I am running SuSE 9.3 on a dual Opteron 246 system (Tyan Thunder K8SR
This machine has been running fine for several weeks using the on board
SCSI controllers & SW RAID. We installed the LSI MegaRAID 320-2X
controller with the TBBU03 battery backup and reinstalled SuSE 9.3.
The installation did not go flawlessly:
- Towards the end of the NFS installation, copying stopped. I switched
consoles and noticed that I couldn't ping any other addresses in our
network. I brought the ethernet interface down then back up and the
- After the install, both megaraid and megaraid_mbox modules were
loaded. I removed megaraid from INITRD_MODULES and ran mkinitrd.
At this point we considered the installation suspect, but pressed on
Now the corruption:
One user was compiling code, another user was loading a database. After
some time, g++ started getting internal compiler errors in cc1plus. I
compared the checksum of this program with another install and they were
different. I reinstalled gcc, verified checksums and everything worked
for a while. Then internal compiler errors and bogus checksum again.
I did not see anything alarming in the log files. So I booted into
rescue mode and ran reiserfsck --check without any problems. Flashed
the controller with the latest firmware and booted again.
This time I set up a 'make clean; make' loop and watched it for about
two hours without any problems. Then I started creating 4GB files with
dd and deleting them and within 15 minutes another internal compiler
- Does anyone run a similar setup without problems?
- Has anyone seen similar problems?
- Can anyone provide me some direction in tracking down this problem?
On Sat, 2005-08-20 at 15:54 -0700, Linda Walsh wrote:
> This ended up way longer than I intended, but it covers
> several factors that can affect disk performance and I
> wanted to try to give you a complete as list as I could
> think of to look for areas of change, potential causes,
> etc. Sorry it ended up so long though, it's a complex
> subject and I probably forgot to mention some items as it
This was hugely helpful. Thank you.
I came in tonight to do some more controlled experimentation with two
machines. One machine is SCSI based HW RAID. The other is IDE based SW
On each machine I partitioned the disks like this:
I ran my 'dd' test three times on each partition from each SuSE version
on the SCSI machine. The differences between 9.1 and 9.3 were small
enough I cannot determine which one is faster.
The IDE machine took so long to install 9.3 on, I took a break, checked
my email and found this note from Linda. I discovered using hdparm that
while both disks were using udma5, hda read speed was *much* slower than
hdc. I used 'raidsetfaulty' to remove the slow drive from the RAID set
and my hdparm reads and dd writes were vastly improved.
Nearly all of the servers I have available to test on have already been
upgraded to 9.3 and are IDE systems with SW RAID. The few comparable
9.1 machines I had to test were IDE systems, but not using SW RAID.
Thus I incorrectly attributed the difference to the OS version.
Thank you all for your help!
I got a Tyan K8E and loaded SuSE 9.3 on it. With the original BIOS
(1.0), hwinfo --cpu reports 1.0GHz. With the latest and greatest BIOS
2.01, hwinfo --cpu reports 1.8GHz. This is with an AMD Athlon 64 3000+.
Two other motherboards (Gigabyte and Asus) report 2.0GHz with the same
CPU. I wrote a device driver and also compared TSCs and jiffies and got
the same results. Does anyone else see this? No answer from Tyan yet.
TIA & Cheers
I've noticed a noticeable disk IO performance decrease in 9.3 vs. 9.1.
To test I've been running the following command:
time dd if=/dev/zero of=bigfile bs=8k count=500000
Here are some representative numbers for three fairly comparable
2*Opteron 242, 2 GB RAM
2 IDE drives (hda,hdc) ST312022A in SW RAID 1
dd time: 103seconds ~ 39.8MB/Sec
2*Opteron 242, 1 GB RAM
2 IDE drives (hda,hdc) ST3120026A in SW RAID 1
dd time: 608seconds ~ 6.7MB/Sec
2*Opteron 246, 1 GB RAM
2 IDE drives (hda,hdc) ST3120026A in SW RAID 1
dd time: 228seconds ~ 17.9MB/Sec
All machines were idle while running the test. All filesystems are
- Has anyone noticed a disk performance degradation moving from 9.1 to
- What can I do to track down the cause of the difference between the
9.1 machine and the two 9.3 machines as well as the fast vs. slow 9.3
When I try to compile a simple C program with -march=i686, I get the following
cc1: error: CPU you selected does not support x86-64 instruction set
I have gcc-32bit installed.
Any help appreciated.
# Department of Applied Physics and Applied Mathematics
# Columbia University
As you may remember from 'dumb guy number 1', I saved my
old /home/mike directory on a separate HD and was able to recover all of
my working files. That's the good news. The bad news is that they all
are now listed as ROOT everything (user, group). I know there's
commands to change individual files, etc. (chmod, chown), but is there a
way I can change the permissions of a directory and all of its contents
without doing each file individually?
Any ideas greatly appreciated. Otherwise I'll be here
until Thanksgiving :-\
I recently purchased an MSI RS480 board and AMD Athlon
Fortunately, the installation had the SATA modules and
I didnt have any trouble with the harddisks.
But the display is still on framebuffer. I tried
changing the driver to ati and raedon, but it didnt
help, the X server dies saying no screens found. I am
not sure if there is any other module that need to be
installed. I guess someone on this list must have run
into similar problems. If so, please help me.
Send a rakhi to your brother, buy gifts and win attractive prizes. Log on to http://in.promos.yahoo.com/rakhi/index.html
I have a few Sun Java Desktop Systems (w2100z) with the dual AMD Opteron
250 processors and the Nvidia FX 3000 DH video card. I have the 7667
driver loaded. It's running SuSE 9.3 pro. I've noticed that there is a
slight pause in mouse and monitor movement every 5 seconds or so then it
jumps to catch up to where it should be.
Does anyone have an idea what maybe causing this? TIA
Recently I made clean installation of 9.3 on X2-based PC.
Then I noticed that I cannot start desktop session properly.
I have tried to find out what is wrong.
Some of facts are as follows:
(1) when users with tcsh as default shell (tcsh users) try
to login to desktop, only xterm window without top bar
appears. It is invoked as `xterm -T Failsafe'.
(2) users with bash as defalut shell (bash users) have got no
problem; they can start their desktop session very soon.
(3) If the session is switched to Gnome-Failsafe, then
tcsh users can start desktop session, with limited environment.
(4) This phenomena do not appear on PCs running 9.3 upgraded
from either 9.1 or 9.2.
I made internet search, but I could not find a clue for my problem.
I appreciate if some kind soul can give me pointers.
Thanks for your attention.