[arch-dev-public] Load_Cycle_Count and storage-fixup
Hi all! I wanted to bring this long time ago when I discovered the problem with my old Seagate Momentus 5400.3, but then forgot since my new WD Scorpio Blue is less affected by this. But recently I noticed a couple of reports about the issue on forums and IRC, so I'm bringing this here. The problem is that on some HDDs (and/or with some BIOSes?) HDD head parking is done too often. This can be seen with Load_Cycle_Count in smartctl output being increased quickly (more than once per minute), and heard as relatively frequent clicks during low IO load. More at http://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequ... Basically, if Load_Cycle_Count divided by Power_On_Hours is more than 30-60 and your HDD is powered for many hours every day - your HDD may reach the limit in a couple of years. More info specifically about WD Green HDDs: http://kerneltrap.org/mailarchive/linux-kernel/2008/4/10/1396844/thread https://lists.ubuntu.com/archives/kernel-bugs/2008-April/035194.html (you can find wdidle3 with google, it's not accessible from WD site). I want to stress that this is _not_specific to WD Green drives. The problem is easy to fix, but that's if you _know_ about it. I think most of users are not aware of it. Now to the point of this message and why it is on this mailing list. What is the best way to handle this? There is a nice script which automatically tunes HDD options to fix this depending on HDD and/or laptop model. http://git.kernel.org/?p=linux/kernel/git/tj/storage-fixup.git Would it be okay if we package it and put it in core & install CD? Would it be enough to put this information in Official Install Guide? Would it be better to not care that much about this, and only mention it in some wiki page plus link in (Post-)Install Guide? I am really not sure what is the best way here. I'd like to hear your feedback on this. -- Roman Kyrylych (Роман Кирилич)
On Wed, Sep 16, 2009 at 10:09 PM, Roman Kyrylych <roman.kyrylych@gmail.com> wrote:
Hi all!
I wanted to bring this long time ago when I discovered the problem with my old Seagate Momentus 5400.3, but then forgot since my new WD Scorpio Blue is less affected by this.
But recently I noticed a couple of reports about the issue on forums and IRC, so I'm bringing this here.
The problem is that on some HDDs (and/or with some BIOSes?) HDD head parking is done too often. This can be seen with Load_Cycle_Count in smartctl output being increased quickly (more than once per minute), and heard as relatively frequent clicks during low IO load. More at http://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequ...
Basically, if Load_Cycle_Count divided by Power_On_Hours is more than 30-60 and your HDD is powered for many hours every day - your HDD may reach the limit in a couple of years.
More info specifically about WD Green HDDs: http://kerneltrap.org/mailarchive/linux-kernel/2008/4/10/1396844/thread https://lists.ubuntu.com/archives/kernel-bugs/2008-April/035194.html (you can find wdidle3 with google, it's not accessible from WD site).
I want to stress that this is _not_specific to WD Green drives.
The problem is easy to fix, but that's if you _know_ about it. I think most of users are not aware of it.
Now to the point of this message and why it is on this mailing list. What is the best way to handle this?
There is a nice script which automatically tunes HDD options to fix this depending on HDD and/or laptop model. http://git.kernel.org/?p=linux/kernel/git/tj/storage-fixup.git
Would it be okay if we package it and put it in core & install CD? Would it be enough to put this information in Official Install Guide? Would it be better to not care that much about this, and only mention it in some wiki page plus link in (Post-)Install Guide?
I am really not sure what is the best way here. I'd like to hear your feedback on this.
+1 for spreading more the noise about this. This was the opinion I expressed in this old thread : http://bbs.archlinux.org/viewtopic.php?id=39258
On Thu, Sep 17, 2009 at 02:09, Xavier <shiningxc@gmail.com> wrote:
On Wed, Sep 16, 2009 at 10:09 PM, Roman Kyrylych <roman.kyrylych@gmail.com> wrote:
Hi all!
I wanted to bring this long time ago when I discovered the problem with my old Seagate Momentus 5400.3, but then forgot since my new WD Scorpio Blue is less affected by this.
But recently I noticed a couple of reports about the issue on forums and IRC, so I'm bringing this here.
The problem is that on some HDDs (and/or with some BIOSes?) HDD head parking is done too often. This can be seen with Load_Cycle_Count in smartctl output being increased quickly (more than once per minute), and heard as relatively frequent clicks during low IO load. More at http://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequ...
Basically, if Load_Cycle_Count divided by Power_On_Hours is more than 30-60 and your HDD is powered for many hours every day - your HDD may reach the limit in a couple of years.
More info specifically about WD Green HDDs: http://kerneltrap.org/mailarchive/linux-kernel/2008/4/10/1396844/thread https://lists.ubuntu.com/archives/kernel-bugs/2008-April/035194.html (you can find wdidle3 with google, it's not accessible from WD site).
I want to stress that this is _not_specific to WD Green drives.
The problem is easy to fix, but that's if you _know_ about it. I think most of users are not aware of it.
Now to the point of this message and why it is on this mailing list. What is the best way to handle this?
There is a nice script which automatically tunes HDD options to fix this depending on HDD and/or laptop model. http://git.kernel.org/?p=linux/kernel/git/tj/storage-fixup.git
Would it be okay if we package it and put it in core & install CD? Would it be enough to put this information in Official Install Guide? Would it be better to not care that much about this, and only mention it in some wiki page plus link in (Post-)Install Guide?
I am really not sure what is the best way here. I'd like to hear your feedback on this.
+1 for spreading more the noise about this. This was the opinion I expressed in this old thread : http://bbs.archlinux.org/viewtopic.php?id=39258
Anyone else has some opinion about how to handle this? -- Roman Kyrylych (Роман Кирилич)
Roman Kyrylych schrieb:
+1 for spreading more the noise about this. This was the opinion I expressed in this old thread : http://bbs.archlinux.org/viewtopic.php?id=39258
Anyone else has some opinion about how to handle this?
Wow, I had this problem when switching to ext4 and thought my drive was broken. Just recently, I tried again and used wdidle3 to disable the idle timer. Now, it's all better.
I do think we should implement a sane default. I've known about this issue for some time and have always added the hdparm line to /etc/rc.local, but I think the vast majority of users, and not just Arch users, remain unaware. I just picked up an MSI Wind U120 this week which has a WD drive. It was clicking so often and so loudly that I thought the disk might need servicing. That's when I realized that the power management values were being reset every time it woke up from standby. It's fixed now, but it's a hassle (and danger) that users shouldn't have deal with. On Sep 21, 2009 at 01:21 AM, Thomas Bächler <thomas@archlinux.org> wrote:
Roman Kyrylych schrieb:
+1 for spreading more the noise about this. This was the opinion I expressed in this old thread : http://bbs.archlinux.org/viewtopic.php?id=39258
Anyone else has some opinion about how to handle this?
Wow, I had this problem when switching to ext4 and thought my drive was broken. Just recently, I tried again and used wdidle3 to disable the idle timer. Now, it's all better.
Roman Kyrylych wrote:
On Thu, Sep 17, 2009 at 02:09, Xavier <shiningxc@gmail.com> wrote:
On Wed, Sep 16, 2009 at 10:09 PM, Roman Kyrylych <roman.kyrylych@gmail.com> wrote:
Hi all!
I wanted to bring this long time ago when I discovered the problem with my old Seagate Momentus 5400.3, but then forgot since my new WD Scorpio Blue is less affected by this.
But recently I noticed a couple of reports about the issue on forums and IRC, so I'm bringing this here.
The problem is that on some HDDs (and/or with some BIOSes?) HDD head parking is done too often. This can be seen with Load_Cycle_Count in smartctl output being increased quickly (more than once per minute), and heard as relatively frequent clicks during low IO load. More at http://ata.wiki.kernel.org/index.php/Known_issues#Drives_which_perform_frequ...
Basically, if Load_Cycle_Count divided by Power_On_Hours is more than 30-60 and your HDD is powered for many hours every day - your HDD may reach the limit in a couple of years.
More info specifically about WD Green HDDs: http://kerneltrap.org/mailarchive/linux-kernel/2008/4/10/1396844/thread https://lists.ubuntu.com/archives/kernel-bugs/2008-April/035194.html (you can find wdidle3 with google, it's not accessible from WD site).
I want to stress that this is _not_specific to WD Green drives.
The problem is easy to fix, but that's if you _know_ about it. I think most of users are not aware of it.
Now to the point of this message and why it is on this mailing list. What is the best way to handle this?
There is a nice script which automatically tunes HDD options to fix this depending on HDD and/or laptop model. http://git.kernel.org/?p=linux/kernel/git/tj/storage-fixup.git
Would it be okay if we package it and put it in core & install CD? Would it be enough to put this information in Official Install Guide? Would it be better to not care that much about this, and only mention it in some wiki page plus link in (Post-)Install Guide?
I am really not sure what is the best way here. I'd like to hear your feedback on this.
+1 for spreading more the noise about this. This was the opinion I expressed in this old thread : http://bbs.archlinux.org/viewtopic.php?id=39258
Anyone else has some opinion about how to handle this?
I'd like to affirm the opinions of Roman and Xavier and take some action on this. Anyone object to my putting storage-fixup in [extra] at least? If no objections by W 9/21, I plan to go ahead with that step. If it works out, we can talk about follow-up steps like: 1) moving it to [core] 2) integrating it into default rc.d scripts It's a pretty serious issue for laptop users with affected drives. And the drives are pretty popular ones, methinks. I'm usually an Arch purist, and believe in not doing much for users aside from what they explicitly set up themselves-- but when it risks hurting their hardware, I revise my position slightly and want to have a discussion about doing something reasonable. - P
participants (5)
-
Paul Mattal
-
Roman Kyrylych
-
Thayer Williams
-
Thomas Bächler
-
Xavier