June 2004 - openSUSE Users - openSUSE Mailing Lists

quick patch request for SuSE 9.0 AMD 3ware users
by Joe Landman 01 Jun '04

01 Jun '04

Hi Folks: I wanted to let you know that the 3ware drivers in 9.0 are somewhat out of date. Also, under 9.0 AMD, there is a simple bug in sd.c with regards to signed vs unsigned ints, which limits the size of the SCSI disk system to under 1 TB (1 TB - 1 byte will work, but 1 TB will overflow). See https://www.3ware.com/kbadmin/attachments/sd.c.patch for the patch to sd.c, and look at the knowledgebase article "Q10753 - Software Configuration: Does 3ware have support for 64-bit Linux operating systems?" on http://www.3ware.com/KB/kb.asp. They mention 'There is a bug with 64-bit Linux 2.4 kernels where SCSI disk targets greater than 1 Terabyte in capacity cannot be accessed beyond one Terabyte (1 TByte). The resultant output for the error is: "I/O error: device 08:00 sector X" ' Please include the latest 3ware drivers and the this patch for 9.0 AMD64 kernels if possible. Thanks! Joe -- Joe Landman <landman(a)scalableinformatics.com>

1 0

Suse 9.1 and modem - problem
by Basil Chupin 01 Jun '04

01 Jun '04

In my message of 18/5 with Subject of "SUSE 9.1 - BAH!" I mentioned that I was having big problems with getting Suse 9.1 to co-operate with my modem/modems (either internal or external - but as I have replaced my internals with externals my comments below only refer to external modems). Since that message of 18/5 I've been doing some testing and have found this (summarised): (a) during the installation of 9.1, at the point where the modem is first configured, is then tested for connectivity with the ISP and then the YOU update is allowed to be done - at this point the modem works perfectly. I get a consistent thruput of 4.4 - 5.9 Kb/s (apart from the occasional drop, of course). However, once the REST of the Suse installation is completed, the modem almost stops working with thruput between 0 and 800+ b/s (not Kb/s but b/s) coupled with periods of total inactivity lasting minutes. My testing has shown that it is NOT the YOU update that causes the modem problem but the completion of the installation- meaning that something gets installed that causes the problem. Configuring the modem during the installation, allowing it to dial the ISP - but not letting YOU to update - and then completing the installation does not solve the modem problem; (b) doing a full install of anything OTHER than KDE-related applications does NOT cause problems with the modem. If, for example, one installs the GNOME Desktop and not KDE then the modem works perfectly after the full installation. In one test I only installed the "Minimum graphical system (without KDE)" - and got the FVWM 'desktop' - and things worked fine but then I added GAMES (which are KDE) to this system. Immediately the modem stopped working correctly; (c) C Hamel and I have been corresponding privately and he mentioned earlier tonight that he found a "funny" about the KDE application "k3b". As a test I re-installed Suse 9.1 with KDE Desktop, Multimedia, Games, plus a couple of other things I normally insall, BUT I deliberately omitted installing k3b. After the full installation the modem was behaving almost normally with thruput again back to 4.4 - 5.9 Kb/s but with quite a number of slowdowns and stops. This degradation in performance could have been caused by the fact that in our part of the world at that time the Net is used pretty heavily. I will be retesting this during the day time (later today) to see if the performance is still below par. In summary, something KDE-related is causing the problem with the modem(s) - and it may be possible that it is k3b. I would pass this info onto to Suse Tech Support but I can't find their address right now. If someone knows it could they please forward this message to them on my behalf? (I got my Suse 9.1 on 13 May but haven't been able to use the darn thing because of the problem with the modem - and I'm getting p***** off :-(.) Cheers. -- I am not young enough to know everything.

5 11

sysvinit bug in 9.1?
by Kastus 01 Jun '04

01 Jun '04

Can anybody running 9.1 confirm the following? sysvinit-2.85-20 shipped with 9.1 has a bug setting HOME environment variable. If you run the command "tr '\0' '\n' < /proc/1/environ" as root, it reports: OME=/ TERM=linux vga=0x317 splash=silent Note missing H in HOME. Possible impact is on apcupsd package. Has anybody reported this to SUSE feedback? If not, I will do that. Thanks, -Kastus

6 7

K3B & Ripping
by Erik Jakobsen 01 Jun '04

01 Jun '04

Hi. I'm using SuSE 9.1, and would rip an audiocd with K3B, but I couldn't do it. Can anybody of you do it, and how did you solve the problem ?. -- Med venlig hilsen - Best regards Erik Jakobsen - eja(a)urbakken.dk SuSE Linux 9.1 Proff.

2 2

Philips Webcam module (pwc) on SuSE 9.1
by Jay Vollmer 01 Jun '04

01 Jun '04

Hi everybody. If I'm not mistaken, the pwc module is hosed in the 9.1 kernel. What are Webcam users among you doing to make use of your webcams, short of recompiling to include pwc? Thanks for any advice - but prefer good advice. -- JAY VOLLMER JVOLLMER(a)CONSOLIDATEDLINT.COM TEXT REFS DOUBLEPLUSUNGOOD SELFTHINK VERGING CRIMETHINK IGNORE FULLWISE

1 0

Looking for an explanation
by eddie 01 Jun '04

01 Jun '04

This morning I couldn't log in (KDE), but could log in at console. After a good deal of searching and experimentation I discovered that my /tmp had run out of space, and that my /tmp/kde-eddie was 689MB! Can anyone say why kde-eddie should have grown to that size? How can I prevent a reoccurrence? By removing kde-eddie I can log in now and everything seem fine (so far). Any thoughts, anyone? Thanks Eddie

3 3

ntegrated card reader problem
by Shawn Tanner 01 Jun '04

01 Jun '04

I have a Gateway M675 laptop and have a small problem with the integrated USB card reader. If I boot with a card in the reader, it behaves as I believe it should. The card will be accessible and switching cards will result in a knoqueror window opening with the contents of the new card. There will also be an icon in the "My Computer" folder which links to the card. If I boot without the card in the reader there will not be a link created in "My Computer" and a konqueror window will not open up. In this case I can mount the card manually (mount -t auto /dev/sdb1 /media/SDCard) and if I do, konqueror will suddenly pop up a window saying new hardware has been detected. This isn't catastrophic because I can acces the card regardless but I dislike having to access it different ways depending on how I booted. Any solutions or explanations would be appreciated.

2 1

PCMCIA help needed
by clarkt＠cnsp.com 01 Jun '04

01 Jun '04

I have 2 amd 64 laptops. Both loaded with Suse 9.1 32 & 64. The first is an Emachines 6807, the 2nd is a Compaq R3140. Problem -- Cards that can be setup and used on the Emachines (generic pcmcia controler) but not on the Compaq. The Compaq has a TI-1620 pcmcia controler that has to be accessed through a pci bridge. I have entered the following into my etc/pcmcia/config.opts --include port 0x3000-0x7fff, memory 0xe8100000-0xe97fffff This allows me to access the controler. Now a modem or aircard 555 are seen, but there is not connection to /dev/ttyS0. >From the system long on the 6807 --- May 30 23:06:39 linux91 kernel: Linux Kernel Card Services May 30 23:06:39 linux91 kernel: options: [pci] [cardbus] [pm] May 30 23:06:39 linux91 kernel: PCI: IRQ 11 for device 0000:00:0a.0 doesn't match PIRQ mask - try pci=usepirqmask May 30 23:06:39 linux91 kernel: PCI: Found IRQ 11 for device 0000:00:0a.0 May 30 23:06:39 linux91 kernel: PCI: Sharing IRQ 11 with 0000:00:10.1 May 30 23:06:39 linux91 kernel: Yenta: CardBus bridge found at 0000:00:0a.0 [161f:2032] May 30 23:06:39 linux91 kernel: Yenta: Enabling burst memory read transactions May 30 23:06:39 linux91 kernel: Yenta: Using CSCINT to route CSC interrupts to PCI May 30 23:06:39 linux91 kernel: Yenta: Routing CardBus interrupts to PCI May 30 23:06:39 linux91 kernel: Yenta: ISA IRQ mask 0x0000, PCI irq 11 May 30 23:06:39 linux91 kernel: Socket status: 30000010 May 30 23:06:39 linux91 kernel: cs: IO port probe 0x0c00-0x0cff: clean. May 30 23:06:39 linux91 kernel: cs: IO port probe 0x0820-0x08ff: clean. May 30 23:06:39 linux91 kernel: cs: IO port probe 0x0800-0x080f: clean. May 30 23:06:39 linux91 kernel: cs: IO port probe 0x03e0-0x04ff: excluding 0x4d0-0x4d7 May 30 23:06:39 linux91 kernel: cs: IO port probe 0x0100-0x03af: clean. May 30 23:06:39 linux91 kernel: cs: IO port probe 0x0a00-0x0aff: clean. May 30 23:06:39 linux91 kernel: cs: memory probe 0xa0000000-0xa0ffffff: clean. May 30 23:06:39 linux91 kernel: spurious 8259A interrupt: IRQ7. May 30 23:06:39 linux91 kernel: ttyS0 at I/O 0x3f8 (irq = 11) is a 16550A May 30 23:06:39 linux91 kernel: NET: Registered protocol family 17 >From the Compaq -- Jun 1 17:36:15 s91 kernel: cs: memory probe 0x0c0000-0x0fffff: excluding 0xc0000-0xfffff Jun 1 17:36:15 s91 kernel: cs: memory probe 0x60000000-0x60ffffff: excluding 0x60000000-0x60ffffff Jun 1 17:36:15 s91 kernel: cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa0ffffff Jun 1 17:36:15 s91 kernel: cs: memory probe 0xe8100000-0xe97fffff: excluding 0xe8100000-0xe826ffff 0xe83e0000-0xe86bffff 0xe8b10000-0xe8c7ffff 0xe8f60000-0xe90cffff 0xe93b0000-0xe951ffff Jun 1 17:36:15 s91 cardmgr[1841]: socket 0: Serial or Modem and the var/lib/pcmcia/stab Socket 0: Serial or Modem 0 serial serial_cs 0 ttyS0 4 64 Socket 1: empty serial_cs is loaded and shown in lsmod. pci=noacpi pci=usepirqmask used on both machines. Any suggestions or help. Clark

1 0