| View previous topic :: View next topic |
| Author |
Message |
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
 Posted: Thu Mar 05, 2009 4:10 pm Post subject: Slow HDD Performance [solved] |
 |
|
Hi,
Now it seems to be that in the last couple of months, the performance of my hard drive has noticeably decreased. If I have concurrent downloads running or am compiling, my computer becomes extremely sluggish. I have the impression that it used to be more responsive before, I certainly do not know what the cause could be. Off the top of my head, I can think of either some configuration errors with the HDD after some upgrades or just old hard drive in general. Anyways, below are a couple of outputs from hdparm (these are with xorg and some other stuff running)
| Code: |
googly rahul # hdparm -tT /dev/sda5
/dev/sda5:
Timing cached reads: 2288 MB in 2.00 seconds = 1144.63 MB/sec
Timing buffered disk reads: 96 MB in 3.01 seconds = 31.86 MB/sec
|
| Code: |
googly rahul # hdparm -iI /dev/sda6
/dev/sda6:
Model=Hitachi HTS541060G9SA00 , FwRev=MB3OC60R, SerialNo= MPBCP0XGJHWD7M
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=DualPortCache, BuffSize=7538kB, MaxMultSect=16, MultSect=?8?
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=117210240
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
Drive conforms to: ATA/ATAPI-7 T13 1532D revision 1: ATA/ATAPI-2,3,4,5,6,7
* signifies the current active mode
ATA device, with non-removable media
Model Number: Hitachi HTS541060G9SA00
Serial Number: MPBCP0XGJHWD7M
Firmware Revision: MB3OC60R
Standards:
Used: ATA/ATAPI-7 T13 1532D revision 1
Supported: 7 6 5 4
Configuration:
Logical max current
cylinders 16383 16383
heads 16 16
sectors/track 63 63
--
CHS current addressable sectors: 16514064
LBA user addressable sectors: 117210240
LBA48 user addressable sectors: 117210240
Logical/Physical Sector size: 512 bytes
device size with M = 1024*1024: 57231 MBytes
device size with M = 1000*1000: 60011 MBytes (60 GB)
cache/buffer size = 7538 KBytes (type=DualPortCache)
Capabilities:
LBA, IORDY(can be disabled)
Queue depth: 32
Standby timer values: spec'd by Vendor, no device specific minimum
R/W multiple sector transfer: Max = 16 Current = 8
Advanced power management level: 128
Recommended acoustic management value: 128, current value: 128
DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=240ns IORDY flow control=120ns
Commands/features:
Enabled Supported:
* SMART feature set
Security Mode feature set
* Power Management feature set
* Write cache
* Look-ahead
* Host Protected Area feature set
* WRITE_BUFFER command
* READ_BUFFER command
* NOP cmd
* DOWNLOAD_MICROCODE
* Advanced Power Management feature set
Power-Up In Standby feature set
* SET_FEATURES required to spinup after power up
SET_MAX security extension
* Automatic Acoustic Management feature set
* 48-bit Address feature set
* Device Configuration Overlay feature set
* Mandatory FLUSH_CACHE
* FLUSH_CACHE_EXT
* SMART error logging
* SMART self-test
* General Purpose Logging feature set
* WRITE_{DMA|MULTIPLE}_FUA_EXT
* Gen1 signaling speed (1.5Gb/s)
* Native Command Queueing (NCQ)
* Host-initiated interface power management
Non-Zero buffer offsets in DMA Setup FIS
DMA Setup Auto-Activate optimization
Device-initiated interface power management
In-order data delivery
* Software settings preservation
Security:
Master password revision code = 65534
supported
not enabled
not locked
frozen
not expired: security count
not supported: enhanced erase
40min for SECURITY ERASE UNIT.
Checksum: correct
|
_________________
Who shall guard the guards?
Last edited by rahulthewall on Fri Mar 13, 2009 3:27 pm; edited 1 time in total |
|
| Back to top |
|
 |
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Thu Mar 05, 2009 7:48 pm Post subject: |
|
|
OK guys, I am getting worried here. KDE-4.2.1 has been compiling for the past 3 hours now and only 70 of the 151 packages have compiled by now and that seems abnormally slow to me.
If there is any way to check my hard drive while it is still mounted, I would like to do that. Otherwise, please point me to other ways of checking my hard disk, I live in abject fear of losing my data and am taking a back up right now.
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54832
Location: 56N 3W
|
| Posted: Thu Mar 05, 2009 8:16 pm Post subject: |
|
|
rahulthewall,
As hard drives are zoned, they are faster at the outside than near the spindle. Zones have a lower number of sectors per track as the track diameter decreases. You can detect this with hdparam -tT for each partition.
As drives wear through life, they remap failing sectors to spare sectors provided for this purpose. When the spares are all used, the drive is end of life. smartmontools can read the drives SMART data and tell you how its doing.
This remapping has the effect of slowing access, as sectors the are logically sequential may not be physically next to one another on the disk surface. This incurrs extra head movement and/or waiting for the sector to come under the head, both of which reduce the data rate.
The on disk cache helps minimise this effect.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Thu Mar 05, 2009 9:23 pm Post subject: |
|
|
So, I installed smartmontools and my HDD had smart mode enabled, so I ran few tests. Here is what I get:
| Code: | googly yawp # smartctl -H /dev/sda
smartctl version 5.38 [i686-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/
=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED |
| Code: |
googly yawp # smartctl -l error /dev/sda
smartctl version 5.38 [i686-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/
=== START OF READ SMART DATA SECTION ===
SMART Error Log Version: 1
ATA Error Count: 82 (device log contains only the most recent five errors)
CR = Command Register [HEX]
FR = Features Register [HEX]
SC = Sector Count Register [HEX]
SN = Sector Number Register [HEX]
CL = Cylinder Low Register [HEX]
CH = Cylinder High Register [HEX]
DH = Device/Head Register [HEX]
DC = Device Command Register [HEX]
ER = Error register [HEX]
ST = Status register [HEX]
Powered_Up_Time is measured from power on, and printed as
DDd+hh:mm:SS.sss where DD=days, hh=hours, mm=minutes,
SS=sec, and sss=millisec. It "wraps" after 49.710 days.
Error 82 occurred at disk power-on lifetime: 11397 hours (474 days + 21 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
40 51 01 32 78 9f e0 Error: UNC at LBA = 0x009f7832 = 10450994
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
40 00 01 32 78 9f e0 00 00:06:42.500 READ VERIFY SECTOR(S)
c8 00 01 0a 62 90 e3 00 00:06:42.500 READ DMA
40 00 01 31 78 9f e0 00 00:06:42.500 READ VERIFY SECTOR(S)
c8 00 01 6e 5f 30 e2 00 00:06:42.500 READ DMA
40 00 02 31 78 9f e0 00 00:06:38.300 READ VERIFY SECTOR(S)
Error 81 occurred at disk power-on lifetime: 11397 hours (474 days + 21 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
40 51 01 32 78 9f e0 Error: UNC at LBA = 0x009f7832 = 10450994
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
40 00 02 31 78 9f e0 00 00:06:38.300 READ VERIFY SECTOR(S)
c8 00 01 00 00 00 e0 00 00:06:38.300 READ DMA
40 00 02 2f 78 9f e0 00 00:06:38.300 READ VERIFY SECTOR(S)
c8 00 01 00 00 00 e0 00 00:06:38.300 READ DMA
40 00 04 33 78 9f e0 00 00:06:38.300 READ VERIFY SECTOR(S)
Error 80 occurred at disk power-on lifetime: 11397 hours (474 days + 21 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
40 51 01 32 78 9f e0 Error: UNC at LBA = 0x009f7832 = 10450994
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
40 00 04 2f 78 9f e0 00 00:06:34.100 READ VERIFY SECTOR(S)
c8 00 01 0a 62 90 e3 00 00:06:34.100 READ DMA
40 00 08 37 78 9f e0 00 00:06:34.100 READ VERIFY SECTOR(S)
c8 00 01 6e 5f 30 e2 00 00:06:34.100 READ DMA
40 00 08 2f 78 9f e0 00 00:06:29.900 READ VERIFY SECTOR(S)
Error 79 occurred at disk power-on lifetime: 11397 hours (474 days + 21 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
40 51 05 32 78 9f e0 Error: UNC at LBA = 0x009f7832 = 10450994
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
40 00 08 2f 78 9f e0 00 00:06:29.900 READ VERIFY SECTOR(S)
c8 00 01 00 00 00 e0 00 00:06:29.900 READ DMA
40 00 10 2f 78 9f e0 00 00:06:25.700 READ VERIFY SECTOR(S)
c8 00 01 00 00 00 e0 00 00:06:25.700 READ DMA
40 00 10 1f 78 9f e0 00 00:06:25.700 READ VERIFY SECTOR(S)
Error 78 occurred at disk power-on lifetime: 11397 hours (474 days + 21 hours)
When the command that caused the error occurred, the device was active or idle.
After command completion occurred, registers were:
ER ST SC SN CL CH DH
-- -- -- -- -- -- --
40 51 0d 32 78 9f e0 Error: UNC at LBA = 0x009f7832 = 10450994
Commands leading to the command that caused the error were:
CR FR SC SN CL CH DH DC Powered_Up_Time Command/Feature_Name
-- -- -- -- -- -- -- -- ---------------- --------------------
40 00 10 2f 78 9f e0 00 00:06:25.700 READ VERIFY SECTOR(S)
c8 00 01 00 00 00 e0 00 00:06:25.700 READ DMA
40 00 10 1f 78 9f e0 00 00:06:25.700 READ VERIFY SECTOR(S)
c8 00 01 00 00 00 e0 00 00:06:25.700 READ DMA
40 00 20 1f 78 9f e0 00 00:06:21.500 READ VERIFY SECTOR(S) |
| Code: |
googly yawp # smartctl -l selftest /dev/sda
googly yawp # smartctl -l selftest /dev/sda
smartctl version 5.38 [i686-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/
=== START OF READ SMART DATA SECTION ===
SMART Self-test log structure revision number 1
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Short offline Completed without error 00% 5528 -
# 2 Short offline Completed without error 00% 0 -
# 3 Short captive Completed without error 00% 0 - |
So, there are apparently 82 errors - are they recent and how do I check this?
Thanks
Rahul
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54832
Location: 56N 3W
|
| Posted: Thu Mar 05, 2009 10:25 pm Post subject: |
|
|
rahulthewall,
| Code: | | Error 82 occurred at disk power-on lifetime: 11397 hours (474 days + 21 hours) |
Only the last 5 errors are stored and they all occured within the same hour.
The table of parameters that smartmontools produces is useful too.
It has things like reallocated sector count.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Thu Mar 05, 2009 10:36 pm Post subject: |
|
|
OK,
Quick question, should I consider changing the HD? I have a three year complete warranty on my laptop and I think I can easily get a replacement for the HD. Would you say it is necessary at this point?
Cheers
Rahul
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54832
Location: 56N 3W
|
| Posted: Thu Mar 05, 2009 10:51 pm Post subject: |
|
|
rahulthewall,
If you can move your data off for a data destructive test, run the hard drive manufactuers test application.
Some of them even check your warranty and print a RMA if the test fails.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Fri Mar 06, 2009 11:26 am Post subject: |
|
|
| NeddySeagoon wrote: | rahulthewall,
If you can move your data off for a data destructive test, run the hard drive manufactuers test application.
Some of them even check your warranty and print a RMA if the test fails. |
My point is that if I am moving the data entirely off the HDD anyways, I might as well get a replacement. The thing is that the University has a contract with the manufacturer and we get 3 years complete warranty + 3 years accidental damage insurance which includes pretty much any damage one can think of. So, if my HDD is in a questionable state, I might just ask for a new one.
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
Cyker
Veteran
Joined: 15 Jun 2006
Posts: 1746
|
| Posted: Fri Mar 06, 2009 12:57 pm Post subject: |
|
|
It might be tricky to get it replaced under warrenty for that sort of error, 'tho 82 seems a bit high.
If it had bad sectors that wouldn't go away, or the SMART test Failed, then the procedure would be easier. The only sure way is if the drive is totally dead, or if the OEM's drive fitness/diagnostic tool (e.g. PowerMax with Maxtor, DFT for IBM/Hitachi, SeaTools for Seagate, Data Lifeguard tools for Western Digital etc.) certifies that the drive needs replacement (Easiest option as the tools usually produce an error code).
I would try running such a tool on the drive.
The problem with these tools is the diagnostics only work in DOS which is a real PITA, esp. if your controller doesn't work in DOS (Ah for the days of BIOS standard hardware protocols!). |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Fri Mar 06, 2009 1:47 pm Post subject: |
|
|
Am gonna try running the DFT, anyways the extended offline test fails with a read error ...
| Code: |
googly rahul # smartctl -l selftest /dev/sda
smartctl version 5.38 [i686-pc-linux-gnu] Copyright (C) 2002-8 Bruce Allen
Home page is http://smartmontools.sourceforge.net/
=== START OF READ SMART DATA SECTION ===
SMART Self-test log structure revision number 1
Num Test_Description Status Remaining LifeTime(hours) LBA_of_first_error
# 1 Extended offline Completed: read failure 80% 13392 10450994
# 2 Short offline Completed without error 00% 5528 -
# 3 Short offline Completed without error 00% 0 -
# 4 Short captive Completed without error 00% 0 -
|
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54832
Location: 56N 3W
|
| Posted: Fri Mar 06, 2009 2:06 pm Post subject: |
|
|
rahulthewall,
Thats a bad sign, the data in LBA 10450994 (if any) has been lost.
smartmontools may show that all the spare sectors have been used and that this sector is waiting for remapping, which will never happen.
The disk vendor tools (well some) have been stand alone for a while now. Follow the instructions on the website.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Fri Mar 06, 2009 4:02 pm Post subject: |
|
|
Ran an extended test with DFT (HDD is Hitachi) and it told me that I have a defective device. So, I called up the service center and they are going to give me a replacement hard drive.
Thanks for the tip guys. 
Cheers
Rahul
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Fri Mar 06, 2009 6:09 pm Post subject: |
|
|
And now, a more pertinent question. I would rather not reinstall windows and Gentoo when I get a new hard drive. So what I would like to know is what is the best wawy to clone my entire hard drive so that it can be replicated on the new hard drive.
Cheers
Rahul
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54832
Location: 56N 3W
|
| Posted: Fri Mar 06, 2009 7:31 pm Post subject: |
|
|
rahulthewall,
As the drive is faulty, there are two thing to try.
1. read the filesystems to backup files with something like tar. If the faults in in unused areas of the filesysem, this will work. It also has the advantage of only copying used space. You will need to reinstall grub to the MBR after you have restored from backup.
This even works for windows as ntfs-3g reads and writes windows files correctly.
2. You can use ddrescue (not dd_rescue) to make an image of the faulty drive to a file, including all the empty space. Then restore it to the new drive. ddrescue will never complete as the drive is faulty, it will leave 'holes' where data is missing. It works in the same way as dd until it hits an error, then does a recursive binary search on the remaining space.
This means it recovers easy to read data first and approaches problem areas by stepping the head in both directions to have the best chance of reading.
Hint: Drive failures are often caused by failure of the platter bearings. You can often get one more read (and thats all you want) by operating the drive in an orientation is not normally used in. i.e. in a laptop, on edge.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
pappy_mcfae
Watchman
Joined: 27 Dec 2007
Posts: 5999
Location: Pomona, California.
|
| Posted: Fri Mar 06, 2009 8:00 pm Post subject: |
|
|
Ghost will work for that. I use the version that comes with Hiren's Boot CD. It can move all partitions in one operation. Once done, you'll have to chroot and setup grub/lilo, and that's it. And yes, definitely try to orient that hard drive outside it's normal operating position while doing the deed. Upside down would be best.
Blessed be!
Pappy
_________________
This space left intentionally blank, except for these ASCII symbols. |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Fri Mar 06, 2009 9:21 pm Post subject: |
|
|
| NeddySeagoon wrote: | rahulthewall,
As the drive is faulty, there are two thing to try.
1. read the filesystems to backup files with something like tar. If the faults in in unused areas of the filesysem, this will work. It also has the advantage of only copying used space. You will need to reinstall grub to the MBR after you have restored from backup.
This even works for windows as ntfs-3g reads and writes windows files correctly.
|
If I use tar, will it preserve the file permissions and everything?
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54832
Location: 56N 3W
|
| Posted: Fri Mar 06, 2009 10:53 pm Post subject: |
|
|
rahulthewall,
Yes. When you restore, you must use the p option.
I don't think p is needed when you make the tarball, tar saves the information anyway but it would do no harm to use p both times
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
rahulthewall
Veteran
Joined: 01 Nov 2007
Posts: 1264
Location: Zürich
|
| Posted: Fri Mar 13, 2009 3:27 pm Post subject: |
|
|
In stead of tar, I used fsarchiver to backup my data. It was faster and looks like a nice option to me.
Plus, I must say I am really impressed with the DELL service. Granted, it took them a week to get the stuff that I wanted (I wanted a new keyboard a new casing for my laptop as well) these guys got the stuff and replaced everything. Even gave me a small 20GB HDD upgrade.
DELL service really rocks, am restoring my data as I type.
_________________
Who shall guard the guards? |
|
| Back to top |
|
|
|
Kernel & Hardware
