G hard disk, an Iomega Jaz 2G, a Plextor CDROM 32X and aPlexwriter CDR. There is nothing on the IDEs. The system is currently running RedHat 5.2. Attempts to install RH and/or Mandrake 6.0, either from floppy or from CDROM(which is bootable), fail very early, before getting to any dialog screens. Here are the last few lines before the installation freezes: RAM disk driver initialized: 16 RAM disks of 4096K size PIIX4: IDE controller on PCI bus 00 dev 09 PIIX4: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xfcd0-0xfcd7, BIOS settings: hda:pio, hdb:pio ide1: BM-DMA at 0xfcd8-0xfcdf, BIOS settings: hdc:pio, hdd:pio PCI_IDE: unknown IDE controller on PCI bus 00 device c9, VID=1042, DID=3020 PCI_IDE: device enabled (Linux) PCI_IDE: will probe irqs later ide2: BM-DMA at 0xfcc0-0xfcc7, BIOS settings: hde:pio, hdf:pio ide3: BM-DMA at 0xfcc8-0xfccf Any help will be greatly appreciated. Dan Stefanescu You need the updated lilo.rpm. The old did just that with SMP systems. tom -- "Everybody is someone else's newbie" (Marilyn Manson, edited) Thomas 'tom' Berger, [EMAIL PROTECTED] No spam, no UCE. 'Nuff said. Get Answers! Visit Mandrake Answers on http://aolmfaq.tsx.org!
Re: [expert] Time jump / Anacrontab
On Mon, Jul 05, 1999 at 04:49:27PM +0200, Eric Simoëns wrote: Le "lun, 05 jui 1999", John Aldrich a écrit : / On lun, 05 jui 1999, John Aldrich wrote: ]On Mon, 05 Jul 1999, you wrote: ] So it seems there is a process that runs periodically to changes the hardware clock... ] But what could it be ? ] ]Hmm...are you running "timed"??? That could explain it, especially if you run it with the "-F ]servername" extension as I do from time to time. ... No, my Linux box is on its own for the time. That's what is surprising me and makes me suspect some internal process... Thanks Eric Hum. Any suspicous looking processes in 'top'? What are you running besides the usual stuff? tom -- "Everybody is someone else's newbie" (Marilyn Manson, edited) Thomas 'tom' Berger, [EMAIL PROTECTED] No spam, no UCE. 'Nuff said. Get Answers! Visit Mandrake Answers on http://aolmfaq.tsx.org!
[expert] Building first kernel after installing Mandrake 6.0
I did: 1) make xconfig 2) selected my FreeBSD file systems and mcdx cdrom 3) make dep 4) make clean 5) make 6) make install 7) make modules 8) make modules_install 9) lilo On reboot using lilo on MBR yields: "modprobe: can't locate module net-pf-1" mount yields: -- [root@localhost hodges]# mount /dev/hda5 on / type ext2 (rw) none on /proc type proc (rw) /dev/hda1 on /boot type ext2 (rw) /dev/hda9 on /home type ext2 (rw) /dev/hda6 on /usr type ext2 (rw) /dev/hda8 on /usr/local type ext2 (rw) /dev/hda7 on /var type ext2 (rw) none on /dev/pts type devpts (rw,mode=0622) -- lilo.conf contains: - boot = /dev/hda delay = 100 timeout = 150 prompt read-only map=/boot/map install=/boot/boot.b image = /boot/vmlinuz-2.2.9-19mdk label = linux root = /dev/hda5 I have grepped the all through /usr/src for something like net-pf to no avail. BTW. when booted from my the floppy the kernel does access my BSD file system and the cdrom!!! Did I do something incorrect in xconfig? or what do i need to do to make a good kernel? Thanks, Bill
RE: [expert] Building first kernel after installing Mandrake 6.0
Did you do [root@localhost hodges] cp zimage /boot/vmlinuz-x.x.x ?? Just want to make sure you did the cp before you ran the Lilo command. James -Original Message- From: BillVirginia Hodges [SMTP:[EMAIL PROTECTED]] Sent: Monday, July 05, 1999 11:35 AM To: Expert Mandrake Subject:[expert] Building first kernel after installing Mandrake 6.0 I did: 1) make xconfig 2) selected my FreeBSD file systems and mcdx cdrom 3) make dep 4) make clean 5) make 6) make install 7) make modules 8) make modules_install 9) lilo On reboot using lilo on MBR yields: "modprobe: can't locate module net-pf-1" mount yields: -- [root@localhost hodges]# mount /dev/hda5 on / type ext2 (rw) none on /proc type proc (rw) /dev/hda1 on /boot type ext2 (rw) /dev/hda9 on /home type ext2 (rw) /dev/hda6 on /usr type ext2 (rw) /dev/hda8 on /usr/local type ext2 (rw) /dev/hda7 on /var type ext2 (rw) none on /dev/pts type devpts (rw,mode=0622) -- lilo.conf contains: - boot = /dev/hda delay = 100 timeout = 150 prompt read-only map=/boot/map install=/boot/boot.b image = /boot/vmlinuz-2.2.9-19mdk label = linux root = /dev/hda5 I have grepped the all through /usr/src for something like net-pf to no avail. BTW. when booted from my the floppy the kernel does access my BSD file system and the cdrom!!! Did I do something incorrect in xconfig? or what do i need to do to make a good kernel? Thanks, Bill
Re: [expert] Bad hdparm settings in rc.sysinit
On Mon, 5 Jul 1999, Wolfgang Bornath wrote: Axalon wrote: This board has the VT82C586/B, but the drive will still be a limit. -- Bus 0, device 7, function 0: ISA bridge: VIA Technologies VT 82C586 Apollo ISA (rev 65). Medium devsel. Master Capable. No bursts. Bus 0, device 7, function 1: IDE interface: VIA Technologies VT 82C586 Apollo IDE (rev 6). Medium devsel. Fast back-to-back capable. Master Capable. Latency=64. I/O at 0xe000 [0xe001]. Bus 0, device 7, function 3: Host bridge: VIA Technologies VT 82C586B Apollo ACPI (rev 16). Medium devsel. Fast back-to-back capable. -- It may not have been an error, just looked at what i posted and noticed i left the -q's out to make it be quiet. You need to add a -q before each option. Check with "hdparm /dev/hd?" -- /dev/hda: multcount= 0 (off) I/O support = 0 (default 16-bit) unmaskirq= 1 (on) using_dma= 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead= 8 (on) geometry = 700/255/63, sectors = 11255328, start = 0 -- I tried Axalons script but it didn't work. I don't know why because I saw only a message about something not found passing by during bootup. It wasn't written in dmesg though. Here's my excerpt from /proc/pci: --- Bus 0, device 7, function 1: IDE interface: VIA Technologies VT 82C586 Apollo IDE (rev 6). Medium devsel. Fast back-to-back capable. Master Capable. Latency=64. I/O at 0xe000. Bus 0, device 7, function 0: ISA bridge: VIA Technologies VT 82C586 Apollo ISA (rev 71). Medium devsel. Master Capable. No bursts. Bus 0, device 1, function 0: PCI bridge: VIA Technologies Unknown device (rev 0). Vendor id=1106. Device id=8598. Medium devsel. Master Capable. No bursts. Min Gnt=4. Bus 0, device 0, function 0: Host bridge: VIA Technologies VT 82C597 Apollo VP3 (rev 4). Medium devsel. Fast back-to-back capable. Master Capable. Latency=16 -- /sbin/hdparm is there so I don't know what it was that didn't work. I didn't get any severe errors though. Everything worked as before. wobo -- Linux Mandrake's Home: http://www.linux-mandrake.com Mandrake Answers(English): http://aolmfaq.tsx.org/faq.html Mandrake Answers(Deutsch): http://people.frankfurt.netsurf.de/wobo ## LLaP (Linux Lovers are Perfect!) # Well thats just not fair you have a newer "82C586 Apollo ISA" ;)
Re: [expert] Building first kernel after installing Mandrake 6.0
Well the mcdx is built as a module, but unless Bernard has changed the .config i don't see bsd partitions supported. net-pf-1 is unix sockets, you either A) did not include it B) Built it as a module and forgot to "alias net-pf-1" in your conf.modules C) Truely didn't want unix sockets, and need to "alias net-pf-1 off" On Mon, 5 Jul 1999, BillVirginia Hodges wrote: I did: 1) make xconfig 2) selected my FreeBSD file systems and mcdx cdrom 3) make dep 4) make clean 5) make 6) make install 7) make modules 8) make modules_install 9) lilo On reboot using lilo on MBR yields: "modprobe: can't locate module net-pf-1" mount yields: -- [root@localhost hodges]# mount /dev/hda5 on / type ext2 (rw) none on /proc type proc (rw) /dev/hda1 on /boot type ext2 (rw) /dev/hda9 on /home type ext2 (rw) /dev/hda6 on /usr type ext2 (rw) /dev/hda8 on /usr/local type ext2 (rw) /dev/hda7 on /var type ext2 (rw) none on /dev/pts type devpts (rw,mode=0622) -- lilo.conf contains: - boot = /dev/hda delay = 100 timeout = 150 prompt read-only map=/boot/map install=/boot/boot.b image = /boot/vmlinuz-2.2.9-19mdk label = linux root = /dev/hda5 I have grepped the all through /usr/src for something like net-pf to no avail. BTW. when booted from my the floppy the kernel does access my BSD file system and the cdrom!!! Did I do something incorrect in xconfig? or what do i need to do to make a good kernel? Thanks, Bill
RE: [expert] Building first kernel after installing Mandrake 6.0
from arch/i386/boot/Makefile, install: $(CONFIGURE) $(BOOTIMAGE) sh -x ./install.sh $(KERNELRELEASE) $(BOOTIMAGE) $(TOPDIR)/System.map "$(INSTALL_PATH)" He used make install, if the kernel wasn't in place when lilo was ran it wouldn't have ran, and if by some fluke he did get it booted like that, he would not be encountering the "can't locate module net-pf-1" cause it's built into the kernel we shiped. On Mon, 5 Jul 1999, James J. Capone wrote: Did you do [root@localhost hodges] cp zimage /boot/vmlinuz-x.x.x ?? Just want to make sure you did the cp before you ran the Lilo command. James -Original Message- From: BillVirginia Hodges [SMTP:[EMAIL PROTECTED]] Sent: Monday, July 05, 1999 11:35 AM To: Expert Mandrake Subject: [expert] Building first kernel after installing Mandrake 6.0 I did: 1) make xconfig 2) selected my FreeBSD file systems and mcdx cdrom 3) make dep 4) make clean 5) make 6) make install 7) make modules 8) make modules_install 9) lilo On reboot using lilo on MBR yields: "modprobe: can't locate module net-pf-1" mount yields: -- [root@localhost hodges]# mount /dev/hda5 on / type ext2 (rw) none on /proc type proc (rw) /dev/hda1 on /boot type ext2 (rw) /dev/hda9 on /home type ext2 (rw) /dev/hda6 on /usr type ext2 (rw) /dev/hda8 on /usr/local type ext2 (rw) /dev/hda7 on /var type ext2 (rw) none on /dev/pts type devpts (rw,mode=0622) -- lilo.conf contains: - boot = /dev/hda delay = 100 timeout = 150 prompt read-only map=/boot/map install=/boot/boot.b image = /boot/vmlinuz-2.2.9-19mdk label = linux root = /dev/hda5 I have grepped the all through /usr/src for something like net-pf to no avail. BTW. when booted from my the floppy the kernel does access my BSD file system and the cdrom!!! Did I do something incorrect in xconfig? or what do i need to do to make a good kernel? Thanks, Bill
Re: [expert] Problem solved -- sort of (IDE problems?)
On Sun, 4 Jul 1999, Mike Abney wrote: To re-cap: Both my external modem and my parallel-port Zip drive were working fine in RH 5.2 (and Win '95, but who cares). I tried installing both RH6.0 and LM6.0 and neither seemed to be able to find either the modem nor the Zip. After that, even RH5.2 would not reinstall and detect them. I'm not sure *exactly* which setting did it, but I got RH5.2 working again by playing with the BIOS. Once that happened, I thought I'd play around with trying to upgrade/install the 6.0 products again. *Neither* worked. They both fail to detect my Zip drive. *However* I did notice this little detail. In RH5.2, during the boot process, the following shows up: Ok. use dmesg and grep for scsi. If you don't find "scsi: 0 hosts found" , then scsi support isn't in the kernel. At a that point, a recompile of the kernel is needed. If you have a Zip drive made before august of 1998, then you need to run modprobe ppa to get it recognized. Do this from a console command line, as the kernel messages won't show up in an X windows terminal session. if it is after 1998, then run modprobe imm to get it recognize. then you can run mount -t vfat /dev/sda4 /zip to mount it on the previously created /zip directory (use mkdir). bug
Re: [expert] Boot troubles
Axalon wrote: On Mon, 5 Jul 1999, Tom Berger wrote: On Mon, Jul 05, 1999 at 03:49:32PM +, Dan C. Stefanescu wrote: I have trouble updating to 6.0. Here are the particulars: My computer is a Micron Powerdigm Xsu with dual CPUs(Pentium II 300MHz), Adaptec 2940 UW SCSI controller on which there is a 9G hard disk, an Iomega Jaz 2G, a Plextor CDROM 32X and aPlexwriter CDR. There is nothing on the IDEs. The system is currently running RedHat 5.2. Attempts to install RH and/or Mandrake 6.0, either from floppy or from CDROM(which is bootable), fail very early, before getting to any dialog screens. Here are the last few lines before the installation freezes: RAM disk driver initialized: 16 RAM disks of 4096K size PIIX4: IDE controller on PCI bus 00 dev 09 PIIX4: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xfcd0-0xfcd7, BIOS settings: hda:pio, hdb:pio ide1: BM-DMA at 0xfcd8-0xfcdf, BIOS settings: hdc:pio, hdd:pio PCI_IDE: unknown IDE controller on PCI bus 00 device c9, VID=1042, DID=3020 PCI_IDE: device enabled (Linux) PCI_IDE: will probe irqs later ide2: BM-DMA at 0xfcc0-0xfcc7, BIOS settings: hde:pio, hdf:pio ide3: BM-DMA at 0xfcc8-0xfccf Any help will be greatly appreciated. Dan Stefanescu You need the updated lilo.rpm. The old did just that with SMP systems. tom Can't be lilo he can't even boot fully to the installer. Dan, if you can disable the ide interfaces do so. Does the system really contain 4 ide chains, I believe the machine has two physical IDE slots and the (Phoenix 1.29)BIOS setup allows you to set/choose 4 devices (hard disks or CDs) corresponding to the 2 IDEs and the Master/Slave position. I ended up setting them all to None. Furthermore, 5.2 boot sequence mentions only ide0 and ide1. and is that exactly where it stops? Yes.
[expert] Talk Problem
hi all, using lm 5.3: a friend telnetted into my machine and we tried to use 'talk'. kept getting the following messages: [No connection yet] [Checking for invitation on caller's machine] and there it would stay and not do another thing. any help will be much appreciated karen
[expert] Modem Icon
anyone an expert on the kde? using LM 5.3 i was connected to the internet and a friend telnetted into my machine. he wanted me to go out of kde to the command, which i did; and when i came back into kde, the modem icon in the tray on the panel was gone. and since i'm hooked on those red/green blinking lights :-), am wondering if there is a way to get it back into the tray and its still connected with my isp. any help will be much appreciated karen
[expert] What's going on??
Hi! I installed Linux-Mandrake and Red Hat , version 6.0 as soon as they were released. I partitioned my hard drives with both Disk Druid and Fdisk and received the same results. All I got was a small boot partition and everthing else dumped into over eight (8) GB of an extended partition with multiple logical drives. This is not what I ordered. I wanted to have three(3) primary partitions and one extended partition, into which I would create as many logical drives as I saw fit for my installation. I came to linux over a year ago because I tried to install a sound card in Windoz 98 at IRQ 5 and DMA 1 and 5, and some son-of-a-bitch in Redmond WA had decided that I could not use those settings. In previous versions of Red Hat, Mandrake and SuSe I had no problem partitioning the HD any way I chose. What gives?? Do we have a case of "WE KNOW BETTER" or is there a reasonable explanation? Is Linux going the way of MickeySoft? Or is this just Red Hat and Mandrake? Perhaps someone who makes these decisions can reply? Rusty
Re: [expert] Modem Icon
On Mon, 05 Jul 1999, Karen R wrote: anyone an expert on the kde? using LM 5.3 i was connected to the internet and a friend telnetted into my machine. he wanted me to go out of kde to the command, which i did; and when i came back into kde, the modem icon in the tray on the panel was gone. and since i'm hooked on those red/green blinking lights :-), am wondering if there is a way to get it back into the tray and its still connected with my isp. Did you log out of KDE? If so, it shut down all programs started under KDE, including kppp. If you want the connection to start automatically everytime you start KDE, just drag a link to kppp into the autostart directory on your desktop. Under LM5.3, you can find kppp under /opt/kde/bin. Just drag it over and select "link" when it asks you. A better way to get to the command line would be 1) start an xterm or konsole; or 2) type ctrl+alt+F2, then type ctrl+alt+F7 to get back to KDE. -- Arandir... ___ http://www.meer.net/~arandir/
[expert] re modem lcon
click on the kppp icon, then click on the setup button ,then the ppp tab ,then just select "dock into panel on connect" under the ppp config panel.
Re: TheRe: [expert] Building first kernel after installing Mandrake6.0
On Mon, 5 Jul 1999, BillVirginia Hodges wrote: The 6.0 makefile appears to do the copying automatically when it does installs. The time stamps appear to confirm this. The suggestion to include sockets got me around the previous problem Now it give s about 8-10 lines more of the boot process and stops on "Finding module dependencies" Do you have a suggestion of where I should look now? Sure.. (cd /lib/modules mv `uname -r` `uname -r`.backup) move the modules out of there, and rerun make modules_install as previously(recently) noted (maybe on the other lists) by Bernard you can hit ctrl-c when it stops there and it will continue to boot so you can get in there to smack it or what ever needs done, in this case mv'ing and reinstalling the modules should work. Thanks, Bill
