Disable hibernate option by default

Bug #812394 reported by Javier Jardón
434
This bug affects 110 people
Affects Status Importance Assigned to Milestone
Ayatana Design
Fix Released
High
John Lea
Pantheon Session Indicator
Invalid
Undecided
Mario Guerriero
PolicyKit
Invalid
Undecided
Unassigned
policykit-desktop-privileges (Ubuntu)
Fix Released
Wishlist
Martin Pitt
Precise
Fix Released
Wishlist
Martin Pitt

Bug Description

-----------------------------------------------------------------------------------------
TO RE-ENABLE HIBERNATE: If the hibernate option is supported by your system and you need it, you can re-enable it as follows: https://help.ubuntu.com/12.04/ubuntu-help/power-hibernate.html

You may be also interested in subscribing to the bug #882040, which tracks the implementation of a configuration tool to toggle hibernate and suspend.
-----------------------------------------------------------------------------------------

I'd like to propose a change in the design of Power preferences in the GNOME control center: https://wiki.ubuntu.com/Power

Instead of going through the "certification" path, we can simply disable the options about suspend/hibernate if it's not supported on the current hardware (we have API for this in upower).

I think this solution is more scalable, as we do not need to maintain a database with "certified" hardware. Also, we can add a button or a informative text when suspend/hibernate doesn't work to guide the users about what to do.

Javier Jardón (jjardon)
summary: - Disable suspend/hibernate option when they are not supported
+ Disable suspend/hibernate options when they are not supported
Revision history for this message
Sebastien Bacher (seb128) wrote : Re: Disable suspend/hibernate options when they are not supported

The "disable suspend unless it has been certified to work" on that specification doesn't seem really practical indeed, suspend works on most laptop configs why would it be disabled?

Changed in gnome-control-center (Ubuntu):
importance: Undecided → Wishlist
Revision history for this message
Martin Pitt (pitti) wrote :

Indeed, I see absolutely no reason to disable suspend on unknown machines. It works on the vast majority of hardware out there, and if it doesn't, then you'll learn to "just don't do that" on that particular machine. But it seems wrong to arbitrarily and unnecessarily break such an important functionality.

Disabling it on known-broken machines indeed seems to be a better way to go, although it'd be even better to actually fix it on these machines :) For many older machines we can apply quirks in pm-utils.

Such a blacklist would be most suitable to maintain in pm-utils, from where upower will pick it up through pm-is-supported reporting False for the respective operation (suspend/hibernate).

Revision history for this message
Matthew Paul Thomas (mpt) wrote :

Javier has asked me to comment on this. But my role here was only to specify how the policy should be presented in the power settings. John Lea proposed the policy and Mark Shuttleworth reviewed it.

As John has explained it to me, the logic behind the new policy is this:
1. In Ubuntu, Suspend and Hibernate often don't work on new hardware models without software fixes.
2. The nature of Suspend/Hibernate bugs is that while (as Martin points out) they can be fixed for individual machines, the only feasible way to prevent them from occurring, again and again with new hardware models, is for hardware to be certified.
3. For Ubuntu to present a Suspend function that doesn't work is unprofessional, and presenting a Hibernate function that doesn't work (and destroys data by never waking up) is even worse.
4. It is feasible for Ubuntu to tell whether the hardware it is running on has been certified.
5. Requiring users to understand that Suspend and Hibernate might not work before turning them on is also bad, but not as bad.
6. Suspend and Hibernate do work on many hardware models that have not been certified.
7. From (1), (2), (3), and (4), Suspend and Hibernate should not be available by default for hardware that is not certified.
8. From (5), (6), and (7), Ubuntu should let people turn Suspend and Hibernate back on again.

I can see various ways in which this policy could be shown to be unsound:
* Show that (1) is false. For example, Martin says that Suspend "works on the vast majority of machines". What percentage is that? Does anyone in the world know?
* Show that (2) is false, e.g. "XYZ new standard or technique means that Suspend and Hibernate will almost always work without certification".
* Show that (3) is false or partly false, e.g. "when Hibernate fails, N % of the time it fails before power off, so there's no data loss".
* Show that (4) is false, e.g. "millions of machines asking a server whether their hardware is certified wouldn't scale".

Revision history for this message
Matthew Paul Thomas (mpt) wrote :

Ted Gould has drawn my attention to these:
<http://upower.freedesktop.org/docs/UPower-up-client.html#up-client-get-can-suspend>
<http://upower.freedesktop.org/docs/UPower-up-client.html#up-client-get-can-hibernate>

And Javier has pointed out that as of today (!), these functions are used by gnome-control-center to disable the Suspend and Hibernate functions if appropriate: <http://git.gnome.org/browse/gnome-control-center/commit/?id=4f08a32570490f837e7c6909b095d2b44f4c05ee>

So an updated way of wording (1) above would be: "On new PC models, up_client_get_can_suspend() and up_client_get_can_hibernate() often return true when they should return false, leading Ubuntu to present Suspend and Hibernate functions that don't work without software fixes."

Revision history for this message
Martin Pitt (pitti) wrote :
Download full text (3.3 KiB)

(1) Of course I don't know percentages, but all of the four laptops that I had so far, and the ~ 5 laptops of other people that run ubuntu or that I tried a live system on had no problem with suspend. I didn't hear a lot of complaints about suspend being broken on UDSes or sprints. That's of course far from representative, but the best I can offer as knowledge. Also, we don't get any bug reports about suspend breakage at upstream pm-utils any more for a year or two; we used to get many, but with the advent of KMS and more drivers supporting it it has become pretty much a non-issue. (This partly ties into (2), too).

What does remain as a complaint that I did hear on UDS was that on some models suspend draws a lot of battery live, due to BIOS or driver bugs. This issue remains, of course.

(2) I seriously doubt that white/blacklisting for hibernate would make any improvement. Most hibernation failures I ran into were due to too little swap space (a problem that should not occur any more since lucid or so, since pm-utils carefully checks swap), unusual partitioning (using LVM, or unsupported crypto schemes). It is very conceivable that e. g. USB devices don't properly restore their state on resuming either. However, none of these problems are tied to the hardware platform itself. Unlike suspend, the process of hibernation is actually quite hardware independent. Hibernation is essentially writing the RAM and state to disk, and for resuming you just to a regular boot all the way until initramfs.

(3) I agree to the statement, I disagree that maintaining a whitelist and only enabling models on it is the, or even a good solution.

(4) Yes, that's feasible.

(5) I see little value in lecturing everyone about the potential failure. On the machines where it works it is confusing at best, and on the machines where it sometimes fails it doesn't help you (as we can't predict when it fails), and on machines where it doesn't work at all, users will quickly learn to "don't do that then" anyway. The second case (works most of the time) is certainly the most frustrating one, but again isn't helped by a whitelist as even on certified models it just tends to fail from time to time (like on many Thinkpads).

My feeling about this is:

 * Suspend: I don't think a whitelist is very helpful here, but if we want to have one, it seems fine to me. I strongly disagree about disabling it by default on non-whitelisted models, though. It doesn't make sense to punish the majority of users on uncertified hardware where suspend is working, just because it doesn't or might not work on some models. If at all, we should blacklist these.

 * Hibernate: As explained above, I don't think a white/blacklist makes sense here. We should enable or disable it by default, but this shouldn't be model specific. (in other words: it sucks on all computers). I have no strong opinion about disabling it or not, and would be fine with either.

As for the can-suspend/can-hibernate properties and the recent g-c-c patch: These flags have been there for a long time, and earlier Ubuntu versions used them as well. (Like the session indicator, gnome-power-manager, etc.). The cited commit just fixes a ...

Read more...

Revision history for this message
Ted Gould (ted) wrote :

I don't mind taking that approach on suspend. But every time I've suggested that we enable suspend by default I've gotten push back from everyone saying that it's not reliable enough. Either it is, and we can set up a 10 min suspend by default or it isn't and we shouldn't ask users to do it either.

Revision history for this message
John Lea (johnlea) wrote :

@Ted; both suspend and hibernate need to be opt-in, e.g. if we are unsure they are disabled by default, if we are explicitly sure that they work on a specific computer, only then is it enabled by default. If they are disabled by default a user can re-enable them at any time at their own risk as shown in https://wiki.ubuntu.com/Power#Power_settings

Revision history for this message
Martin Pitt (pitti) wrote :

> @Ted; both suspend and hibernate need to be opt-in, e.g. if we are unsure they are disabled by default

With my desktop TL hat on, I veto this. See comment 5 for the details, which haven't gotten any response yet.

Disabling a working suspend functionality on a lot of hardware out there just because we don't know about it is providing a very poor service to our users. Conversely, enabling it on known boxes is still not going to magically fix the bugs which break suspend/hibernate in some cases. We need to get the actual bugs fixed!

Would you stop shipping GNOME, Evolution, or Firefox, just because they have some bugs which sometimes crash them in some circumstances?

Changed in gnome-control-center (Ubuntu):
status: New → Won't Fix
Revision history for this message
Jeremy Bícha (jbicha) wrote :

John, on what hardware is suspend broken?

I confirm pitti's observations that I haven't seen a computer fail to suspend and recover from suspend in quite a while. I believe we have been shipping Ubuntu for years to suspend by default on battery power after a reasonable idle time; therefore I don't see this as a real concern.

Hibernation of course is quite different and still broken for too many machines.

Revision history for this message
Ted Gould (ted) wrote : Re: [Bug 812394] Re: Disable suspend/hibernate options when they are not supported

On Thu, 2011-08-18 at 09:57 +0000, Martin Pitt wrote:
> With my desktop TL hat on, I veto this. See comment 5 for the details,
> which haven't gotten any response yet.

So do you feel we can assume that on all machines suspend works?

Revision history for this message
Martin Pitt (pitti) wrote :

Ted Gould [2011-08-18 14:13 -0000]:
> So do you feel we can assume that on all machines suspend works?

It certainly doesn't, but at least it "works" on far more machines
than where it doesn't. Reportedly suspend sometimes fails randomly,
when you have particular peripherals attached, or suspend too many
times, etc.), but that wouldn't be helped with a by-machine whitelist
either.

Revision history for this message
Ted Gould (ted) wrote :

On Thu, 2011-08-18 at 15:55 +0000, Martin Pitt wrote:
> Ted Gould [2011-08-18 14:13 -0000]:
> > So do you feel we can assume that on all machines suspend works?
>
> It certainly doesn't, but at least it "works" on far more machines
> than where it doesn't. Reportedly suspend sometimes fails randomly,
> when you have particular peripherals attached, or suspend too many
> times, etc.), but that wouldn't be helped with a by-machine whitelist
> either.

I guess the point here Martin is that you're saying "we don't know" --
and that leads to very confused designs that say we don't know to the
user, and thus the user doesn't know. Then they have no confidence in
us to tell them anything!

I think that we need to come up with a "yes" or "no" answer to the
question: "Does suspend work?" Or "yes, with a blacklist" or "No, with
a whitelist" would also work.

Do you think this would be a good thing to put on the technical board's
agenda?

Revision history for this message
Martin Pitt (pitti) wrote :

Ted Gould [2011-08-18 20:27 -0000]:
> I guess the point here Martin is that you're saying "we don't know" --
> and that leads to very confused designs that say we don't know to the
> user, and thus the user doesn't know. Then they have no confidence in
> us to tell them anything!

I'm afraid I don't understand what you mean here by "us tell them
anything" -- you mean if we had a whitelist of certified models, and
suspend would still fail because of a special case/USB device/etc.,
they would not know?

> I think that we need to come up with a "yes" or "no" answer to the
> question: "Does suspend work?"

Depends what "yes" is -- "always, 100%" is probably "no" everywhere,
as both hardware and to a bigger degree software have a tendency to
deterioate after some time of running (bugs corrupting the memory
state, memleaks), and we can't ever test suspend with all USB or
bluetooth hardware plugged in in the world.

I think "yes" should be "good and reliable enough to be useful", which
seems to be the case on most platforms these days? This could be
measured with some synthetic test cases like "survives 20 suspends in
a row after a clean boot" and "survives a 24 hour suspend cycle".

> Or "yes, with a blacklist" or "No, with a whitelist" would also work.

A blacklist might help, but I don't think that the immense effort in
keeping it up to date justifies the little gain that you get from it
(it would be better to spend that time figuring out pm-utils quirks to
fix it, etc.)

> Do you think this would be a good thing to put on the technical board's
> agenda?

If John and/or you still want to disable suspend by default, then the
TB is the appropriate forum to decide about this, indeed.

Thanks,

Martin

--
Martin Pitt | http://www.piware.de
Ubuntu Developer (www.ubuntu.com) | Debian Developer (www.debian.org)

Revision history for this message
John Lea (johnlea) wrote : Re: Disable suspend/hibernate options when they are not supported

Ok, I have discussed this with Jason W and I think we have a final decision on this item for Oneiric:

1. Suspend is enabled by default on all computers. A option to disable Suspend is shown in the power indicator, see https://wiki.ubuntu.com/Power

2. Hibernate is disabled by default on all computers, unless the computer is on a whitelist. The whitelist should include all "Ubuntu Certified" ( http://www.ubuntu.com/certification/ ) computers on which hibernate is tested regularly and proven to work. Perhaps in the future this list could also be augmented by a community certification process? A user can re-enable hibernate at any time by going to the power settings, as shown in https://wiki.ubuntu.com/Power

cheers,
John

Revision history for this message
Martin Pitt (pitti) wrote : Re: Disable hibernate option when it is not supported

Retitling and reopening as per previous comment.

However, I still don't understand the rationale for having a machine type whitelist.; see (2) in comment 5. Do we have hard data that hibernation is consistently broken on some particular hardware, and the reason isn't just partitioning?

My gut feeling is that we should just disable hibernation by default, period (and perhaps provide an Ui configuration bit to enable it). Maintaining and checking a whitelist will just cause a lot of unnecessary effort which won't actually help a lot.

summary: - Disable suspend/hibernate options when they are not supported
+ Disable hibernate option when it is not supported
Changed in gnome-control-center (Ubuntu):
status: Won't Fix → Incomplete
John Lea (johnlea)
Changed in ayatana-design:
assignee: nobody → John Lea (johnlea)
importance: Undecided → High
status: New → Fix Committed
tags: added: udo
removed: needs-design
Revision history for this message
Ted Gould (ted) wrote : Re: [Bug 812394] Re: Disable hibernate option when it is not supported

On Mon, 2011-08-22 at 11:21 +0000, Martin Pitt wrote:
> My gut feeling is that we should just disable hibernation by default,
> period (and perhaps provide an Ui configuration bit to enable it).
> Maintaining and checking a whitelist will just cause a lot of
> unnecessary effort which won't actually help a lot.

The question becomes: how do HW vendors that *do* care about Ubuntu
enough to ensure that hibernation works on their system express that so
that we know we're on one of those systems?

I think that we'd like them to be able to self select by adding
themselves to some sort of list.

Revision history for this message
Martin Pitt (pitti) wrote : Re: [Bug 812394] Re: Disable hibernate option when it is not supported

Ted Gould [2011-08-25 15:41 -0000]:
> The question becomes: how do HW vendors that *do* care about Ubuntu
> enough to ensure that hibernation works on their system express that so
> that we know we're on one of those systems?

Again, hibernation often doesn't fail because of a particular system,
but because of how you set it up, use it, and which peripherals you
connect. So what does that "works" mean in that case? It works "for
some cases with a default partitioning and no extra devices attached
after a clean boot"? That should be the case for pretty much any
system these days.

I thought we wanted to disable it because it's utterly slow, and
confusing to have two ways of sleeping. And that isn't hw dependent
either?

Revision history for this message
John Lea (johnlea) wrote : Re: Disable hibernate option when it is not supported

When OEM role a custom Ubuntu release for new hardware they will be able to include the new hardware on the white list. And all 'Ubuntu Certified' hardware (where it is testing and know to work) will also be included, so I think we are good here for now. It would also be great to have a community hardware certification process in the future.

John Lea (johnlea)
Changed in ayatana-design:
status: Fix Committed → Fix Released
Changed in gnome-control-center (Ubuntu):
assignee: nobody → Canonical Desktop Team (canonical-desktop-team)
status: Incomplete → Triaged
Revision history for this message
Martin Pitt (pitti) wrote : Re: Disable hibernate option when it is not supported

/var/lib/polkit-1/localauthority/10-vendor.d/com.ubuntu.desktop.pkla already has a commented out stanza how to disable this. Easiest to disable it there.

affects: gnome-control-center (Ubuntu) → policykit-desktop-privileges (Ubuntu)
Changed in policykit-desktop-privileges (Ubuntu):
assignee: Canonical Desktop Team (canonical-desktop-team) → Martin Pitt (pitti)
Revision history for this message
Martin Pitt (pitti) wrote :

That wouldn't include a black/whitelist, of course; I don't believe that this is going to work anyway. OEM installs can modify the .pkla to turn it back on if desired.

summary: - Disable hibernate option when it is not supported
+ Disable hibernate option by default
Changed in policykit-desktop-privileges (Ubuntu Precise):
status: New → Triaged
importance: Undecided → Wishlist
assignee: nobody → Martin Pitt (pitti)
John Lea (johnlea)
tags: added: udp
Changed in ayatana-design:
status: Fix Released → Fix Committed
John Lea (johnlea)
Changed in ayatana-design:
importance: High → Critical
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package policykit-desktop-privileges - 0.8

---------------
policykit-desktop-privileges (0.8) precise; urgency=low

  * Disable hibernation by default. (LP: #812394)
 -- Martin Pitt <email address hidden> Thu, 20 Oct 2011 11:49:35 +0200

Changed in policykit-desktop-privileges (Ubuntu Precise):
status: Triaged → Fix Released
John Lea (johnlea)
Changed in ayatana-design:
importance: Critical → High
Nick Tait (jnick-tait)
Changed in ayatana-design:
status: Fix Committed → Fix Released
tags: added: reviewedbydesignp
removed: oneiric udo udp
Revision history for this message
Maxim Levitsky (maximlevitsky) wrote :

And I almost filled a bugreport with KDE about this.
What a bunch of a****, there are just no more words for this.

Revision history for this message
Jeremy Bícha (jbicha) wrote :

Maxim, please calm down; it's just software. Anyway, take a look at
http://askubuntu.com/questions/94754/how-to-modify-policykit-to-allow-hibernation-in-upower

Revision history for this message
Maxim Levitsky (maximlevitsky) wrote :

True, but you are just keeping of harming already broken system.
There is *lot* of bugs.
I already enabled hibernation, I am smart enough to Google and find what is going on eventually.
Its just stinks when you just remove/disable more and more options instead on focusing on usability.

Usability? Tell me please how I can take few images and stitch them together in Gimp?
Even in MS paint its easy.

Ubuntu is full of glaring usability problems, I don't even complain about, and all developers of it are busy removing features.
And KDE is even worse (but at least it has the features).

I understnd very well that ubuntu and all free software it distrubutes is free, and I didn't yet paid a dime for using it.
So I never would complain about lack of some feature, etc.
But when you remove already existing features, make already good product worse, I think I do have right to complain.
Besides I did contribute a bit by writing few kernel drivers, fixing a bunch of bugs (including many bugs that caused unreliable suspend on my system... the irony...)

https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.22/+bug/160763

Good, keep up the good work, I already mostly use Win7.

Revision history for this message
Phillip Susi (psusi) wrote :

It sounds like the plan was to disable by default, but still have a simple, easy to discover option to turn it back on. If this is not the case, this bug needs reopened until it is added, or enabled by default again.

Disabling hibernation by default because it is less used than suspend and having two options is a little confusing is fine, but it should be easy to discover how to turn it back on ( meaning: without having to dig around in gconf ). I still don't understand this sentiment that it works less often than suspend; I have always found the reverse to be true, which makes sense since it doesn't rely on the machine's acpi bios which is sometimes broken.

Revision history for this message
ibidem (ibid-ag) wrote :

On my Aspire One, hibernate works as soon as I add ath_pci & co. to PM_MODULES (IIRC). Suspend, however, is quite broken.

Of course, I'm using pm-hibernate instead of gnome/policykit; and I have to use a local dkms package of the (unmaintained) madwifi-hal 0.10.5.6 branch because every other wireless driver does not work (1 option out of four drivers...)

And I haven't bothered testing either hibernate or suspend, especially with Ubuntu, on my new laptop, a Thinkpad X100e. Judging by this, I won't--because I don't care for a distro like this.

Revision history for this message
RussianNeuroMancer (russianneuromancer) wrote :

I can understand disabling hibernate by default for Ubuntu or Gnome3-based distro, but I can't understand why that disabled for Kubuntu. KDE users expect more features, and broken by default hibernate functional is simply harm to users who have hardware where hibernate is works.

I ask developers look into this issue again and change decision, at least for Kubuntu.

Revision history for this message
Michel-Ekimia (michel.ekimia) wrote :

Disabling hibernation is a good choice, but why not disabling suspend also ? on UX31 it corrupts memory for example.

ubuntu should provide a graphical way for a user to Reenable Hibernation with a warning message about the probability of failure.

You can't ship 12.04 without this IMO.

Martin Pitt (pitti)
Changed in policykit:
status: New → Invalid
Revision history for this message
Nils Rennebarth (nils-rennebarth) wrote :

Yes, please provide an easy to find method to switch it on. It took me some googling to find how
to enable it, and it was not obvious at all, that it was intentionally disabled. It looked like a bug
(on xfce, a window pops up: not authorized)

And although I count myself as reasonably computer literate, I know next to nothing about policykit et al,
and would never thought by myself to edit /var/lib/polkit-1/localauthority/10-vendor.d/com.ubuntu.desktop.pkla

Revision history for this message
Laurent Dinclaux (dreadlox) wrote :

Let the option available in energy settings, with the hibernate disabled on not certified hardwares, and enabled on certified one.

Indeed ....

(How can this bug be solved ? A grayed out option with no notice can't be a solution.)

Revision history for this message
Phillip Susi (psusi) wrote :

There is no certified hardware because hibernation is not hardware specific and maintaining such a list would be a great deal of effort for little to no gain, and never be complete. All that is needed is a simple check box in the power settings to enable hibernation, that is all.

Revision history for this message
Cas (calumlind) wrote :

Although I can see the reasoning to disable by default I find the lack of any trivial way to revert ridiculous. My Dell Precision used have issues with hibernation until it was fixed in Oneiric but now in Precise a working feature is disabled by default! If my battery is running critically low, powering off the system rather than hibernating is a major pita.

Revision history for this message
quadra (info06) wrote :

Agree with 2 previous posts.
Disabling hibernation is an effective way to destroy our planet faster.
Thousands of ubuntu users will leave their PC running all day and waste energy.

It's good that standby is still enabled.
But please, please provide a user-friendly way to enable hibernation from the GUI...

In xubuntu / xfce, as Nils wrote, we get an ugly error pop-up complaining about permissions. And hibernation is not grayed out.

Suggestion
1) re-enable hibernation by default
2) upon first hibernation, display a warning:

===========
Warning:
Hibernation may fail if your hardware is not _certified_ [weblink], or if you use incompatible peripherals.
Link: [Enable / disable hibernation]

[x] Show this warning next time
===========

Revision history for this message
Hermes (PL) (hermes85pl) wrote :

I totally agree with those that complain about the decisions that were made regarding the options to hibernate and suspend. I also find what Maxim Levitsky said in post #22 completely absolutely justified. It is really sad when some people irresponsibly take decisions and break things that work.

Nevertheless, I see no reason why not to put an option to enable/disable either hibernation or suspending to RAM in the settings. Usability first.

Revision history for this message
papukaija (papukaija) wrote :

I've opened bug 976654 asking developers to allow people to re-enable hibernation from the GUI or even make it default again. Please use that bug's "this bug affects me too link" to make the change to happen.

Revision history for this message
Leuke (leuke) wrote :

Though I understand the reasons to disable hibernation by default, I strongly disagree with this decision.
This feature is simply too important to disable it with no option to easily restore it.
When I first tried the beta of 12.04 and couldn't find hibernation where it is supposed to be, I thought it was a bug. Then I realized that it was done on purpose and I was really negatively surprised. Now, I work with computers so googling for a solution was enough for me, but think about regular users: they won't find a really useful feature that competitors provide (Microsoft Windows for example, but of course also other Linux distributions).
I never had problems with hibernation, but the rare cases where it causes problems shouldn't be enough to disable it for all the users. In that cases as Martin said about suspend, people will learn to "just don't do that"!!
In my opinion the decision should be reconsidered but if it is not possible to enable hibernation by default for the reasons discussed here, please provide an easy graphical way to re-enable it as suggested in the previous comments.

Revision history for this message
Jeremy Bícha (jbicha) wrote :

While I agree that re-enabling hibernate should probably be more visible, that's a separate bug that probably won't be fixed for Ubuntu 12.04.

There is a help page in the default pre-installed help entitled "How do I hibernate my computer?" that explains how to test if hibernate works and how to re-enable it manually.

One reason why the bug is unlikely to be fixed is that it would require a User Interface Freeze Exception, which would require a Docs Freeze Exception and the translators are nearly out of time to translate Ubuntu 12.04 before release as it is.

Revision history for this message
Eduard Gotwig (gotwig) wrote :

My toshiba notebook supports hibernate and is in the ubuntu friendly database, but I see no hibernate function - WHY ?...

Revision history for this message
Phillip Susi (psusi) wrote : Re: [Bug 812394] Re: Disable hibernate option by default

On 4/10/2012 3:37 PM, Eduard Gotwig wrote:
> My toshiba notebook supports hibernate and is in the ubuntu friendly
> database, but I see no hibernate function - WHY ?...

Reading ( or even just glancing at ) this bug before commenting on it
would answer that question.

Revision history for this message
Eduard Gotwig (gotwig) wrote :

I realy have to say that I cant believe what the arguments to mark this "bug" as a feature are.

Power to the people, not to hibernate.

I think you know the rest.

' I also find what Maxim Levitsky said in post #22 completely absolutely justified '

And that for a LTS release...

Revision history for this message
SoloTurn (soloturn) wrote :

It makes me sad that you just disabled this. Either you underestimate the number of people of users using Ubuntu, or you overestimate the few bug reports you got about it.

Revision history for this message
Roshan (rospkos-07) wrote :

it is too late as the release is one week away.
My suggestion is that after installing the ubuntu, a program similar to Kaptan that would provide the option to enable hibernate.
it is just the step to usability fix. the program could make a warning message if the machine is not listed in the white-list. For 12.10 we need to be effective in the validation of machines for enabling hibernate/suspend.

Disable Hibernate option is somewhat a 50-50 decision and it must be rather unfortunate. I hope the hibernate bug is fixed in 12.10 version or precise-update.

Revision history for this message
epek (epek) wrote :

I am testing Precise for about two weeks now. I just found out about disabled hibernation on UL30A.
Do you consider it a wise decision, to disable a vital function on some machines without informed consent - especially on a LTS edition? I wouldn't.

Revision history for this message
epek (epek) wrote :

Latest test: On that specific laptop Asus UL30A, hibernation works well (but a little slow compared to the formerly used tuxonice) with the latest set of packages, if you uninstall hibernate and uswsusp.
With hibernate installed it will hang on restore.

Revision history for this message
Bob Bib (bobbib) wrote :

It would be wise to rollback this change.

Revision history for this message
Jose Bernardo (bernardo-bandos) wrote :

On my Toshiba NB550D, hibernate always worked, and suspend needs a grub option (acpi_sleep=s3_bios). Also, as I use KDE, the only symptom I've had of this "feature" after installing the precise beta was that the "suspend to disk" option in the shutdown menu suddenly only locks the screen and nothing else. At least I know what I am doing and after some googling I was able to find out this has been done on purpose and wasn't some bug. To that I can only add:

' I also find what Maxim Levitsky said in post #22 completely absolutely justified '

description: updated
Bob Bib (bobbib)
description: updated
description: updated
Revision history for this message
Tom Chiverton (bugs-launchpad-net-falkensweb) wrote :

The hell ?!? Hibernate has been working well on every Linux machine I've used, for years.
And suddenly it gets turned off by default *even if you were using it before* ?!?
How is this usable. There are now buttons all over the (KDE) GUI that don't do anything. This has to be *worse*. And because pm-hibernate works, it looks like a bug in the GUI...

I could understand making the default for new installs to disable it if there really were serious issues, but I've not heard anyone having any (except with very cutting edge hardware) for years. Even then you should *ask* or *explain* if you are removing functionality that most people expect to have from a modern Linux distribution.

Whitelisting isn't going to work. People will just assume Ubuntu is broken. Or, worse, 3rd party packages like KDE are broken.

Revision history for this message
Tom Chiverton (bugs-launchpad-net-falkensweb) wrote :

Re #37: what UI needs to be changed ?
There isn't any UI exposing this new anti-feature, it just doesn't work.

As to the docs, all you need to do is delete the new page you wrote, no translation overhead required.

Revision history for this message
hoijui (czech-freesurf) wrote :

i am also in favor of it being enabled by default.
you say it is unprofessional if Ubuntu shows the hibernate option but it does not work, though that is exactly what is the case now, just that Ubuntu does not really try to hibernate. for the user, these are both the same: "does not work!". sure... there are scenarios where you would loose data in one way and not in the other, but such scenarios can be drawn either way. for example, if i have very low battery, i have to leave, and have lots of apps open.. i click on hibernate, close the lid and off i go... while underway, the rest of the battery gets emptied, and i loose data.

i am using xubuntu, which shows an error message box:
"Failed to hibernate session - not authorized"
... which suggests a system/software configuration error. this error starts appearing after updating to 12.04, which suggest that the update mis-configured the system, which... would be unprofessional. ;-)
i read that on Ubuntu, you do not get any error message, which is even worse.
The least i would expect, is a worthwhile and accurate error message. something like: "there is lots of hardware ... certified ... if your system ... therefore, hibernation has been disabled for your system, to prevent you from using data."
and this error should appear right after clicking on hibernate, and an error sound should appear, and the screen should not be locked.

The best thing you could do, would be to organize a whistle blower at MS, that hands over their list of hardware+fixes for hibernation. if then you only fail to properly hibernate for systems where MS fails too, you are on the safe side.
i would even find it morally just to hack into MS, for getting this list, as it is morally totally wrong and bad for the end user, for hardware creators to help MS but not others in this issue.

Revision history for this message
Adam Porter (alphapapa) wrote :

"One reason why the bug is unlikely to be fixed is that it would require a User Interface Freeze Exception, which would require a Docs Freeze Exception and the translators are nearly out of time to translate Ubuntu 12.04 before release as it is."

Bureaucracy does not behoove Ubuntu. If paperwork and red tape and silly excuses are the direction Ubuntu is going, other distros will be the direction users are going.

Here's a new motto for Ubuntu to follow: "Just fix it."

These poor decisions are going to drive me back to Debian one of these days, I just know it. So many problems, so many bugs, so many regressions--caused not by technical issues but by unwise decision-making. It's such a shame, because they could have been avoided.

There's a bug #1--how about a rule #1: Keep moving forward. Keep making progress. Keep getting better. Stand on the shoulders of the giants--including the previous release. Don't regress.

Bug #1 will never be resolved as long as Ubuntu keeps going back and forth between progress and regress.

Revision history for this message
Martin Pitt (pitti) wrote :

Adam Porter [2012-04-30 15:59 -0000]:
> "One reason why the bug is unlikely to be fixed is that it would require
> a User Interface Freeze Exception, which would require a Docs Freeze
> Exception and the translators are nearly out of time to translate Ubuntu
> 12.04 before release as it is."

That's not really true, too. It is unlikely to be fixed because it is
not a bug in the sense of "oops, that wasn't intended", but it was an
explicit and deliberate design decision.

Martin

Revision history for this message
Adam Porter (alphapapa) wrote :

A new release should be an improvement--an "upgrade" should be an upgrade. If you took your car in for service and found the radio was missing when you picked it up, you'd be very upset. Then imagine the mechanic said, "Well, some people don't like some of the stations it plays, so we ripped it out." Hey, dude--it's my car! Let me leave the radio off if I don't like it. Don't make me jump through hoops to get it back. It's not a perfect analogy, but it's basically the same principle.

I really don't want to see Ubuntu go the way of GNOME, and even Google, recently, by not listening to its users and just doing whatever it wants.

Revision history for this message
Stephen Warren (srwarren) wrote :

If hibernate must be disabled by default (which IMHO is a silly idea, since it makes the whole fact that hibernate even exists completely undiscoverable), then rather than directing users to create a file in /etc (hardly something that the average end-user is going to know how to do), why not create a package that can be installed that enables the option. This would show up in whatever GUI package management tool users get to use these days, with a description like "install this to enable hibernate on your system". That'd be far easier for end-users to do; much more in-line with other tasks they probably already perform.

Revision history for this message
Joachim Durchholz (jo-durchholz) wrote :

Ubuntu currently has two modes of "thinking" about the current hardware's ability to hibernate:
"I know it work, therefore it should be available to the user" and "I know it does not work, therefore it should be hidden from the user".

Ubuntu is missing the third mode:
"I do not know whether it will work, so the user should know the option exists, but be warned that there's a risk involved; also, if hibernating has been known to work, I should record the current hardware as 'I know it works'."

Revision history for this message
Al Markov (info-markoval) wrote :

The first - hibernate mode is 'grey' without any explanation.

The second - my battery live is short and I want to power off or hibernate my notebook by closed lid.

I propose add description with link (https://help.ubuntu.com/12.04/ubuntu-help/power-hibernate.html) to the disabled point and add point (disabled by default?) with 'power off' functionality.

Eduard Gotwig (gotwig)
Changed in indicator-pantheon-session:
assignee: nobody → Mario Guerriero (mefrio-g)
status: New → Confirmed
Revision history for this message
MestreLion (mestrelion) wrote :

What a sad decision for an LTS.

1) Disable hibernate by default: ok, I can understand the many issues around it

2) Not providing a GUI way to turn it back on - BAD

3) Not allowing users to even know that this feature /exists/ - VERY bad

Solutions in #53 and #54 are great:

A) Provide a package in Software Center to re-establish the Hibernate option in the UI. Warning messages like "this may not work for you" is fine. #33 suggest a /very/ nice example.

B) If white/black lists are maintained, allow machines to "certify themselves" (or blacklist themselves): After a (or N) successful hibernate/resume cycle(s), send data to servers.

Revision history for this message
Dv8tion (dv8tion) wrote :

Lenovo T61 - Windows7, Hibernate mode works fine. Suspend Works fine and has for the last 3 years.
Lenovo T61 - Suddenly 12.04 comes along and apparently that laptop is not good enough for hibernate, even though I carefully created a swap drive large enough for such things. Hrrm. Suddenly I can't just just close the lid because I have to make damn sure all my docs are shut because the battery might not last. Well, I can if I'm using Windows. Irony, anyone?

On a laptop where every last drop of battery life is necessary, I find this disabling of Hibernation to be pretty lame. Sure it can be enabled but hibernation should be made an option in Power Management. You know, like it used to be.

So count me on the list of people who have laptops who have and need hibernation, but that Ubuntu doesn't play nice with.

Revision history for this message
Marcos Roriz (marcosrorizinf) wrote :

This is just ridiculous. I don't care if you guys DISABLE Hibernate but YOU SHOULD PROVIDE AN OPTION TO TURN IN EASILY. Many peopple are used to hibernate, if they don't see on the menu they just think that it doesn't work.... OMG guys..

Revision history for this message
gatopeich (gatoguan-os) wrote :

Add my to the "this is just ridiculous" list.
The sort of thing that is driving me out of Ubuntu any day now...

Revision history for this message
Nathan Collins (ntc2) wrote :

I strongly support the solution in #33: give a warning when the user uses hibernate, with a link to more information.

Does anyone know if this trend to remove features is motivated by user studies? Or is it just intuition? I find this trend very annoying, but I'm probably not an average user. I switched from Debian to Ubuntu in 2005, because things (e.g. wireless, mplayer (?)) "just worked" out of the box. Probably Debian is much more usable "out of the box" now. Maybe it's time to go back?

Maybe there could a be survey in the distribution upgrade UI, that included questions related to the "simplification" of interfaces? This way we could find out what proportion of users really want their options limited. It might be hard to word this survey in a neutral way, e.g. I might word the question as "do you prefer to be treated like a mentally disabled 8 year old", whereas the people that want to remove the hibernate (and suspend !!!) feature might say "do you prefer that options, which may not work with your setup and could cause data loss if you don't save before using them, are disabled by default?". So, it would make sense to have concrete examples in the survey. E.g.
- the current hibernate example.
- the restriction to a fixed set of zoom levels in evince.
- the newish requirement that you hold a modifier key while right-clicking on the Gnome menu, if you want a menu to pop up.
- the hard-coding, with no option to override, of WIn+P to toggle display settings.

Revision history for this message
Andy (amo999) wrote :

Can someone please make it clear to me if the suspend to disk feature is now stable (fixed and released), or if it's still a general issue and is still disabled by default due to results for differents hardware?

Revision history for this message
Bob Bib (bobbib) wrote :

andygmorris,
hibernate works on most machines, but was disabled to prevent problems with hypothetical newbie users & their buggy hardware;
though it's easy for an experienced user to re-enabled it (look at the bug description).

Revision history for this message
Paddy Landau (paddy-landau) wrote :

This may be a bit late, but...

Have you considered encrypted folders?

Hibernation *must* be disabled with encrypted folders, because the swap is also encrypted.

It is possible to have hibernation with encrypted folders, but there is a caveat: you need a separate password just to boot, and all users need to know it.

https://help.ubuntu.com/community/EnableHibernateWithEncryptedSwap

Revision history for this message
Ubuntu QA Website (ubuntuqa) wrote :

This bug has been reported on the Ubuntu laptop testing tracker.

A list of all reports related to this bug can be found here:
http://laptop.qa.ubuntu.com/qatracker/reports/bugs/812394

tags: added: laptop-testing
Revision history for this message
angelique (ange) wrote :

My suggestion:

Include a lens to find hardware id in certificate database on hibernate and suggest compatible laptops from Amazon!

Revision history for this message
Jordan P (japzone) wrote :

Why the heck the Devs thought it was smart to Disable Hibernate by Default and not include a GUI to Enable it is beyond me. Looking at the Early Comments it seems the Devs didn't think this through and consider what the End-User wants.

How Hard is it to Give us a freakin switch in the Power Settings to Enable Hibernation! It's Not Complicated. Just have a Warning Pop-Up when people click it so that they know it might not work properly. Heck, even mark it "Beta" so people will understand it even more. Just give the average people the option. When I recommend Ubuntu to a Friend and he goes "Where the Heck is Hibernate?", I don't want to tell him "You have to create this file in a system folder and then copy&paste this text into it." He'll think I'm crazy.

Revision history for this message
Paddy Landau (paddy-landau) wrote :

@Jordan P: The hibernate option was turned off by default because of problems associated with it. I don't know why Canonical did not give a simple GUI option to turn it on, but I would imagine that it is precisely because only "advanced" users should do so.

Your answer to your friend could quite simply have been, "There are problems with hibernation partly due to new security procedures and partly due to hardware problems, and it has been disabled as a result." You don't want to tell him about the system folder and cut-and-paste, unless he is sufficiently advanced to cope with any resulting problems — in which case, he wouldn't think you crazy anyway!

Cody Garver (codygarver)
Changed in indicator-pantheon-session:
status: Confirmed → Invalid
Revision history for this message
Paul Ortyl (ortylp) wrote :

As of today all publicly known workarounds (changes in policy file) for this *BUG* stopped working for me on Ubuntu 13.10.
So no hibernation in UI any more. Manual entries in /etc/acpi/* (event and script calling pm-hibernate) do work.

Revision history for this message
Karma Dorje (taaroa) wrote :

#68
Yes, also confirm this.

Revision history for this message
Bob Bib (bobbib) wrote :

IMHO, it's much more productive to open a new bug like "Please enable hibernate in Ubuntu", "Hibernate doesn't work in Ubuntu" etc.
than posting complaints to this one;

the current bug report is a suggestion to DISABLE the hibernation in Ubuntu;

if you make a "YES" vote ("Does this bug affect you?") for this bug, you support DISABLING hibernation in Ubuntu,
and not taking it back to function ;)

----
Best wishes, Bob

Revision history for this message
Bob Bib (bobbib) wrote :

Bugs #882040 & #976654 are right ones, I think.

----
Best wishes, Bob

Revision history for this message
Jérôme (jerome-bouat) wrote :

I don't agree with a default behaviour which disable hibernation by default.

Especially on laptops, the hibernation should be available in order to properly shut down when the battery level reaches the critical limit.

Could you at least keep hibernation by default for laptops ?

Revision history for this message
Paddy Landau (paddy-landau) wrote :

@Jérôme, Regardless of how you go about it, you'd need to enter a complex passphrase each time you start your computer, whether it's a cold boot or after hibernation.

To give a comparison, imagine if Windows introduced a security feature requiring every user to enter a complex passphrase when turning on the computer; furthermore, every user on a single computer would need to know the same shared passphrase (in addition to their own login password). There would be a worldwide revolt.

Ubuntu would have the very same restrictions.

That's why hibernation cannot reasonably be available by default; the majority of Ubuntu users are not technically expert, but rather are "everyday" users and would revolt against this.

Full-disk encryption is available via LUKS (now an option during installation in 13.10) for those who have the need, and hibernation can work then; perhaps it should be available by default, but only when LUKS is being used.

Revision history for this message
Jérôme (jerome-bouat) wrote :

@Paddy, Ok. However, each user can start a new session in addition to the session which is blocked.

Revision history for this message
jhoechtl (johann-hoechtl) wrote :

@Jerome: Paddy was referring to the password when full disk encryption is in use, not the one used to protect a user account. Normally the pass-phrase used for full disk encryption this is an automatically created, awkward and long phrase, hard to type in without making mistakes, let alone memorize.

However, @Paddy, your argumentation is flawed in the sense to assume that full disk encryption necessarily is a sane default and should be forced on a user. True, it is an elegant excuse instead fixing bugs when the kernel is communicating with the BIOS, UEFI or the graphics stack, which are the most occurring reasons for failing resume from S3, S4 suspends.

Revision history for this message
Phillip Susi (psusi) wrote :

Who said anything about full disk encryption? It is completely unrelated to this bug.

Revision history for this message
TomaszChmielewski (mangoo-wpkg) wrote :

I'm using Samsung Ativ 5 laptop.

Unfortunately suspend to RAM does not work here - the laptop does suspend, but never wakes up.

On the other hand, suspend to disk works fine - the laptop does suspend and wakes up correctly.

Leaving a broken suspend to RAM option is a bug in my opinion, especially that the only working option of suspending to disk was removed.

Revision history for this message
Bob Bib (bobbib) wrote :

2014-10-20 from TomaszChmielewski:
> I'm using Samsung Ativ 5 laptop.
> ...
> Leaving a broken suspend to RAM option is a bug in my opinion,
> especially that the only working option of suspending to disk was removed.

1) Please report a new bug about the suspend-to-RAM problem on your system: https://wiki.ubuntu.com/DebuggingKernelSuspend
2) Regarding the hibernation feature, please check the bug #882040; meanwhile, probably you can still use the "sudo pm-hibernate" from console.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.