The noise comes from the "parking" of the heads when HDD is idle. The problem is that moments after that the heads go back because of I/O activity then it parks again and again. I got that suddenly after 2 years of using my GRT815M. Interestingly enough, running the computer under DOS or Knoppix (Linux LiveCD distribution) did not reproduce the clicking nightmare... I will leave the conclusions for the reader.
Possible (brutal force) solution:
Check with some "S.M.A.R.T." tool the count of the "parks" - if this is the reason for the "clicks" it will increase the number after every click.
Get to the HDD vendor web page and get the software they are providing to configure the HDD. In my case - IBM/Hitachi Feature Tool 2.01 [link] and read the manual. Usually it goes to a parameter they relate to performance or power save. Sliding toward performance increases the time before the HDD parks the heads. You can increase the value a little bit and it might work, otherwise increase more. Above some limit it will stop parking at all. If you are lucky to have Hitachi Travelstar HDD check this tool [link] that allows you to to change the Advanced Power Management (APM) level directly from Windows.
I use my "laptop" as a desktop replacement so I don't care that much if the heads are safely parked. Power saving is not an issue since on my computer there are so many other things that drain more power than the hard disk itself (I have 3 fans and one of them is constantly running - I have difficulties to classify this as a laptop).
Check with some "S.M.A.R.T." tool the count of the "parks" - if this is the reason for the "clicks" it will increase the number after every click.
Get to the HDD vendor web page and get the software they are providing to configure the HDD. In my case - IBM/Hitachi Feature Tool 2.01 [link] and read the manual. Usually it goes to a parameter they relate to performance or power save. Sliding toward performance increases the time before the HDD parks the heads. You can increase the value a little bit and it might work, otherwise increase more. Above some limit it will stop parking at all. If you are lucky to have Hitachi Travelstar HDD check this tool [link] that allows you to to change the Advanced Power Management (APM) level directly from Windows.I use my "laptop" as a desktop replacement so I don't care that much if the heads are safely parked. Power saving is not an issue since on my computer there are so many other things that drain more power than the hard disk itself (I have 3 fans and one of them is constantly running - I have difficulties to classify this as a laptop).
One can do the same APM adjustments from Linux with hdparm -B /dev/hda etc.
from the "hdparm" manual: -B Set Advanced Power Management feature, if the drive supports it. A low value means aggressive power management and a high value means better performance. A value of 255 will disable apm on the drive.
Under Windows one can use this command line tool to check the SMART values:
http://www.beyondlogic.org/solutions/smart/smart.htm
As administrator (in "Command Prompt" window) run smart.exe
C:>smart.exe
SMART & Simple for Windows NT/2000/XP V1.01
Copyright 2001-2003 Craig.Peacock@beyondlogic.org
Opened Drive \\.\c: . .
SMART Enabled : Yes
Model Number : IC25N080ATMR04-0
Firmware Version : MO4OAD1A
Serial Number : xxxxxxxxxxxxxxxxxxxx
Drive Size : 80.026 GB
ID Attribute Type Threshold Value Worst Raw St
---- -------------------------- ----- --------- ----- ----- ---------- --
[01] Raw Read Error Rate Prefailure 62 100 100 0 OK
[02] Throughput Performance Prefailure 40 106 106 5825 OK
[03] Spin Up Time Prefailure 33 125 125 1 OK
[04] Start/Stop Count Advisory 0 98 98 3409 OK
[05] Reallocated Sector Count Prefailure 5 100 100 0 OK
[07] Seek Error Rate Prefailure 67 100 100 0 OK
[08] Seek Time Performance Prefailure 40 100 100 45 OK
[09] Power On Hours Count Advisory 0 84 84 7284 OK
[0A] Spin Retry Count Prefailure 60 100 100 0 OK
[0C] Power Cycle Count Advisory 0 99 99 1997 OK
[BF] Unknown SMART Attribute Advisory 0 100 100 0 OK
[C0] Power-Off Park Count Advisory 0 100 100 3 OK
[C1] Load/Unload Cycle Count Advisory 0 90 90 106573 OK
[C2] Drive Temperature Advisory 0 157 157 1179683 OK
[C4] Re-Allocated Data Count Advisory 0 100 100 0 OK
[C5] Pending Sector Count Advisory 0 100 100 0 OK
[C6] UnCorrectable Sector Count Advisory 0 100 100 0 OK
[C7] CRC Error Count Advisory 0 200 200 0 OK
During my clicking nightmare the parameter “[C1] Load/Unload Cycle Count” was changing every time I heard the click. This is a certain way to identify the reason for the clicks.
I will suggest this tool as well, since it offers more features:
http://smartmontools.sourceforge.net/
C:>smartctl -A /dev/hda #(Linux conventions)
will show you the device SMART vendor-specific attributes and values as the previous command line tool.
You can run some additional internal self-test in case you want to make sure that there is nothing wrong with the drive on the first place (you may continue to work while the test is running):
You can run some additional internal self-test in case you want to make sure that there is nothing wrong with the drive on the first place (you may continue to work while the test is running):
C:>smartctl -t short /dev/hda #(Linux conventions)
smartctl version 5.34 [i686-pc-linux-gnu] Copyright (C) 2002-5 Bruce Allen
Home page is http://smartmontools.sourceforge.net/
=== START OF OFFLINE IMMEDIATE AND SELF-TEST SECTION ===
Sending command: "Execute SMART Short self-test routine immediately in off-line mode".
Drive command "Execute SMART Short self-test routine immediately in off-line mode" successful.
Testing has begun.
Please wait 2 minutes for test to complete.
Test will complete after Mon Oct 9 10:23:00 2006
Use smartctl -X to abort test.
and after finishing (about 2 minutes for the short test) you may run the following command to check the result. BE CAREFUL WITH THAT TOOL!!!
C:>smartctl.exe -l selftest /dev/hda
smartctl version 5.33 [i386-pc-mingw32] Copyright (C) 2002-4 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) LB
# 1 Short offline Completed without error 00% 7269 -
# 2 Short offline Completed without error 00% 7149 -
# 3 Short offline Completed without error 00% 6189 -
# 4 Short offline Completed without error 00% 5452 -
# 5 Short offline Completed without error 00% 5347 -
# 6 Short offline Completed without error 00% 5254 -
# 7 Short offline Completed without error 00% 5192 -
# 8 Extended offline Completed without error 00% 5186 -
# 9 Short offline Completed without error 00% 5184 -
I have seen on Internet that this is not model or vendor specific. It appears that the problem is triggered by settings that set some services or program to write frequently to the disk. If your disk is set to park after 5 seconds of inactivity and there is a program that writes every 7 seconds - there you are.
http://www.thinkwiki.org/wiki/Problem_with_hard_drive_clicking
Continue reading: clicking hard disk:II
http://www.thinkwiki.org/wiki/Problem_with_hard_drive_clicking
Continue reading: clicking hard disk:II
No comments:
Post a Comment