Hi, Since I upgraded my 2 PCs at home and 9 PCs at the university to SuSE 9.3 (clean install, all partitions ReiserFS) I had 4 hard disks either crashing or showing bad sectors. In all cases the failure is (I still need to confirm one case) hardware related and there is no apparent pattern: they were Maxtor, Western Digital (2), and Seagate disks, EIDE and SATA, new and old, on Intel and on AMD boards, so I can't pin the blame on any particular brand or type. But I can say that it is definitely an unusually high number of hard disk failures. Since I am one of only 2 in the department running SuSE and my colleague also had a hard disk crash while running 9.3, whereas the rest of the MSerfs (and a couple of Mac addicts) in the department may have lost one HD in the same period, it is starting to look bad. :( I wonder if it is just here, or if someone else also noticed higher number of crashes. I remember a previous thread from July, where people tried to blame ReiserFS and others convinced them it was mainly hardware. Well, hardware seems to be the case here, but then again, SuSE and ReiserFS were the only things in common among the 4 failed HDs I had in the past 4 months. Am I seeing things, or what? Carlos -- Carlos Frederico Lange -------------------------------------------------------------------- Scientific Programmer: professional who loves to _find_ own errors.
On 03/11/05, Carlos F Lange <carlos.lange@ualberta.ca> wrote:
Hi,
Since I upgraded my 2 PCs at home and 9 PCs at the university to SuSE 9.3 (clean install, all partitions ReiserFS) I had 4 hard disks either crashing or showing bad sectors. In all cases the failure is (I still need to confirm one case) hardware related and there is no apparent pattern: they were Maxtor, Western Digital (2), and Seagate disks, EIDE and SATA, new and old, on Intel and on AMD boards, so I can't pin the blame on any particular brand or type.
But I can say that it is definitely an unusually high number of hard disk failures. Since I am one of only 2 in the department running SuSE and my colleague also had a hard disk crash while running 9.3, whereas the rest of the MSerfs (and a couple of Mac addicts) in the department may have lost one HD in the same period, it is starting to look bad. :(
I wonder if it is just here, or if someone else also noticed higher number of crashes. I remember a previous thread from July, where people tried to blame ReiserFS and others convinced them it was mainly hardware. Well, hardware seems to be the case here, but then again, SuSE and ReiserFS were the only things in common among the 4 failed HDs I had in the past 4 months. Am I seeing things, or what?
Carlos --
Forgive me, but how on earth could a file system be held responsible for destroying a hard drive? I don't see how ext2, 3, ReiserFS, HFS, FAT, NTFS or any other could damage the drive. All they are doing is ordering the magnetic blocks of data on the platters. Aren't they? -- ============================================== I am only human, please forgive me if I make a mistake it is not deliberate. ============================================== Take care. Kevan Farmer 34 Hill Street Cheslyn Hay Staffordshire WS6 7HR
Carlos F Lange wrote:
Hi,
Since I upgraded my 2 PCs at home and 9 PCs at the university to SuSE 9.3 (clean install, all partitions ReiserFS) I had 4 hard disks either crashing or showing bad sectors. In all cases the failure is (I still need to confirm one case) hardware related and there is no apparent pattern: they were Maxtor, Western Digital (2), and Seagate disks, EIDE and SATA, new and old, on Intel and on AMD boards, so I can't pin the blame on any particular brand or type.
But I can say that it is definitely an unusually high number of hard disk failures. Since I am one of only 2 in the department running SuSE and my colleague also had a hard disk crash while running 9.3, whereas the rest of the MSerfs (and a couple of Mac addicts) in the department may have lost one HD in the same period, it is starting to look bad. :(
I wonder if it is just here, or if someone else also noticed higher number of crashes. I remember a previous thread from July, where people tried to blame ReiserFS and others convinced them it was mainly hardware. Well, hardware seems to be the case here, but then again, SuSE and ReiserFS were the only things in common among the 4 failed HDs I had in the past 4 months. Am I seeing things, or what?
Carlos
Most SuSE/Mandriva/gentoo (I have them all) installs are reiserfs by default, so guess under what conditions failures occur. I have had a number of run-arounds with HD failures, some were due to bad IDE controllers so the disks were reclaimed after swapping out the motherboard and reformatting the drives, all others, 2 in about 2 years have been genuine HD failures. My bet is that if you install Fedora core which is ext3, you'd have the same problems. I shot down reiserfs as the cause the last time the issue was raised on the grounds that if reiserfs was that bad, the kernel mailing list would be full of complaints, customers would be frothing at the mouth and RedHat/Fedora/Windows would be replacing the other distros in short order. Regards Sid. -- Sid Boyce ... Hamradio License G3VBV, licensed Private Pilot Retired IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support Specialist Microsoft Windows Free Zone - Linux used for all Computing Tasks
On Thursday 03 November 2005 21:42, Sid Boyce wrote: <snip>
Most SuSE/Mandriva/gentoo (I have them all) installs are reiserfs by default, so guess under what conditions failures occur. I have had a number of run-arounds with HD failures, some were due to bad IDE controllers so the disks were reclaimed after swapping out the motherboard and reformatting the drives, all others, 2 in about 2 years have been genuine HD failures. My bet is that if you install Fedora core which is ext3, you'd have the same problems. I shot down reiserfs as the cause the last time the issue was raised on the grounds that if reiserfs was that bad, the kernel mailing list would be full of complaints, customers would be frothing at the mouth and RedHat/Fedora/Windows would be replacing the other distros in short order. Regards Sid.
I agree. Reiser and Ext3 formats are not the problem. Another thing that can be a serious problem is cooling on the hard drives, or the system in general. Being involved in PC hardware for 2 decades I can tell you that ALL manufactures have problems. I've had Linksys network cards fail in batches of 30+, (30 outta 50) and I've seen the same from hard drives from WD, SG, IBM, Maxtor, etc. IBM's drives bit me in the *ss , 4 failed in 7 months..., not happy. The IBM bug bit me while running Debian unstable (ha!) about 14 months ago. I used both Reiser and ext3, but it just came back to bad hardware. Almost *all* of these failures are related to a certain serial # or manufaturing date / batch. As a FYI, just cause it's new doesn't mean it isn't DOA. ;) Dana
On Thursday 03 November 2005 23:16, Dana J. Laude wrote:
On Thursday 03 November 2005 21:42, Sid Boyce wrote:
<snip>
Most SuSE/Mandriva/gentoo (I have them all) installs are reiserfs by default, so guess under what conditions failures occur. I have had a number of run-arounds with HD failures, some were due to bad IDE controllers so the disks were reclaimed after swapping out the motherboard and reformatting the drives, all others, 2 in about 2 years have been genuine HD failures. My bet is that if you install Fedora core which is ext3, you'd have the same problems. I shot down reiserfs as the cause the last time the issue was raised on the grounds that if reiserfs was that bad, the kernel mailing list would be full of complaints, customers would be frothing at the mouth and RedHat/Fedora/Windows would be replacing the other distros in short order. Regards Sid.
I agree. Reiser and Ext3 formats are not the problem. Another thing that can be a serious problem is cooling on the hard drives, or the system in general.
Being involved in PC hardware for 2 decades I can tell you that ALL manufactures have problems. I've had Linksys network cards fail in batches of 30+, (30 outta 50) and I've seen the same from hard drives from WD, SG, IBM, Maxtor, etc. IBM's drives bit me in the *ss , 4 failed in 7 months..., not happy.
The IBM bug bit me while running Debian unstable (ha!) about 14 months ago. I used both Reiser and ext3, but it just came back to bad hardware.
Almost *all* of these failures are related to a certain serial # or manufaturing date / batch.
As a FYI, just cause it's new doesn't mean it isn't DOA. ;)
Dana
OK, Thank you all for your comments. I think I just needed to hear from a wider sample of SuSE users that my small sample statistics is not representative of a trend. It is also comforting to know that there is no fundamental problem with the setup of my systems, with which I am otherwise very happy. I will make a note to keep the fan filters free of dust to prevent overheating. And thanks to backup, I haven't lost any data. And I will fend off cynical comments from colleagues, saying that millions of other SuSE users are not having the same problem, so it is just a wave of bad luck that will pass. Good weekend to all, Carlos -- Carlos Frederico Lange -------------------------------------------------------------------- Scientific Programmer: professional who loves to _find_ own errors.
Dana J. Laude wrote: <STUFF DELETED>
I agree. Reiser and Ext3 formats are not the problem. Another thing that can be a serious problem is cooling on the hard drives, or the system in general.
Being involved in PC hardware for 2 decades I can tell you that ALL manufactures have problems. I've had Linksys network cards fail in batches of 30+, (30 outta 50) and I've seen the same from hard drives from WD, SG, IBM, Maxtor, etc. IBM's drives bit me in the *ss , 4 failed in 7 months..., not happy.
They didn't unjustly get the tag "IBM DeathStar". Regards Sid.
The IBM bug bit me while running Debian unstable (ha!) about 14 months ago. I used both Reiser and ext3, but it just came back to bad hardware.
Almost *all* of these failures are related to a certain serial # or manufaturing date / batch.
As a FYI, just cause it's new doesn't mean it isn't DOA. ;)
Dana
PC hardware is still designed for cheap so diagnostics hardly exist and resilience hasn't yet been dreamt of. Regards Sid. -- Sid Boyce ... Hamradio License G3VBV, licensed Private Pilot Retired IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support Specialist Microsoft Windows Free Zone - Linux used for all Computing Tasks
<STUFF DELETED>
i have monitored this thread b/c my suse box have given these kind of problems.... i run 3 linux @ home, slackware and suse and my suse machine in 5 years i am on my 2 HD and i remeber on another suse box had similar issue. i thought it was justthat HD now last shorter but this is an interesting thread
On Saturday 05 Nov 2005 10:27, boricua wrote:
<STUFF DELETED>
i have monitored this thread b/c my suse box have given these kind of problems.... i run 3 linux @ home, slackware and suse and my suse machine in 5 years i am on my 2 HD and i remeber on another suse box had similar issue.
i thought it was justthat HD now last shorter but this is an interesting thread
It would appear that modern Hard Drives do not like or are not made to run continuousley or maybe i should say the retail versions of them at any rate the IBM drives very nice fast but aptly nicknamed deathstars i have had 1 die compleyely and another one started to fade so i took it out Maxstor likewise Seagate the cheap ones wiht the small cache on them seem to drop like dying flies the one with the 8Mb cache so far sem ok not had the latest one show problems yet (Xx) crossed fingers and toes i gettin sick of changing them the IBM by the way died on an Slackware 10.0 box running EXT2 fs I see it all as a ploy to get us to spend more money make the drives crap so we have to replace them more frequently OR :-) ..maybe Mr Gates &Co have teamed up with the hdd manafacturers (SP) to make crap drives cus he is losing money on windBloZe now so he is making a slice of the replacement hdd market one never knows what tricks he will get up to .. Cheers folks .. Pete . -- If Bill Gates had gotten LAID at High School do YOU think there would be a Microsoft ? Of course NOT ! You gotta spend a lot of time at your school Locker stuffing underware up your ass to think , I am going to take on the worlds Computer Industry -------:heard on Cyber Radio.:------- AFFA
participants (6)
-
boricua -
Carlos F Lange -
Dana J. Laude -
Kevanf1 -
Peter Nikolic -
Sid Boyce