A7V133-C primary slave problem

T

Thomas Wendell

My PC has problems when using 2nd HD. It works OK under WinXPpro (I think,
haven't stress tested), but I want to use it as a backup disk with
DriveImage 2002 and here the problems start (also under Linux). When making
a image, I naturally have to reboot to let DI load the DOS-version to backup
the C- and E-drive (E because my TEMP-env variable is set to E:\TEMP). Ok,
so far everything works, but about 15-20 min into the imagining my PC starts
to sound like an emergency vehicle (you know, uu-aa-uu-aa-uu-aa-uu at least
in Finland). It continues on, but that sound is NOT something I want to
listen to nighttime (when making my image). I'm just asking, any ideas??

(I can't check the temp when doing this, but if I exit DI and boot back to
XP, DriveHealth shows it as 42degrees C (about 105F??).
Under winXP it usually is at 39-41 degrees.
Cpu runs at 57degC, system at 41 and 1st HD at 40-41degrees, so I don't
think it's overheating). It is in a removable cradle with its own fan.
It's running with UDMA5, as my mobo can't(?) do more than that.

And the sound is definitely coming from the system, I guess it's the
S.M.A.R.T. monitoring the BIOS is doing.

The drive in question is a Seagate ST380011A 80GB disk, new as of 1 week (it
started straight away) ago
(exchanged on warranty for this problem)
HD0 is an internal Samsung SV1203N 120GB


System is: (HW) ASUS A7V133-C m/b (BIOS 1009), Athlon (T-bird) 1.2 GHz(FSB
100MHz), 384MB (SDRAM @133MHZ), 120+80GB HDs(primary IDE), DVD and
CD-RW(sec. IDE), GF2MX (32MB), Creative SB Live!Value, D-Link
NetNic(DRN-32TX),235W power supply (I know, I know, too puny), ADSL=ZyXEL
645MP-A1
OS=Win98SE_fi, all updates (incl. IE6SP1)



Any help would be appreciated, as I really don't like the sound of an
emergency vehicle when I'm trying to sleep....
(It's usually some 3 to 4 hours the imagining takes)


More info (and a guess)

I did run the test (SeaTools) both with SMART enabled or disabled in BIOS.
The test goes through w/o errors (as seen from log), but it also "generates"
this same sound after about the same time (~20min). Now I've started
thinking (with what?) :

SMART (or DriveHealth) is reporting errors in Raw Read Error Rate, Seek
Error Rate and Hardware ECC Recovered
But has these HAVE to be from the disk itself? Is it possible for the
chipset southbridge to be about to go south ( :) )? (Chipset VIA KT133A, NB
8363A
and SB 686B).Mobo was bought on 07.01.2002 (ddmmyyyy).

But why don't I get this same problem under Windows, just when accessing the
drive under Linux or DriveImage (I have no other means I can think of to
test this, I had the swap-file (pagefile.sys) on it, but no error). I'm
about to move one of my partitions from HDD-0 to HDD-1 to
test. Didn't matter..

(No, I can NOT afford a new PC or mobo/processor/memory-combination, and I'm
not wild about the idea of changing only the MB. It'd be outmoded when I
bought it. I'm unemployed)

Still looking for ideas/solutions
Below is output from SeaTools.exe






SEATOOLS.EXE
No errors detected (same as the previus HD) as seen from log:

-----------------------------------------------------------------

DIAGATA.EXE Version 3.07.30214ML
Copyright (c) 2002 by Seagate Technology, LLC. All rights reserved.

-----------------------------------------------------------------
Timer Resolution: 0.000131
Short Test Begin: 20-Apr-2004 06:59:54
Cable Test - 0 Errors
Buffer Test - 0 Errors
Identify Data
Model Number: ST380011A
Serial Number: 5JV6A4GC
Firmware Revision: 3.06
Default CHS: 16383-16-63
Current CHS: 16383-16-63
Current Capacity: 16514064 Sectors
Total Capacity: 156301488 Sectors
80 Conductor Cable Reported
ID Method: Jumper
SMART Check: Passed
Outer Diameter Scan (0 to 65025) - Passed
Elapsed Time: 30.02s
Estimated Internal Transfer Rates
Min: 35.48Mb/s Max: 347.05Mb/s Avg: 307.56Mb/s
Inner Diameter Scan (156268976 to 156301488) - Passed
Elapsed Time: 15.07s
Estimated Internal Transfer Rates
Min: 35.32Mb/s Max: 347.05Mb/s Avg: 296.97Mb/s
Random Seek (128 Cycles) - Passed
Estimated Access Times
Min: 4.449ms Max: 19.235ms Avg: 11.753ms
Random Read (128 Cycles) - Passed
Estimated Access Times
Min: 6.804ms Max: 27.872ms Avg: 17.908ms
PIO/DMA Data Compare
Total Bytes = 3346432
Estimated PIO Transfer Rates
Min: 2.97MB/s Max: 3.07MB/s Avg: 3.03MB/s
Estimated DMA Transfer Rates
Min: 74.34MB/s Max: 99.12MB/s Avg: 74.56MB/s
SMART Check: Passed
Short Test Passed: 20-Apr-2004 07:01:16

Scanning sectors...

Performing Complete Surface Scan on drive: ST380011A
Scan Completed... NO PROBLEMS FOUND



--
Tumppi
Reply to group
=================================================
Most learned here on nntp://news.mircosoft.com
Helsinki, Finland (remove NO _SPAM)
(translations from FI/SE not always accurate)
=================================================



--
Tumppi
Reply to group
=================================================
Most learned here on nntp://news.mircosoft.com
Helsinki, Finland (remove NO _SPAM)
(translations from FI/SE not always accurate)
=================================================
 
P

Paul

"Thomas Wendell" said:
My PC has problems when using 2nd HD. It works OK under WinXPpro (I think,
haven't stress tested), but I want to use it as a backup disk with
DriveImage 2002 and here the problems start (also under Linux). When making
a image, I naturally have to reboot to let DI load the DOS-version to backup
the C- and E-drive (E because my TEMP-env variable is set to E:\TEMP). Ok,
so far everything works, but about 15-20 min into the imagining my PC starts
to sound like an emergency vehicle (you know, uu-aa-uu-aa-uu-aa-uu at least
in Finland). It continues on, but that sound is NOT something I want to
listen to nighttime (when making my image). I'm just asking, any ideas??

(I can't check the temp when doing this, but if I exit DI and boot back to
XP, DriveHealth shows it as 42degrees C (about 105F??).
Under winXP it usually is at 39-41 degrees.
Cpu runs at 57degC, system at 41 and 1st HD at 40-41degrees, so I don't
think it's overheating). It is in a removable cradle with its own fan.
It's running with UDMA5, as my mobo can't(?) do more than that.

And the sound is definitely coming from the system, I guess it's the
S.M.A.R.T. monitoring the BIOS is doing.

The drive in question is a Seagate ST380011A 80GB disk, new as of 1 week (it
started straight away) ago
(exchanged on warranty for this problem)
HD0 is an internal Samsung SV1203N 120GB


System is: (HW) ASUS A7V133-C m/b (BIOS 1009), Athlon (T-bird) 1.2 GHz(FSB
100MHz), 384MB (SDRAM @133MHZ), 120+80GB HDs(primary IDE), DVD and
CD-RW(sec. IDE), GF2MX (32MB), Creative SB Live!Value, D-Link
NetNic(DRN-32TX),235W power supply (I know, I know, too puny), ADSL=ZyXEL
645MP-A1
OS=Win98SE_fi, all updates (incl. IE6SP1)



Any help would be appreciated, as I really don't like the sound of an
emergency vehicle when I'm trying to sleep....
(It's usually some 3 to 4 hours the imagining takes)


More info (and a guess)

I did run the test (SeaTools) both with SMART enabled or disabled in BIOS.
The test goes through w/o errors (as seen from log), but it also "generates"
this same sound after about the same time (~20min). Now I've started
thinking (with what?) :

SMART (or DriveHealth) is reporting errors in Raw Read Error Rate, Seek
Error Rate and Hardware ECC Recovered
But has these HAVE to be from the disk itself? Is it possible for the
chipset southbridge to be about to go south ( :) )? (Chipset VIA KT133A, NB
8363A
and SB 686B).Mobo was bought on 07.01.2002 (ddmmyyyy).

But why don't I get this same problem under Windows, just when accessing the
drive under Linux or DriveImage (I have no other means I can think of to
test this, I had the swap-file (pagefile.sys) on it, but no error). I'm
about to move one of my partitions from HDD-0 to HDD-1 to
test. Didn't matter..

(No, I can NOT afford a new PC or mobo/processor/memory-combination, and I'm
not wild about the idea of changing only the MB. It'd be outmoded when I
bought it. I'm unemployed)

Still looking for ideas/solutions
Below is output from SeaTools.exe

<<seatools output snipped>>

First off, I think the "police siren" is a temperature or voltage
warning. Get a copy of Motherboard Monitor (MBM5)

http://mbm.livewiredev.com/download.html
http://mbm.livewiredev.com/mobolist.html

MBM5 has an option to record the temperatures, fan speeds, and
voltages to a text file, roughly every ten seconds. When the
"police siren" sounds, let MBM5 collect data for another
minute or so, before stopping whatever other test you are running.
MBM5 runs in the background, so even if your rendering program
is running, you can still be collecting motherboard data at
the same time.

For a CPU load test, get a copy of Prime95 and use the "Torture Test"
option. This will load the CPU and allow you to do some thermal
testing. Since Prime95 doesn't take over the Windows desktop, you
will be able to look at the MBM5 monitor window while the Prime95
test is running. You should be able to see which sensor is setting
off the alarm by watching MBM5.

http://www.mersenne.org/freesoft.htm (Prime95)

Chances are, there is something wrong with the cooling of your
machine. For example, if the room temperature is 22C and the case
air temperature (termed the motherboard temperature) is 40C or
higher, then the computer case needs more ventilation. A quick
check would be, if removing the side of the computer makes the
CPU temperature come down, then the case needs more fans.

The Athlon 1.2 is one hot chip. To cool it, a heatsink is touching
just the small die of the processor, and so it is hard for
the heat to get out and use the fins of the heatsink effectively.
A well designed heatsink will have a copper core, as copper
spreads the heat from the CPU into the surrounding aluminum.
An all-copper heatsink is another option - these are heavier
and if you move the computer around a lot, an all-copper heatsink
can break the clips if the computer receives a mechanical shock.

So, start with some simple CPU load tests and monitor the
temperatures. Either the temp is rising too high, or perhaps
the Vcore voltage is dropping too low. With MBM5, you will soon
know what is the problem.

As for disk drives, it doesn't pay to run them in a computer case
where the air temperature is high. Modern IDE drives cannot take
high temperatures the way that some of the old Seagate 35Watt
SCSI drives could. If you cannot bring the case temperature down,
expect a premature drive failure when summer heats up your
computer room (do frequent backups!).

Also on the topic of disks, search in Google for "686B Bug",
including the double quotes. There were some issues with the
Via chipset and disk transfers, and you may want to read up on
the background information on the issue, in case you have
some future disk problems. (I don't remember all the details,
of what was fixed by 4in1 and what was fixed in the BIOS...)

http://groups.google.com/groups?hl=...m=120420011245473311%[email protected]

For A7V motherboards, you can also consult
www.a7vtroubleshooting.com , as that site specializes in
A7Vxxx issues.

HTH,
Paul
 
T

Thomas Wendell

Well, prime95 has been running for 3h now (with chassis side open) , and
I've seen no changes.
Ie. CPU runs at 57-58C, MB(sys)at 42-43C

Now, I opened OE and AsusProbe says Fan1 dropped to below 600rpm (set as
warning level by me)(look like~0rpm) but it didn.t stop, I looked
And +5V spiked to +6.854V

These it started doing abot about once every 2min???

AsusProbe I just d/l'd for this and it's probe22206 from Asus homepage.
MBM5 haven't had anything to say.....

Any more ideas??


--
Tumppi
Reply to group
=================================================
Most learned here on nntp://news.mircosoft.com
Helsinki, Finland (remove NO _SPAM)
(translations from FI/SE not always accurate)
=================================================
 
P

Paul

"Thomas Wendell" said:
Well, prime95 has been running for 3h now (with chassis side open) , and
I've seen no changes.
Ie. CPU runs at 57-58C, MB(sys)at 42-43C

Now, I opened OE and AsusProbe says Fan1 dropped to below 600rpm (set as
warning level by me)(look like~0rpm) but it didn.t stop, I looked
And +5V spiked to +6.854V

These it started doing abot about once every 2min???

AsusProbe I just d/l'd for this and it's probe22206 from Asus homepage.
MBM5 haven't had anything to say.....

Any more ideas??

You should only run one monitoring program at a time.
The authors of MBM5, have a semaphore set up, to organize
accesses to the SMBUS and the hardware monitor chip. Asus
Probe doesn't use the semaphore and if two programs attempt
to do a read on the SMBUS at the same time, the data
gets corrupted. Only install one of the two programs at
any one time. Maybe even the BIOS code occasionally
accesses the SMBUS ?

A motherboard temperature of 42C is high, and is not a good
environment for your disk drives. Is there anything you can do
to improve the flow of air through the computer case ?

Using MBM5, try to look at all the measured parameters, to
see if there is something different when the alarm goes
off after 20 minutes. There must be something different.

The fact that it takes 20 minutes, implies to me that something
is heating up, and fails to work properly at that point. The
trick will be figuring out exactly what makes the "police
siren".

Paul
 
T

Thomas Wendell

I can't run MBM when doing DI-backup (Drive Image 2002), as it runs in its
own DOS environment.


Short of changing chassis, there's not much I can do. I'm not very handy, so
chassis mod is not a choice. There's no places for aux fan on this chassis
(came with a 400Celeron/440BX motherboard somewhere around spring -98).

Guess I have to "run across" some spare money (~65EUR) for a new chassis
(incl. 400W power) and spare fan

(Unemployed means abot 400EUR/month, from that subtract rent/260EUR and
there's not much to play with)..


--
Tumppi
Reply to group
=================================================
Most learned here on nntp://news.mircosoft.com
Helsinki, Finland (remove _NOSPAM)
(translations from FI/SE not always accurate)
=================================================
 
P

Paul

"Thomas Wendell" said:
I can't run MBM when doing DI-backup (Drive Image 2002), as it runs in its
own DOS environment.


Short of changing chassis, there's not much I can do. I'm not very handy, so
chassis mod is not a choice. There's no places for aux fan on this chassis
(came with a 400Celeron/440BX motherboard somewhere around spring -98).

Guess I have to "run across" some spare money (~65EUR) for a new chassis
(incl. 400W power) and spare fan

(Unemployed means abot 400EUR/month, from that subtract rent/260EUR and
there's not much to play with)..

Options at this point are:

1) Continue testing with specific test programs.
The purpose of the testing is to try to find what is
causing the alarm, so you know what to fix.

2) Rearrange the fans in the case, for better cooling.
Try to improve on your 41C case temperature.

Do you have a web link / URL that describes your computer case ?
What is the brand name and model number of the computer case ?
(There is nothing wrong with a 235W power supply, if it puts the
power where it is needed. My measurements of total system power
show that 100 to 150W is being drawn, at full CPU load, for
typical simple computer configurations. Your system might be
closer to the 100W end of the spectrum.)

I would run Motherboard Monitor and Prime95, and see if Prime95
causes the police siren sound. Run whatever tests you can find
in Windows, so you can monitor the computer with Motherboard
Monitor, as that may indicate what the problem is.

To test the disk, is there some way you can do a surface
(read only) scan, as a means of exercising the disk ? I don't
believe there is any value in doing "seek tests" and a
continuous read of the disk surface may trigger the
alarm sound. Maybe it is the disk case temperature that is
causing S.M.A.R.T. to trigger ? Maybe something like
"scandisk" or "chkdisk" can be run ?

I don't think there is any reason to spend money yet, until
you are certain of the problem.

Is there some way the new disk can be supported outside the
computer case ? For example, if your computer case has two
openings for floppy drives, you could punch out the spare
floppy drive opening and run an IDE cable and drive power
cable outside the case. Try running the new drive outside
the case, and see if the problem disappears. (Note - try to
operate the disk on one of its major axis. A drive can be
operated flat on its belly, it can be operated on edge,
but should be exactly at 0 degrees or 90 degrees, so balance
the forces on the disk.) Make sure the disk is held securely,
so no accidents happen!

Another way to test for temperature issues, is to remove the
side of the computer case. See if the measured case temperature
drops. Then try repeating the Drive Image test, to see
if the alarm is raised or not.

Ah! I see a fan mounted on the Northbridge. I bet the
fan has stopped spinning. Have a look at the fan and
see if it has stopped running. Take the fan to a used
computer store and see if they have one that matches the
size and design of the current fan. The picture in the
manual shows some screws holding the fan in place, so
maybe it won't be too difficult to remove. If there is
some other attachment method, post back with the details,
and maybe someone can suggest an easy way to get the fan
off the Northbridge. Sometimes all it needs is to be
cleaned.

Paul
 
T

Thomas Wendell

Case temp dropped about 2C when run w/o sidepanel, CPU temp didn't drop

The 2nd HD is in a 51/2" bay, in its own removable caddy equipped with its
own small fan
All IDE connections are in use (and I'm not going to bui a separate
controller board)

Northbridge fan works as normal, as have the others (CPU,PS)

Haven't got a clue as to the maker and model of case, as it's originally an
Octek 400Celeron built by BT-mikro in Helsinki

When I have the time and inclination with nothing else to do(sometime after
midsummer, I guess), I'm going to (temporarily) change disk configuration
and use the 80G as boot and system disk for a while (I know I have to
install Windows again for that, but so what?).

I'm going to get the new case anyway, I have use for it when I build a new
computer anyway.
Case is PowerMAX CP0327PL ( http://www.powmax.com/CASE_GENREAL_PAGE.htm) and
can be had for 40EUR here now




(Economic discussions with father underway)

Thinking about (budget around 650EUR)
ASRock K7S8XE+ about 55EUR
Athlon XP2600 boxed 97EUR
512MB PC2700 95EUR (now, changing all the
time)
Samsung V80 SP1604N 160GB ~100EUR
Radeon 9200 70EUR
LG GSA-4082 DVD-RW 109EUR





--
Tumppi
Reply to group
=================================================
Most learned here on nntp://news.mircosoft.com
Helsinki, Finland (remove _NOSPAM)
(translations from FI/SE not always accurate)
=================================================
 
P

Paul

"Thomas Wendell" said:
Case temp dropped about 2C when run w/o sidepanel, CPU temp didn't drop

The 2nd HD is in a 51/2" bay, in its own removable caddy equipped with its
own small fan
All IDE connections are in use (and I'm not going to bui a separate
controller board)

Northbridge fan works as normal, as have the others (CPU,PS)

Haven't got a clue as to the maker and model of case, as it's originally an
Octek 400Celeron built by BT-mikro in Helsinki

When I have the time and inclination with nothing else to do(sometime after
midsummer, I guess), I'm going to (temporarily) change disk configuration
and use the 80G as boot and system disk for a while (I know I have to
install Windows again for that, but so what?).

I'm going to get the new case anyway, I have use for it when I build a new
computer anyway.
Case is PowerMAX CP0327PL ( http://www.powmax.com/CASE_GENREAL_PAGE.htm) and
can be had for 40EUR here now

(Economic discussions with father underway)

Thinking about (budget around 650EUR)
ASRock K7S8XE+ about 55EUR
Athlon XP2600 boxed 97EUR
512MB PC2700 95EUR (now, changing all the
time)
Samsung V80 SP1604N 160GB ~100EUR
Radeon 9200 70EUR
LG GSA-4082 DVD-RW 109EUR

http://www.asrock.com.tw/Drivers/Manual/K7S8XE+_UM.pdf

The new case has room for some 80mm case fans. I suggest you include
money in the budget for two 80mm case fans. The fan in the PSU is
not sufficient to cool a computer. The PSU fan is temperature controlled
in many cases, and the temperature setting of the PSU fan is too
high to maintain a safe case temperature. (The disk drives should be
kept as cool as possible, whereas everything else in the computer can
take more heat.) That is the reason for adding at least one more 80mm
case fan, and having two on hand is convenient for experimenting with
placement. You can observe the results using the monitor chip, assuming
the Asrock board has a monitor chip like an Asus board would have.

The 400W power supply listed on that web page looks to have a
good mix of power output levels. I wouldn't expect trouble
with that.

When it comes to DVD devices, visit cdfreaks.com , as they
have review articles and a forum, and you can learn a lot
about what drives are good and what drives are bad.

http://www.cdfreaks.com/article/search/B

Discussions about CD/DVD drives are here:

http://club.cdfreaks.com/

I bought a Liteon DVDRW as a Christmas gift for someone,
and it turned out to be just as picky about media, as the
CDfreaks review said it would be. Had to buy a few packs
of disks until I found some that had low error rates.
To use DVDRW well, costs a lot of money for good software
and packs of disks. If you are on a budget, CDRW media is
a cheaper option.

Have fun,
Paul
 
T

Thomas Wendell

Removed cartridge with 80G HD in it.
MBM5 now reports mb temp at 40C. CPU is at 52C
Temps dropper 2-3C

And this cartridge has its own internal fan????



--
Tumppi
Reply to group
=================================================
Most learned here on nntp://news.mircosoft.com
Helsinki, Finland (remove _NOSPAM)
(translations from FI/SE not always accurate)
=================================================
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top