Hello!
A few days ago in my old Debian box, my computer randomly locked up
twice and rebooted once. Once, I got system failure during POST. I
turned off and on the machine. It didn't boot up (no POST). Turned off
and turned on again, then computer worked for another 24 hours. Then
after that, I got no POST (all fans inside computer worked though), no
video signals, no beeps, no keyboard lights, etc. Clearing CMOS,
changing video card, etc. did not help.
What helped was removing, swapping, testing, etc. the two installed 512
MB of Kingston RAM modules (together and separately). My friend and I
found the one module was going/went bad. I never had problem with this
for the last 4-5 years (bought it new). Since it is lifetime warranty,
it was sent in for a replacement.
I am curious. How often do old RAM (machine runs almost 24/7) go bad
like this? I haven't had any more lockups, reboots, and other weirndness
with the other 512 MB of memory. I didn't see any
burns/melted/discolored areas and smelled any funny odor on the bad
memory. The temperatures have been cool [under 70 degrees(F)] due to
winter time (lots of rain lately though).
Thank you in advance. :)
--
"For while the giants have just been talking about an information
superhighway, the ants have actually been building one: the Internet."
From "The Accidental Superhighway." The Economist: A Survey of the
Internet, 1-7 July 1995, insert.
/\___/\
/ /\ /\ \ Phil/Ant @ http://antfarm.home.dhs.org (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Remove ANT from e-mail address: philpi@earthlink.netANT
( ) or ANTant@zimage.com
Ant is currently not listening to any songs on his home computer.
A few days ago in my old Debian box, my computer randomly locked up
twice and rebooted once. Once, I got system failure during POST. I
turned off and on the machine. It didn't boot up (no POST). Turned off
and turned on again, then computer worked for another 24 hours. Then
after that, I got no POST (all fans inside computer worked though), no
video signals, no beeps, no keyboard lights, etc. Clearing CMOS,
changing video card, etc. did not help.
What helped was removing, swapping, testing, etc. the two installed 512
MB of Kingston RAM modules (together and separately). My friend and I
found the one module was going/went bad. I never had problem with this
for the last 4-5 years (bought it new). Since it is lifetime warranty,
it was sent in for a replacement.
I am curious. How often do old RAM (machine runs almost 24/7) go bad
like this? I haven't had any more lockups, reboots, and other weirndness
with the other 512 MB of memory. I didn't see any
burns/melted/discolored areas and smelled any funny odor on the bad
memory. The temperatures have been cool [under 70 degrees(F)] due to
winter time (lots of rain lately though).
Thank you in advance. :)
--
"For while the giants have just been talking about an information
superhighway, the ants have actually been building one: the Internet."
From "The Accidental Superhighway." The Economist: A Survey of the
Internet, 1-7 July 1995, insert.
/\___/\
/ /\ /\ \ Phil/Ant @ http://antfarm.home.dhs.org (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Remove ANT from e-mail address: philpi@earthlink.netANT
( ) or ANTant@zimage.com
Ant is currently not listening to any songs on his home computer.
wrote:
we now have tin wiskers growing on everything. I don't
think your modules are post-ROHS though, so I would instead
suspect the motherboard is failing and subjecting them to
excessive voltage, OR there was a power surge that caused
damage.
There's not much point in wondering about the 1 time in a
million things didn't work out as expected. Literally,
there are a million other systems out there that didn't have
this problem and it could be anything from a DRAM defect to
a power surge to a mild damage from a ESD long ago. Unless
we can find what actually caused the fault, there is nothing
we can do to prevent them except continue following best
practices.
If I had to play odds, I'd say the capacitors in either the
PSU or motherboard are failing, but you didn't clarify what
testing means, if you took the questionable module and then
tried it in another system (and if an older system, knowing
that it has clean memory contacts and likewise it's
capacitors are not aged onto the point of degraded
performance).
Sadly, nothing is made to last forever. Some failures are
sooner than others.
we now have tin wiskers growing on everything. I don't
think your modules are post-ROHS though, so I would instead
suspect the motherboard is failing and subjecting them to
excessive voltage, OR there was a power surge that caused
damage.
There's not much point in wondering about the 1 time in a
million things didn't work out as expected. Literally,
there are a million other systems out there that didn't have
this problem and it could be anything from a DRAM defect to
a power surge to a mild damage from a ESD long ago. Unless
we can find what actually caused the fault, there is nothing
we can do to prevent them except continue following best
practices.
If I had to play odds, I'd say the capacitors in either the
PSU or motherboard are failing, but you didn't clarify what
testing means, if you took the questionable module and then
tried it in another system (and if an older system, knowing
that it has clean memory contacts and likewise it's
capacitors are not aged onto the point of degraded
performance).
Sadly, nothing is made to last forever. Some failures are
sooner than others.
On 1/30/2008 12:37 AM PT, kony typed:
What's ESD? Power surge. Hmm. Wouldn't my APC Back-UPS XS BX1500
prevented those small ones? I know these consumer based ones can't
handle big ones. My other more powerful machine and 19" LCD monitor are
on this UPS too. Here are the only power hiccups my Debian box recorded
recently via USB (only posted the power issue errors):
# cat /var/log/apcupsd.events
Tue Aug 28 01:30:19 PDT 2007 apcupsd 3.12.4 (19 August 2006) debian
startup succeeded
Sun Sep 02 21:46:55 PDT 2007 Power failure.
Sun Sep 02 21:47:01 PDT 2007 Running on UPS batteries.
Sun Sep 02 21:47:13 PDT 2007 Mains returned. No longer on UPS batteries.
Sun Sep 02 21:47:13 PDT 2007 Power is back. UPS running on mains.
Sun Sep 09 10:41:36 PDT 2007 Power failure.
Sun Sep 09 10:41:38 PDT 2007 Power is back. UPS running on mains.
Mon Sep 10 07:25:54 PDT 2007 Power failure.
Mon Sep 10 07:25:58 PDT 2007 Power is back. UPS running on mains.
Sat Sep 22 18:09:25 PDT 2007 Power failure.
Sat Sep 22 18:09:27 PDT 2007 Power is back. UPS running on mains.
Thu Sep 27 22:24:19 PDT 2007 Power failure.
Thu Sep 27 22:24:22 PDT 2007 Power is back. UPS running on mains.
Thu Oct 11 10:46:29 PDT 2007 Power failure.
Thu Oct 11 10:46:32 PDT 2007 Power is back. UPS running on mains.
Sun Oct 21 08:23:46 PDT 2007 Power failure.
Sun Oct 21 08:23:49 PDT 2007 Power is back. UPS running on mains.
Sun Oct 28 08:42:14 PDT 2007 Power failure.
Sun Oct 28 08:42:17 PDT 2007 Power is back. UPS running on mains.
Wed Nov 07 06:25:56 PST 2007 Power failure.
Wed Nov 07 06:25:59 PST 2007 Power is back. UPS running on mains.
Fri Nov 09 06:25:57 PST 2007 Power failure.
Fri Nov 09 06:26:00 PST 2007 Power is back. UPS running on mains.
Sat Nov 10 07:12:11 PST 2007 Power failure.
Sat Nov 10 07:12:14 PST 2007 Power is back. UPS running on mains.
Thu Nov 15 06:25:57 PST 2007 Power failure.
Thu Nov 15 06:26:00 PST 2007 Power is back. UPS running on mains.
Sun Dec 30 09:16:56 PST 2007 Power failure.
Sun Dec 30 09:16:58 PST 2007 Power is back. UPS running on mains.
Tue Jan 01 08:28:12 PST 2008 Power failure.
Tue Jan 01 08:28:14 PST 2008 Power is back. UPS running on mains.
Sat Jan 05 07:41:30 PST 2008 Power failure.
Sat Jan 05 07:41:33 PST 2008 Power is back. UPS running on mains.
Sun Jan 06 17:03:28 PST 2008 Power failure.
Sun Jan 06 17:03:31 PST 2008 Power is back. UPS running on mains.
According to my computer hardware customization logs, I did have to
change my Debian's PSU on 5/14/2007 with a new Fortron FSP650-80GLC PSU
(650 watts) due to its old Antec PSU stopped working (computer no longer
booted up). I don't know if that was related to damage/kill my RAM. It
took this long to cause the symptoms if it is related.
We also tried different memory slots. We did not try another computer
since there wasn't one available. My friend also tried underclocking
speed and that still crashed. The memory module was already sent out for
RMA, so we can't test it on another machine if there was one.
True, but I wasn't expecting a non-movable component to die that fast. I
can understand movable parts like drives. My friend and I thought it was
a dead motherboard too. We saw nothing fried, no weird shaped
capitactors, no discolorations, no odd odors, etc.
--
"He who storms in like a whirlwind returns like an ant." --Borneo
/\___/\
/ /\ /\ \ Phil/Ant @ http://antfarm.home.dhs.org (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Remove ANT from e-mail address: philpi@earthlink.netANT
( ) or ANTant@zimage.com
Ant is currently not listening to any songs on his home computer.
What's ESD? Power surge. Hmm. Wouldn't my APC Back-UPS XS BX1500
prevented those small ones? I know these consumer based ones can't
handle big ones. My other more powerful machine and 19" LCD monitor are
on this UPS too. Here are the only power hiccups my Debian box recorded
recently via USB (only posted the power issue errors):
# cat /var/log/apcupsd.events
Tue Aug 28 01:30:19 PDT 2007 apcupsd 3.12.4 (19 August 2006) debian
startup succeeded
Sun Sep 02 21:46:55 PDT 2007 Power failure.
Sun Sep 02 21:47:01 PDT 2007 Running on UPS batteries.
Sun Sep 02 21:47:13 PDT 2007 Mains returned. No longer on UPS batteries.
Sun Sep 02 21:47:13 PDT 2007 Power is back. UPS running on mains.
Sun Sep 09 10:41:36 PDT 2007 Power failure.
Sun Sep 09 10:41:38 PDT 2007 Power is back. UPS running on mains.
Mon Sep 10 07:25:54 PDT 2007 Power failure.
Mon Sep 10 07:25:58 PDT 2007 Power is back. UPS running on mains.
Sat Sep 22 18:09:25 PDT 2007 Power failure.
Sat Sep 22 18:09:27 PDT 2007 Power is back. UPS running on mains.
Thu Sep 27 22:24:19 PDT 2007 Power failure.
Thu Sep 27 22:24:22 PDT 2007 Power is back. UPS running on mains.
Thu Oct 11 10:46:29 PDT 2007 Power failure.
Thu Oct 11 10:46:32 PDT 2007 Power is back. UPS running on mains.
Sun Oct 21 08:23:46 PDT 2007 Power failure.
Sun Oct 21 08:23:49 PDT 2007 Power is back. UPS running on mains.
Sun Oct 28 08:42:14 PDT 2007 Power failure.
Sun Oct 28 08:42:17 PDT 2007 Power is back. UPS running on mains.
Wed Nov 07 06:25:56 PST 2007 Power failure.
Wed Nov 07 06:25:59 PST 2007 Power is back. UPS running on mains.
Fri Nov 09 06:25:57 PST 2007 Power failure.
Fri Nov 09 06:26:00 PST 2007 Power is back. UPS running on mains.
Sat Nov 10 07:12:11 PST 2007 Power failure.
Sat Nov 10 07:12:14 PST 2007 Power is back. UPS running on mains.
Thu Nov 15 06:25:57 PST 2007 Power failure.
Thu Nov 15 06:26:00 PST 2007 Power is back. UPS running on mains.
Sun Dec 30 09:16:56 PST 2007 Power failure.
Sun Dec 30 09:16:58 PST 2007 Power is back. UPS running on mains.
Tue Jan 01 08:28:12 PST 2008 Power failure.
Tue Jan 01 08:28:14 PST 2008 Power is back. UPS running on mains.
Sat Jan 05 07:41:30 PST 2008 Power failure.
Sat Jan 05 07:41:33 PST 2008 Power is back. UPS running on mains.
Sun Jan 06 17:03:28 PST 2008 Power failure.
Sun Jan 06 17:03:31 PST 2008 Power is back. UPS running on mains.
According to my computer hardware customization logs, I did have to
change my Debian's PSU on 5/14/2007 with a new Fortron FSP650-80GLC PSU
(650 watts) due to its old Antec PSU stopped working (computer no longer
booted up). I don't know if that was related to damage/kill my RAM. It
took this long to cause the symptoms if it is related.
We also tried different memory slots. We did not try another computer
since there wasn't one available. My friend also tried underclocking
speed and that still crashed. The memory module was already sent out for
RMA, so we can't test it on another machine if there was one.
True, but I wasn't expecting a non-movable component to die that fast. I
can understand movable parts like drives. My friend and I thought it was
a dead motherboard too. We saw nothing fried, no weird shaped
capitactors, no discolorations, no odd odors, etc.
--
"He who storms in like a whirlwind returns like an ant." --Borneo
/\___/\
/ /\ /\ \ Phil/Ant @ http://antfarm.home.dhs.org (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Remove ANT from e-mail address: philpi@earthlink.netANT
( ) or ANTant@zimage.com
Ant is currently not listening to any songs on his home computer.
wrote:
Electrostatic discharge, would typically happen prior to
installing the memory or while installing.
In an ideal world yes, in the real world parts are built to
a budget and restrained by their reaction time and impedance
to earth ground. We'd have to suspect it was the cause to
make more time spent on this fruitful as memory generally
isn't as likely damaged by a surge as other parts, mainly at
least those through which the surge entered the system and
you report no other failures.
Either your power is flaky or you have set too high a
threshold in your APC software. (At least, IIRC, some of the
APC software I've seen allows the user to set the threshold
at which it goes from line to backup power). If you have an
AFPC PSU (no 110/220V switch on it), it is quite tolerant of
changing line voltage and you can set your UPS threshold
very broad.
It can't be direclty assumed, Antecs are not only popular
but certain models are known to fail prematurely due to bad
and/or marginal capacitors, but in those cases where they
did fail I dont' ever recall it damaging the memory. Modern
memory is behind at least one addt'l regulation stage on the
motherboard.
There's really not much point in being too concerned about a
single part failure. Random failures are just that, but if
you have more memory failing in the same board in the
future, I would think about retiring the motherboard.
Electrostatic discharge, would typically happen prior to
installing the memory or while installing.
In an ideal world yes, in the real world parts are built to
a budget and restrained by their reaction time and impedance
to earth ground. We'd have to suspect it was the cause to
make more time spent on this fruitful as memory generally
isn't as likely damaged by a surge as other parts, mainly at
least those through which the surge entered the system and
you report no other failures.
Either your power is flaky or you have set too high a
threshold in your APC software. (At least, IIRC, some of the
APC software I've seen allows the user to set the threshold
at which it goes from line to backup power). If you have an
AFPC PSU (no 110/220V switch on it), it is quite tolerant of
changing line voltage and you can set your UPS threshold
very broad.
It can't be direclty assumed, Antecs are not only popular
but certain models are known to fail prematurely due to bad
and/or marginal capacitors, but in those cases where they
did fail I dont' ever recall it damaging the memory. Modern
memory is behind at least one addt'l regulation stage on the
motherboard.
There's really not much point in being too concerned about a
single part failure. Random failures are just that, but if
you have more memory failing in the same board in the
future, I would think about retiring the motherboard.
Ah, if that was the case, then it would probably happen in the end of
2006. Or maybe May 2006 when replacing the PSU (doubt the memory was
even touched) by something else?
Hm.
Maybe. I do notice sometimes my lamp blink when this happens if it is
on. Probably a power issue. I live in an old house (built in the late
70s/early 80s). Power is done underground. I have had this UPS since
9/25/2005 according to logs on
http://alpha.zimage.com/~ant/antfarm/about/toys.html ... Before it, I
had an APC Back-UPS 650VA (BK650MC) that had to be replaced because it
kept shutting down even if there was AC power! That is when I replaced
it.
Hmm, I am not sure how to set this. I used the default configurations
from apcaccess package in Debian. I didn't set my OS to shut down the
computer or anything when battery is low. Maybe my current UPS
statistics help to see what's funky?
$ /sbin/apcaccess
APC : 001,037,0895
DATE : Wed Jan 30 17:43:16 PST 2008
HOSTNAME : FooBar
RELEASE : 3.14.2
VERSION : 3.14.2 (15 September 2007) debian
UPSNAME : ANTian
CABLE : USB Cable
MODEL : Back-UPS RS 1500
UPSMODE : Stand Alone
STARTTIME: Sat Jan 26 23:02:42 PST 2008
STATUS : ONLINE
LINEV : 117.0 Volts
LOADPCT : 11.0 Percent Load Capacity
BCHARGE : 100.0 Percent
TIMELEFT : 70.9 Minutes
MBATTCHG : 5 Percent
MINTIMEL : 3 Minutes
MAXTIME : 0 Seconds
SENSE : High
LOTRANS : 097.0 Volts
HITRANS : 138.0 Volts
ALARMDEL : Always
BATTV : 26.9 Volts
LASTXFER : Low line voltage
NUMXFERS : 0
TONBATT : 0 seconds
CUMONBATT: 0 seconds
XOFFBATT : N/A
SELFTEST : NO
STATFLAG : 0x07000008 Status Flag
MANDATE : 2005-03-16
SERIALNO : QB0512132444
BATTDATE : 2001-09-25
NOMINV : 120
NOMBATTV : 24.0
FIRMWARE : 8.g8 .D USB FW:g8
APCMODEL : Back-UPS RS 1500
END APC : Wed Jan 30 17:44:11 PST 2008
Hmm. OK.
OK. Will note. I hope Kingston doesn't say the RMA'ed RAM is fine and
sends it back. Again, didn't have another machine to test it on.
--
"I used to own an ant farm but had to give it up. I couldn't find
tractors small enough to fit it." --Steven Wright
/\___/\
/ /\ /\ \ Ant @ http://antfarm.home.dhs.org (Personal Web Site)
| |o o| | Ant's Quality Foraged Links (AQFL): http://aqfl.net
\ _ / Please remove ANT if replying by e-mail.
( )
This Thread
  Re: How often do RAM just die when used for severa...
| Timothy Daniels | 01-30-2008 |
Please Register and login to reply and use other advanced options
- Optical Mouse used on eMachine T5052 (actually a Gateway desktop)
- Microsoft Windows MediaCenter
- 2007-05-04
- USB 2.0 on Dell Dimension.
- Computer Hardware
- 2013-01-24
- Recommendations?
- Computer Hardware
- 2012-11-07
- An Advent Wierdo?
- Computer Hardware
- 2013-02-28
- Linksys wrt54g Router Problem
- Computer Hardware
- 2013-02-16
- Hard Drive slows down and stops
- Computer Hardware
- 2012-09-17
- Scanner drivers for Windows 7
- Computer Hardware
- 2012-08-20
- New Toshiba Satellite Laptop
- Computer Hardware
- 2012-07-04
- PSU as Car Stereo Power
- Computer Hardware
- 2012-09-26
XML Sitemap
>A few days ago in my old Debian box, my computer randomly locked up
>twice and rebooted once. Once, I got system failure during POST. I
>turned off and on the machine. It didn't boot up (no POST). Turned off
>and turned on again, then computer worked for another 24 hours. Then
>after that, I got no POST (all fans inside computer worked though), no
>video signals, no beeps, no keyboard lights, etc. Clearing CMOS,
>changing video card, etc. did not help.
>What helped was removing, swapping, testing, etc. the two installed 512
>MB of Kingston RAM modules (together and separately). My friend and I
>found the one module was going/went bad. I never had problem with this
>for the last 4-5 years (bought it new). Since it is lifetime warranty,
>it was sent in for a replacement.
>I am curious. How often do old RAM (machine runs almost 24/7) go bad
>like this?