Go Back   nV News Forums > Linux Support Forums > General Linux

Newegg Daily Deals

Reply
 
Thread Tools
Old 01-03-07, 09:28 AM   #37
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by netllama
I reviewed the LKML thread that you referenced and the problem descriptions there sound vastly different from what you're reporting. Your issue sounds like filesystem and/or in-memory corruption, however the issue on LKML isn't occuring on the filesystem level, but in the files themselves.
See the later posts in our thread for our solutions/workarounds (i.e. bios setting memhole mapping=disabled or kernel parameter iommu=soft). You're right it most likely happens below filesystem level...

Quote:
Originally Posted by netllama
Additionally, you stated that "it boots with some scary messages" and "mplayer segfaults". It sounds like the data on your disk(s) is getting corrupted even when its not being actively written to, which is not the same issue as was reported on LKML.
The corruption definitely occurs also on simple reads because my continuing sha512sums checks bring errors on different files...

Quote:
Originally Posted by netllama
At this point, the information that you've provided suggests a hardware problem (faulty RAM or disk). If you can provide information that suggests otherwise, I can look into your issue further.
A simple test for domasj would be to use iommu=soft as a kernel parameter. If it works with this setting he is probably suffering from the same problem.
(Note that the issue probably cannot occur with Windows and/or Intel CPUs and/or main memory below some value).
calestyo is offline   Reply With Quote
Old 01-03-07, 09:32 AM   #38
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by krader
Also, while others have reported corruption with PATA disks I was unable to do so. It appeared that the speed of the disks is a factor.
Yes I suppose this too (as you can read in my lkml posts )

Chris.
calestyo is offline   Reply With Quote
Old 01-03-07, 09:36 AM   #39
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by netllama
I've run with your files & script for 3 iterations, and have not been able to reproduce the problem.
Which disk drives do you use? Exact type/model-# and vendor.
In the beginning of me research I found some reports (although I think it is unlikely that they're true) where people claimed it would only happen with disk drives from some vendors (e.g. Hitachi).
Some others claimed they solved the problem by applying a firmware update to their Maxtor disks.

In the meantime I got a firmware update from Hitachi but due to our latest results with iommu/memhole mapping I didn't try them out (because I won't be able to flash the firmware back to the older version to continue my tests).

Chris.
calestyo is offline   Reply With Quote
Old 01-03-07, 09:38 AM   #40
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by domasj
regardless the OS.
Yes and no... If you're actually having the same problem as we have at lkml... The issue is actually OS independant. But some OS (i.e. Windows) won't suffer from it, as they don't use the hardware iommu at all (even if available).

Chris.
calestyo is offline   Reply With Quote
Old 01-03-07, 09:40 AM   #41
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by netllama
You referenced filesystem corruption, whereas the LKML posts are not filesystem corruption, but rather file corruption. You've not provided any additional information to suggest that you're hitting the same issue as in the LKML posts. Are you seeing file corruption or filesystem corruption or both? Please provide detailed instructions on how can I reproduce the problem(s) that you initially reported.
It is true, that none of us at lkml reported filesystem corruption. But it is not sure that this wouldn't happen. The corruption occurs so rarely that file corruption is simply more likely thant filesystem corruption.

Chris.
calestyo is offline   Reply With Quote
Old 01-03-07, 09:43 AM   #42
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by netllama
No one in the LKML thread you referenced reported files getting corrupted on-disk.
What do you mean?
calestyo is offline   Reply With Quote
Old 01-03-07, 09:44 AM   #43
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by krader
> 2) Does this still reproduce if you reduce the RAM to 1GB or 512MB
> (booting with the mem= kernel parameter)?

Yes, the problem still occurred after booting with "mem=1g". Problem also
occurs if I remove half the memory (2 x 1 GiB DIMMS). Problem still occurs
if I swap the pairs of DIMMs I removed with the ones still in the system.
Memtest86 was run 24 hours on the full 4 GiB without error.
I cannot confirm this. IIRC most people (if not all) had the problem solved when using MEM=2G or so...
calestyo is offline   Reply With Quote
Old 01-03-07, 09:46 AM   #44
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by domasj
I recently got tired of testing my problem on Linux and decided to try that out in windows. However, I soon ran into a more obvious problem - installation even doesn't complete. It crashes either after restart or somewhere in the middle with different error every time. This behavior is seen on both of my disks. Now it seems like it is really hardware fault not software. Thus, I contacted the seller and they agreed to change the motherboard and the RAM stick to see if that helps. More information will come by the end of this week.

P. S. memtest86+ ran 3 passes on my system without any error whatsoever.
Windows should not suffer from the problem at all.

Chris.
calestyo is offline   Reply With Quote

Old 01-03-07, 09:48 AM   #45
calestyo
Christoph Anton Mitterer
 
Join Date: Dec 2006
Location: München, Germany
Posts: 48
Send a message via ICQ to calestyo Send a message via AIM to calestyo Send a message via MSN to calestyo Send a message via Yahoo to calestyo
Default Re: nForce 4 corrupting data written to HDD

Quote:
Originally Posted by netllama
What kind of errors was the FC6 kernel reporting? Have you considered the possibility that perhaps you have a faulty motherboard?
This is highly unlikely considering the number of people that suffer from this issue.

Another reason why should our two workarounds (iommu=soft and BIOS memholemapping=disabled) solve the issue on a faulty board?
calestyo is offline   Reply With Quote
Old 01-03-07, 03:35 PM   #46
krader
Registered User
 
Join Date: Dec 2006
Posts: 11
Default Re: nForce 4 corrupting data written to HDD

> I cannot confirm this. IIRC most people (if not all) had the problem
> solved when using MEM=2G or so.

That's because there appear to be two distinct problems. I just got off the phone with my counterpart at Red Hat. What they're seeing is a 4 KiB page being zerod unexpectedly. When they don't use the IOMMU (such as booting with "mem=2g") the problem goes away. That failure mode could be a Linux kernel defect or a hardware defect. The failure mode I'm seeing on my system are random 32-bit words of four of nine files being corrupted. My problem could not possibly be a kernel defect and definitely has nothing to do with the use of the hardware IOMMU as the corruption occurs with only 2 GiB installed, "mem=1g" as well as "iommu=soft".
krader is offline   Reply With Quote
Old 01-04-07, 11:58 AM   #47
domasj
Registered User
 
Join Date: Dec 2006
Posts: 11
Default Re: nForce 4 corrupting data written to HDD

Hello,

I got my motherboard and RAM stick replaced last week with identical parts. The happiest part is that it finally works without any problems! I got XP installed successfully then I formatted it and got my Debian installed.
After that I was able to correct my FSs using fsck (I couldn't correct that with the old system) and I haven't gotten any FS inconsistencies. I don't get any md5sum or sha512sum variations even with large files.
To conclude, my problems are gone. Appearantly the corruption was caused by the hardware issues. Several days ago I added an additional fan to be on the safer side in the future I hope the problem won't happen any more.
domasj is offline   Reply With Quote
Old 01-06-07, 09:18 PM   #48
jpskidmore
Registered User
 
Join Date: Jan 2007
Posts: 1
Default Re: nForce 4 corrupting data written to HDD

Hello,

I've been watching this thread for a while

I've been getting this problem on my M2PV-VM too - ever since i bought the board in november 06.

setting the drive speeds to 1.5G instead of 3.0G via jumpers (samsung drives) or the Hitachi Disk Utilities seems to have cured the problem. Does this mean I've got a faulty motherboard, or is this just par for the course with nforce 4?

thanks for any advise / time etc...

John, England, UK
jpskidmore is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


Similar Threads
Thread Thread Starter Forum Replies Last Post
Maintain Your Privacy by Manually Accepting and Rejecting "Cookies" (nV News) MikeC Open Forum 2 02-02-13 07:15 PM
Verizon's shared data plans won't save solo users much money News Archived News Items 0 06-12-12 10:40 AM
Verizon announces 'Share Everything' plans ' the future of mobile data (sort of News Archived News Items 0 06-12-12 10:40 AM
New Paper: MPI-ACC ' An Integrated Approach to Data Movement in Accelerators News Archived News Items 0 06-02-12 03:00 AM

All times are GMT -5. The time now is 06:46 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ©1998 - 2014, nV News.