Developer comments
This *bug* is now fixed, see the bug report.
But remember : keep bug reports on Launchpad, the Ubuntu bug tracker; Ubuntu Brainstorm is only for ideas.
This bug, submitted at the launch of Brainstorm in feb 2008, was not closed since the guidelines were not set at this time.
1534
votes
|
|
1660
0
126
|
|
|
|
|
Attachments
Duplicates
Comments
|
|
|
It happens on my desktop also. It is really a bit worrying.
|
|
|
|
If would prefer setting it up some reasonable defaults during installation. Another possibility is to show user some warning and provide him a configuration option to change it.
|
|
|
This happens to me on a HP Pavilion dv6449us laptop. I have
had to stop using Ubuntu because of this. I hope that it can be fixed quickly so that I can return.
|
|
|
@warrencarl
There are tutorials out there that fix this, but you have to disable advanced power management.
|
|
|
|
Yes, I have seen many tutorials. Most are followed by comments that they don't work or have certain problems. Is there an official Ubuntu fix? If not, could someone please direct me to the best of the many fixes?
|
|
tchough
wrote on the 29 Feb 08 at 16:53
|
|
|
My understanding is that this problem only occurs when laptop_mode is enabled. Fortunately, laptop_mode is not enabled by default on Ubuntu.
Only if a user goes out of their way to enable laptop_mode will they encounter a problem.
|
|
fria
wrote on the 29 Feb 08 at 21:49
|
|
|
|
I don't want to have to go to the command line to fix a problem that the OS has created. Neither does my grandmother. It's great that this OS is very configurable, but it is also supposed to be an alternative to Windows.
|
|
|
This was actually documented by the Gnome project years ago when there was discussion about integrating advanced hdd power saving functionality to gnome.
The modern drives are very smart and attempting to apply more power saving to them actually mostly just increases the power usage. You really shouldn't have enabled anything extra in the first place.
|
|
feNNec
wrote on the 1 Mar 08 at 12:28
|
|
|
|
I'm worried too; I got a laptop hard drive showing smart errors already because of that!
|
|
|
I believe they've already fixed this for 8.04 (I think).
But yeah, this is very worrying if it hasn't been fixed. We don't want Canonical to get any law suites.
|
|
laga
wrote on the 2 Mar 08 at 13:20
|
|
|
Yes, it'd be great to have that fixed.
There are some settings in /etc/default/acpi-support which seem to be related to this problem. When I tried to use them a few months ago on Gutsy, they didn't seem to affect the shutdown/idle time. Can someone confirm this?
|
|
|
I have quite a bit about that issue in launchpad and it's NOT a trivial issue to fix. Some HDD / Bios ship will inappropriate default and it's not only a Linux problem.
However something really need to be done to make sure those HDD won't get broken because of the load cycle... if you can avoid it nobody want dead hard drives.
A general HDD monitory tool might be a good idea. Warning the user there's a problem and maybe propose a way to stop the insane head parking that happen on some configuration
|
|
|
|
This is cruical! An operating system that in essence destroys hardware will never be popular.
|
|
mbv
wrote on the 2 Mar 08 at 16:26
|
|
|
At my laptop, with laptop-mode enabled, I changed the APM settings to 253, and it still would spin-down after 1 minute (which is usual while browsing). I guess the reason of this is that the hardware (HDD) controls spin-down using APM.
However, it would be way smarter to predict disk-usage using more info than just disk-reads: while moving the mouse around, and typing stuff, you should never spindown.
|
|
Aleks
wrote on the 3 Mar 08 at 02:21
|
|
|
|
This bug has been "critical" for a few months now. It's absolutely ridiculous. This needs to be fixed.
|
|
|
This is one of the things that may stop potential users from trying this out on their laptop. It's a very worrying issue and does need to be resolved without any further delay, especially in an LTS.
Fix this issue and [Ubuntu] Linux might start getting a much bigger slice of the pie.
|
|
stevec
wrote on the 5 Mar 08 at 07:08
|
|
|
|
Already voted for, but want to comment to agree that issues which affect hardware life are critical. I've read about this issue before and I was originally under the impression it affected desktops too. Reading the comments here I'm hoping that's not the case. While I know how to resort to tutorials and hacks on the internet to fix problems like this, I don't like that I have to.
|
|
jespdj
wrote on the 14 Mar 08 at 19:39
|
|
|
|
Although this is not really a bug in Ubuntu (see the discussion in the bug report), I agree that this is a critical issue for which Ubuntu needs to provide a solution.
|
|
buggyman
wrote on the 21 Mar 08 at 05:30
|
|
|
dito, annoying clacks and the fear of killing my hard drive.
I did
/dev/sda
apm = 255
spindown_time = 0
as a default - consequence is loss of power saving capabilities in favor of prolonging my hdds lifespan...
|
|
|
ubuntuforums thread :
http://ubuntuforums.org/showthread.php?t=591503
unofficial ugly fix :
http://ubuntuforums.org/showpost.php?p=3675960&postcount=26
Don't apply any unofficial ugly fixes unless you understand what you are doing and you understand how to revert. Only apply this fix if you are heavily affected. After applying this fix keep an eye on your Load_Cycle_Count and on your harddrive temperate and make sure it remains below the maximum temperature specification of your harddrive. Everything you do is on your own risk.
|
|
|
Turning off power management of your notebook HD is NO GOD IDEA as a general advice! Stop telling Linux beginners to do that - they might fry their drives! It is irresponsible to post such advices in the forums!
I know what I am talking about since my blog is currently being operated from a notebook - I wasted a HD by following this -B 255 talk, it died within 2 weeks after having turned off HD APM because of a broken ball bearing.... and boy it made some really ugly noises before it died.....
There is no simple solution, as the same APM parameter has completely different results on the various HD drives. Also, there are different scenarios, and in some of them you actually want to have agressive APM setting in order to save maximum power.
What you guys want is something like a squared circle or getting washed without getting wet. It is not going to happen wihtout a HD database and/or vendor support, as all approaches have to respect different situations and different hardware.
The best way currently for some drives seems to be a HD management via good old laptop-mode and low APM values while on AC and more agressive settings while on battery. If you can combine this with a HD drive database, maybe you get what you want. But will not get a "one fits it all" solution.
Read this article here for some more information and a flexible "solution" for some Fujitsu HD drives, if you are new to the issue. And don't switch to Vista because of this issue - Vista seems to have the same problem on many drives.
http://vale.homelinux.net/wordpress/?p=199
|
|
steve196
wrote on the 21 Mar 08 at 21:13
|
|
|
For me hdparm -B 255 /dev/sda does not even change anything. The drive still spins down and restarts every few minutes when on battery. And the whole thing happens, while ACPI is disabled due to the age of the BIOS. Also WinXP on the same computer does NOT have an issue.
This is a big issue directly but even more of an issue indirectly.
The first contact with Ubuntu is always by "trying it out". If a cost in hardware wear and tear is attached to that, people won't do that.
Also it could lead to Ubuntu-exceptions on warranties, which is not the kind of publicity, that is needed. ( and it could give service personnel an easy way out of fixing your computers, even if such an exception wasn't explicitly stated)
|
|
flitzjoy
wrote on the 26 Apr 08 at 04:52
|
|
|
|
Anyone please can confirm if this is fixed or not in hardy.
|
|
AbtZ
wrote on the 5 May 08 at 17:19
|
|
|
|
It is not fixed. My Load_Cycle_Count increases rapidly, at approx 4-5 per minute.
|
|
Hammer89
wrote on the 26 May 08 at 03:00
|
|
|
I just finished a "fix" for my laptop with this issue... but now my HD is running hotter than before (46 degrees Celsius compared to 38 degrees Celsius before)... still within the operational temp range... but it's a little unsettling. My load cycle is much better now though... I was at 11 load cycles every 26 minutes before... after the fix I haven't even had one load cycle for multiple hours.
This should definitely be a high-priority task for the Ubuntu dev's though.
|
|
muteXe
wrote on the 30 Jun 08 at 12:24
|
|
|
"And don't switch to Vista because of this issue - Vista seems to have the same problem on many drives."
i dual boot vista and ubuntu 8.04. There is no problem with my drive in vista, but there is in 8.04. How can this be a HD problem and not a OS problem?
|
|
|
|
tchough, that is absolutely incorrect, this problem can occur with or without laptop_mode enabled. In fact, one of the most common workarounds for this problem is to enable laptop_mode with certain settings.
|
|
|
|
In fact, it does happen in Vista for me... the only difference is that smart capability of my hard drive is somehow disabled under vista and there's no way for me to stop it. the suggested 'ugly' fix works pretty well in hardy.
|
|
edajai
wrote on the 9 Oct 08 at 05:29
|
|
|
|
it seems the issue has been fixed for the intrepid release
|
|
minisori
wrote on the 13 Oct 08 at 17:42
|
|
|
|
No it is not, at least in my aspire 5930g
|
|
Auzy
wrote on the 13 Oct 08 at 23:20
|
|
|
If it hasn't been fixed, post bug reports, this is the wrong place, because there is no feedback loop anyway.
Enabling AHCI (to give you NCQ) could potentially help a bit too, because it allows the heads to move more efficiently, but never actually checked.
Talking about it here wont help, because developers have no idea about the specs of your computer, cant ask you to do checks, etc.
|
|
|
Everyone, please cast your votes in support of idea #15153 "Canonical And Dell Should Issue Official Statement About Hard-Drive Killer Bug" if you think it is as crucial as I do. It will help eliminate a lot of controversy and serve to reassure users who'd like to install Linux on their laptops.
Thanks.
~
erythrocyte
|
|
tc111
wrote on the 4 Nov 08 at 06:12
|
|
|
This has not been fixed... my stop_start_count, equivalent to the load_cycle_count has started climbing rapidly again since my upgrade to Intrepid.
Thanks,
TC111
|
|
|
|
Latest update: This 'bug' is no longer marked as Fixed on its Launchpad tracker page. Its status is "In Progress" as of today.
|
|
ceap80
wrote on the 10 Dec 08 at 06:20
|
|
|
|
In any case if it's not fixed, it should NOT appear here as "Done!", that is misinformation...
|
|
Endolith
wrote on the 15 Dec 08 at 17:31
|
|
|
This has NOT been fixed. Please remove the "Done" tag. My two-year old drive is at 1,348,046 load cycles
and climbing.
|
Post your comment
|
Ubuntuforums.org thread #591503
