You are not logged in.

#1 2005-05-11 19:48:32

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

memtest

This program testes the ram memory right?
Will this program return wich ram stick that has errors or just that someone has?
and, for me it seems to take ages and seems to doing the same thing over and over again, is it supposed to or is it just redoing?

Offline

#2 2005-05-11 23:38:37

Kern
Member
From: UK
Registered: 2005-02-09
Posts: 464

Re: memtest

This program testes the ram memory right?

Yes  .... but not exclusively.

Will this program return wich ram stick that has errors or just that someone has?

Not necessarily either of those. its more involved than that. CPU / L1 L2 cache etc.

it seems to take ages and seems to doing the same thing over and over again,

memtest86 executes indefinitely.
......

all  this assumes you are using the memtest86 package from the Arch repo :-

pacman -Ss memtest
extra/memtest86+ 1.55.1-1
    An advanced memory diagnostic tool

if so, heres a good explanation of its workings and general howto on troubleshooting memory errors.
http://www.memtest86.com/

Offline

#3 2005-05-12 09:51:25

FUBAR
Member
From: Belgium
Registered: 2004-12-08
Posts: 1,029
Website

Re: memtest

I don't think it can test a memory stick: the RAM it tests is non-permanent, ie. when you switch off your computer the contents are gone. The memory on a memory stick is permanent, the contents are gone when you delete them. If you're having trouble with your memory stick, try repartitioning it. This is a data destructive procedure!. I had trouble with my memory stick: I couldn't mount it so I couldn't access the data anyway.


A bus station is where a bus stops.
A train station is where a train stops.
On my desk I have a workstation.

Offline

#4 2005-05-12 15:41:12

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

Uhum, I have memtest+ ( = memtest86?)
I need to test my ram coz i think that it is faulty (from time to time the computer stops and a beep that should take 1 sec takes seven sec)
memtest never gave me an error, does this mean that the ram is ok?

Offline

#5 2005-05-12 17:08:08

i3839
Member
Registered: 2004-02-04
Posts: 1,185

Re: memtest

Yes, most likely it is. It may be a heat problem, did you overclock anything? Check your cpu temperature in the bios if possible, and also check the coolers. But if your pc never crashes then there's no big problem, is the extra beeping only at startup?

Offline

#6 2005-05-12 17:17:18

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

i3839 wrote:

Yes, most likely it is. It may be a heat problem, did you overclock anything? Check your cpu temperature in the bios if possible, and also check the coolers. But if your pc never crashes then there's no big problem, is the extra beeping only at startup?

The beep comes when the computer gets very slow and i restart it with ctrl+alt+del (it is supposed to give a short beep) and it gives a very long beep and takes long time to shutdown.
No I havent overclocked anything and I cant beleve it is a heating problem becouse it happens sometimes some minutes after the computer has started, and it reeboots opk aftera 'crach'

It crashes about twice a day, sometimes moore.

Offline

#7 2005-05-12 17:23:23

Kern
Member
From: UK
Registered: 2005-02-09
Posts: 464

Re: memtest

If you're not sure its Ram maybe also try a few other diagnostic tools or stress tests, checkout  www.ultimatebootcd.com.

tests are best run outside your host operating system, so the bootable CD is good idea for isolating a problem.

Might be an idea to check the ram is seated correctly, the CPU is getting cooled sufficiently as i3839 suggested, as heat could be a cause. Bad contacts also give sporadic problems, as might a dodgy IDE cable if you swap h/d's a lot.I

EDIT

cant beleve it is a heating problem becouse it happens sometimes some minutes

depends what the processor is doing. it could happen quicker than you think.

Offline

#8 2005-05-12 17:27:11

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

Might be an idea to check the ram is seated correctly

What do you mean?

I will check the bootcd, what tools are recomended to use on the cd?

Offline

#9 2005-05-12 17:34:28

Kern
Member
From: UK
Registered: 2005-02-09
Posts: 464

Re: memtest

What do you mean?

(seated)

RAM modules usually sit in a long thin slot, with a white tab at either end. these should be open, you then fit the ram into the slot and the white tags close in slightly. sometimes folks forget to then pinch the white tags in to sit in the recesses made for them. its a locking action.

As for which tools to use, try System Burn in / CPU test / memory tests and possiby a hard disk test.

Offline

#10 2005-05-12 18:38:49

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

uh, i did some cpu burn , said it would take seven days...
well il do it this way, il take out one ramstick a time and if its not crashing i know wichone is wrong.
all ramsticks where with the white flaps on.

Offline

#11 2005-05-12 18:47:33

i3839
Member
Registered: 2004-02-04
Posts: 1,185

Re: memtest

I don't think it's a ram problem, memtest would have found it most likely. You don't have to run the cpuburn a whole week, just until it crashes (one hour should be plenty). Nothing funny in dmesg or logs after a crash?

Offline

#12 2005-05-14 06:35:24

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

i3839 wrote:

I don't think it's a ram problem, memtest would have found it most likely. You don't have to run the cpuburn a whole week, just until it crashes (one hour should be plenty). Nothing funny in dmesg or logs after a crash?

Im sorry, where is the logs?
Im quite sure its not the ram. becouse I took out one ramstick a time and the computer crashed with all combinations (um, i mean like ram 1,2 & 1,3 & 3,2)

Offline

#13 2005-05-14 07:01:46

Snowman
Developer/Forum Fellow
From: Montreal, Canada
Registered: 2004-08-20
Posts: 5,212

Re: memtest

The logs are in /var/log

Offline

#14 2005-05-14 07:08:04

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

yay, i have like auth.log.1 to auth.log.4 same with errors,daemon,kernel,messages etc
wichone do I want to look in? kernel.log or kernel.log.4?

Offline

#15 2005-05-14 09:48:05

T-Dawg
Forum Fellow
From: Charlotte, NC
Registered: 2005-01-29
Posts: 2,736

Re: memtest

auth.log
authlog.4 is the oldest and next in line for logrotate to delete.

Offline

#16 2005-05-14 11:51:01

i3839
Member
Registered: 2004-02-04
Posts: 1,185

Re: memtest

Look at /var/log/kernel. Though if it's a real hard crash then either no info was generated or the data didn't made it to disk. But perhaps there is something interesting earlier on.

Since when do you have these problems? What changed?

Offline

#17 2005-05-14 16:47:50

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

i3839 wrote:

Look at /var/log/kernel. Though if it's a real hard crash then either no info was generated or the data didn't made it to disk. But perhaps there is something interesting earlier on.

Since when do you have these problems? What changed?

I have had theese problems since I installed arch, never happened in fedora and happened once on the ubuntu live cd.

Im not sure if its a crash, the computer works, but VERY VERY slow (In the time of a beep i am able to go to the sofa , sit down and still hear the beep)

Offline

#18 2005-05-14 17:39:34

i3839
Member
Registered: 2004-02-04
Posts: 1,185

Re: memtest

Weird, what cpu and other hardware do you have and do you have a temperature sensor in your pc? If so, look in the bios to see what the temperature is, or install and setup lm_sensors. I suspect a cpu cooler that gets stuck once in a while or something, causing the slowness. It may also be a broken harddisk, but then you'll see errors in the logs for sure.

Look in your logs and at dmesg output, if there's nothing interesting there try booting with acpi=off to see if that makes a difference. What kernel are you running? If something before 2.6.11 then upgrade. Also posting the dmesg output just when such sloweness begins would be very interesting.

Offline

#19 2005-05-14 20:02:37

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

i3839 wrote:

Weird, what cpu and other hardware do you have and do you have a temperature sensor in your pc? If so, look in the bios to see what the temperature is, or install and setup lm_sensors. I suspect a cpu cooler that gets stuck once in a while or something, causing the slowness. It may also be a broken harddisk, but then you'll see errors in the logs for sure.

Look in your logs and at dmesg output, if there's nothing interesting there try booting with acpi=off to see if that makes a difference. What kernel are you running? If something before 2.6.11 then upgrade. Also posting the dmesg output just when such sloweness begins would be very interesting.

[firedance@NetBox ~]$ uname -r
2.6.11.7-ARCH

but, if it was overheating wouldnt the slowness come slowly?
the computer goes slow like Poff!

[root@NetBox firedance]# pacman -S lm_sensors
:: lm_sensors-2.9.1-1: is up to date.  Upgrade anyway? [Y/n] n

I haave lm_sensors, just dont know howto use it, il check with google.

Offline

#20 2005-05-14 20:28:12

i3839
Member
Registered: 2004-02-04
Posts: 1,185

Re: memtest

If you have e.g. a P4 and it overheats then the cpu will throttle itself down so it's very slow, to protect burning like the old cpu's did when overheating. That's why I asked what cpu you have, if you have an athlon then it probably won't slow down, but either burn or reboot. It isn't the heat that causes the slowdown, but the cpu detecting the overheating and then slowing tiself down to prefent burning. That is, if this is the cause at all.

To get lm_sensors working run and follow the instructions of sensors-detect. After that customize your /etc/sensors.conf if needed (if get get very weird results then change some thing until it gives reasonable output. For more info read the stuff in the config file).

But for now the output of the logfiles and dmesg are much more interesting.

Offline

#21 2005-05-18 17:02:58

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

ok, i havent been able to get the sensors working and i dont know what cpu I have.

my computer 'crashed' for 10 minutes ago this is what happened when i tried rebooting

first it crashed, i rebooted and it stopped at hotplug , rebooted three times and this happened.
after that i stopped at the partition check and dit fsck -y ; reboot  after that it said 'ARCH has gone 49709 days without beeing checked' <- thats a lot of days!

I copied all logs to another folder, wich log is theone i want to look in?
--edit-
crashed again, yay, maby it really is overheating. thistime i did fsck -y directly and it worked fine, saved logs into diffrent folder so now I maby could do diff or something.

Offline

#22 2005-05-18 20:23:18

i3839
Member
Registered: 2004-02-04
Posts: 1,185

Re: memtest

To see what cpu you have, do cat /proc/cpuinfo. The logfile needed is /var/log/kernel.

Offline

#23 2005-05-19 12:42:00

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

[firedance@NetBox ~]$ cat /proc/cpuinfo
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 2
model name      : Intel(R) Pentium(R) 4 CPU 3.20GHz
stepping        : 9
cpu MHz         : 3192.351
cache size      : 512 KB
physical id     : 0
siblings        : 2
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe cid xtpr
bogomips        : 6324.22


The kernel.log, what to look for, theres alot of info. I found this, maby it is what im looking for.

May  8 05:43:25 NetBox ACPI: PCI Interrupt Routing Table [_SB_.PCI0._PRT]
May  8 05:43:25 NetBox ACPI: PCI Interrupt Routing Table [_SB_.PCI0.SLOT._PRT]
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKA] (IRQs 3 4 5 7 9 10 *11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKB] (IRQs 3 4 5 7 *9 10 11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKC] (IRQs 3 4 *5 7 9 10 11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKD] (IRQs 3 4 5 7 9 *10 11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKE] (IRQs 3 4 5 7 9 10 *11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKF] (IRQs *3 4 5 7 9 10 11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKG] (IRQs 3 4 5 7 *9 10 11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Link [LNKH] (IRQs 3 4 5 7 *9 10 11 12 14 15)
May  8 05:43:25 NetBox ACPI: PCI Interrupt Routing Table [_SB_.PCI0.AGP_._PRT]

but i have no idea... I gotta set the system clock right so that I can see the right time in the log, last time i set it wiuth date but it seems to change each time I reboot

Offline

#24 2005-05-19 13:19:27

i3839
Member
Registered: 2004-02-04
Posts: 1,185

Re: memtest

Try looking for something with "warning" or "error" in front of it instead, not the ACPI routing table.

Offline

#25 2005-05-19 15:15:32

firedance
Member
From: Stockholm, Sweden
Registered: 2005-04-18
Posts: 131

Re: memtest

[firedance@NetBox log]$ cat kernel.log | grep error
May  8 00:05:09 NetBox NTFS-fs error (device sda2): ntfs_ucstonls(): Unicode name contains characters that cannot be converted to character set iso8859-1.
[firedance@NetBox log]$ cat kernel.log | grep warning
[firedance@NetBox log]$ cat kernel.log.1 | grep warning
[firedance@NetBox log]$ cat kernel.log.1 | grep error 
May  2 00:05:03 NetBox NTFS-fs error (device sda2): ntfs_ucstonls(): Unicode name contains characters that cannot be converted to character set iso8859-1.
May  3 18:58:20 NetBox usb 4-2: string descriptor 0 read error: -19
May  3 18:58:20 NetBox usb 4-2: string descriptor 0 read error: -19
May  3 18:58:20 NetBox usb 4-2: string descriptor 0 read error: -19
May  5 12:11:50 NetBox usbhid: probe of 1-2:1.0 failed with error -5
May  5 12:11:50 NetBox usbhid: probe of 1-2:1.1 failed with error -5
May  7 13:50:04 NetBox usbhid: probe of 1-2:1.0 failed with error -5
May  7 13:50:04 NetBox usbhid: probe of 1-2:1.1 failed with error -5
[firedance@NetBox log]$

does not seem to be any big errors.

Offline

Board footer

Powered by FluxBB