Show notifications on low disk space.

Show notifications on low disk space.

The notifications are shown at most once per hour, or whenever the severity of the notification changes.

BUG=622590
TEST=./out/Release/unit_tests --gtest_filter=LowDiskNotificationTest.*

Committed: https://crrev.com/0d43514effc1b793aef2794815bc6f5b5e2811db
Cr-Commit-Position: refs/heads/master@{#403342}

[yoshiki, 2016-06-23 08:00:44]

yoshiki@chromium.org changed reviewers:
+ yoshiki@chromium.org

[yoshiki, 2016-06-23 08:00:45]

Thank you for doing this! The direction is good for me. I added some tiny
comments.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/chr...
File chrome/browser/chromeos/chrome_browser_main_chromeos.cc (right):

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/chr...
chrome/browser/chromeos/chrome_browser_main_chromeos.cc:686: new
LowDiskNotification(DBusThreadManager::Get()->GetCryptohomeClient()));
Please free this instance explicitly before DBusThreadManager is free'd.

Maybe we should do it in
ChromeBrowserMainPartsChromeos::PostMainMessageLoopRun()?

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
File chrome/browser/chromeos/ui/low_disk_notification.cc (right):

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:28: const base::TimeDelta
kNotificationInterval = base::TimeDelta::FromMinutes(2);
constexpr is better than const.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:34: :
cryptohome_client_(cryptohome_client) {
I think it's better to use DBusThreadManager::Get()->GetCryptohomeClient()
directly, instead of passing it from the caller.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:35: if (cryptohome_client_)
{
cryptohome_client_ is necessary for this class, so should you add
DCHECK(cryptohome_client_) instead of if-clause?

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:37:
&LowDiskNotification::OnLowDiskSpace, base::Unretained(this)));
You might need to close the notification when the disk gets enough space back?

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:41:
LowDiskNotification::~LowDiskNotification() {
Please DCHECK(DBusThreadManager::Get()->GetCryptohomeClient()) to prevent them
from being destructed in wrong order.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:50: if (severity >
last_notification_severity_ ||
Shouldn't we update the notification content when the severity is changed down
to MEDIUM from HIGH?

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:52:
g_browser_process->message_center()->AddNotification(
You shouldn't pass nullptr to AddNotification.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:66:
l10n_util::GetStringUTF16(IDS_LOW_DISK_NOTIFICATION_TITLE),
Did you forget to add the string grd file to this patchset?

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:73: notification =
message_center::Notification::CreateSystemNotification(
Parhaps, pinned notification is better?
https://cs.chromium.org/chromium/src/ui/message_center/notification.h?rcl=0&l...

fyi, pinned notification is a special type of notification which can't be
removed from the message center by user.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:87:
l10n_util::GetStringUTF16(IDS_LOW_DISK_NOTIFICATION_BUTTON)));
Don't you add a click handler for this button?

https://codereview.chromium.org/2082363004/diff/1/chromeos/dbus/cryptohome_cl...
File chromeos/dbus/cryptohome_client.cc (right):

https://codereview.chromium.org/2082363004/diff/1/chromeos/dbus/cryptohome_cl...
chromeos/dbus/cryptohome_client.cc:1174:
low_disk_space_handler_.Run(disk_free_bytes);
low_disk_space_handler_ may be null so null check is necessary.

[dspaid, 2016-06-27 05:40:05]

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/chr...
File chrome/browser/chromeos/chrome_browser_main_chromeos.cc (right):

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/chr...
chrome/browser/chromeos/chrome_browser_main_chromeos.cc:686: new
LowDiskNotification(DBusThreadManager::Get()->GetCryptohomeClient()));
On 2016/06/23 08:00:45, yoshiki wrote:
> Please free this instance explicitly before DBusThreadManager is free'd.
> 
> Maybe we should do it in
> ChromeBrowserMainPartsChromeos::PostMainMessageLoopRun()?

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
File chrome/browser/chromeos/ui/low_disk_notification.cc (right):

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:28: const base::TimeDelta
kNotificationInterval = base::TimeDelta::FromMinutes(2);
On 2016/06/23 08:00:45, yoshiki wrote:
> constexpr is better than const.

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:34: :
cryptohome_client_(cryptohome_client) {
On 2016/06/23 08:00:45, yoshiki wrote:
> I think it's better to use DBusThreadManager::Get()->GetCryptohomeClient()
> directly, instead of passing it from the caller.

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:35: if (cryptohome_client_)
{
On 2016/06/23 08:00:45, yoshiki wrote:
> cryptohome_client_ is necessary for this class, so should you add
> DCHECK(cryptohome_client_) instead of if-clause? 

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:37:
&LowDiskNotification::OnLowDiskSpace, base::Unretained(this)));
On 2016/06/23 08:00:45, yoshiki wrote:
> You might need to close the notification when the disk gets enough space back?

We don't actually have this information.  Notifications are sent when there is
not enough space, but there's no notification for having enough space.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:41:
LowDiskNotification::~LowDiskNotification() {
On 2016/06/23 08:00:45, yoshiki wrote:
> Please DCHECK(DBusThreadManager::Get()->GetCryptohomeClient()) to prevent them
> from being destructed in wrong order.

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:50: if (severity >
last_notification_severity_ ||
On 2016/06/23 08:00:45, yoshiki wrote:
> Shouldn't we update the notification content when the severity is changed down
> to MEDIUM from HIGH?

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:52:
g_browser_process->message_center()->AddNotification(
On 2016/06/23 08:00:45, yoshiki wrote:
> You shouldn't pass nullptr to AddNotification.

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:66:
l10n_util::GetStringUTF16(IDS_LOW_DISK_NOTIFICATION_TITLE),
On 2016/06/23 08:00:45, yoshiki wrote:
> Did you forget to add the string grd file to this patchset?

They were in a separate change which has now been submitted.
https://codereview.chromium.org/2090913002

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:73: notification =
message_center::Notification::CreateSystemNotification(
On 2016/06/23 08:00:45, yoshiki wrote:
> Parhaps, pinned notification is better?
>
https://cs.chromium.org/chromium/src/ui/message_center/notification.h?rcl=0&l...
> 
> fyi, pinned notification is a special type of notification which can't be
> removed from the message center by user.

Done.

https://codereview.chromium.org/2082363004/diff/1/chrome/browser/chromeos/ui/...
chrome/browser/chromeos/ui/low_disk_notification.cc:87:
l10n_util::GetStringUTF16(IDS_LOW_DISK_NOTIFICATION_BUTTON)));
On 2016/06/23 08:00:45, yoshiki wrote:
> Don't you add a click handler for this button?

Done.

https://codereview.chromium.org/2082363004/diff/1/chromeos/dbus/cryptohome_cl...
File chromeos/dbus/cryptohome_client.cc (right):

https://codereview.chromium.org/2082363004/diff/1/chromeos/dbus/cryptohome_cl...
chromeos/dbus/cryptohome_client.cc:1174:
low_disk_space_handler_.Run(disk_free_bytes);
On 2016/06/23 08:00:45, yoshiki wrote:
> low_disk_space_handler_ may be null so null check is necessary.

Using WeakPtr now, which will result in a no-op if low_disk_space_handler_ is
already destroyed.

[dspaid, 2016-06-27 07:28:31]

dspaid@chromium.org changed reviewers:
+ satorux@chromium.org

[satorux1, 2016-06-27 07:46:40]

The patch is relatively big. How about splitting the d-bus stuff into a separate
patch?

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification.cc (right):

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:27: namespace {
nit: add a blank line?

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification.h (right):

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.h:21: class LowDiskNotification
{
class comment is missing.

https://google.github.io/styleguide/cppguide.html#Class_Comments

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.h:40: void
SetNotificationIntervalForTest(base::TimeDelta interval);
function comments are missing.

https://google.github.io/styleguide/cppguide.html#Function_Comments

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
File chromeos/dbus/cryptohome_client.cc (right):

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
chromeos/dbus/cryptohome_client.cc:1174:
low_disk_space_handler_.Run(disk_free_bytes);
Check if low_disk_space_handler_.is_null() before Running?

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
File chromeos/dbus/cryptohome_client.h (right):

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
chromeos/dbus/cryptohome_client.h:120: virtual void ResetLowDiskSpaceHandler() =
0;
I think resetting is possible by something like
SetLowDiskSpaceHandler(LowDiskSpaceHandler()). Having one less method would be
better?

[satorux1, 2016-06-27 07:46:46]

The patch is relatively big. How about splitting the d-bus stuff into a separate
patch?

[yoshiki, 2016-06-27 08:14:02]

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification.cc (right):

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:47:
~LowDiskNotificationDelegate() override {}
nit: please add DISALLOW_COPY_AND_ASSIGN(LowDiskNotificationDelegate);

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:69: void
LowDiskNotification::OnLowDiskSpace(uint64_t free_disk_bytes) {
I don't know much about DBusThreadManager, but is that this callback method
called on the UI thread? Notification methods must be called on the UI thread.
Please double-check.

Adding the following check might help.
DCHECK_CURRENTLY_ON(BrowserThread::UI);

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:118:
notification->set_pinned(true);
If you use pinned notification, you need to add the code removing the
notification. This is because user can't remove a pinned notification from the
message center manually. (sorry for lack of my explanation)

You need do either of:
- Add the code removing the notification when disk gets enough space
- Make the notification non-pinned

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification_unittest.cc (right):

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification_unittest.cc:39:
low_disk_notification_ = new LowDiskNotification();
nit: using unique_ptr is better for consistency?

[dspaid1, 2016-06-28 00:39:04]

dspaid@google.com changed reviewers:
+ dspaid@google.com

[dspaid1, 2016-06-28 00:39:06]

dbus changes split out into https://codereview.chromium.org/2101853002#ps1

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification.cc (right):

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:27: namespace {
On 2016/06/27 07:46:39, satorux1 wrote:
> nit: add a blank line?

Done.

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:47:
~LowDiskNotificationDelegate() override {}
On 2016/06/27 08:14:02, yoshiki wrote:
> nit: please add DISALLOW_COPY_AND_ASSIGN(LowDiskNotificationDelegate);

Done.

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:69: void
LowDiskNotification::OnLowDiskSpace(uint64_t free_disk_bytes) {
On 2016/06/27 08:14:02, yoshiki wrote:
> I don't know much about DBusThreadManager, but is that this callback method
> called on the UI thread? Notification methods must be called on the UI thread.
> Please double-check.
> 
> Adding the following check might help.
> DCHECK_CURRENTLY_ON(BrowserThread::UI);
> 

As far as I can tell based on comments this should be on the UI thread (and
things seem to be working without warnings/errors, so I assume that is the
case).  I've added the DCHECK to be safe.

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.cc:118:
notification->set_pinned(true);
On 2016/06/27 08:14:02, yoshiki wrote:
> If you use pinned notification, you need to add the code removing the
> notification. This is because user can't remove a pinned notification from the
> message center manually. (sorry for lack of my explanation)
> 
> You need do either of:
> - Add the code removing the notification when disk gets enough space
> - Make the notification non-pinned

Since we don't have an easy way to determine when there is enough space I've
opted for making it non-pinned.

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification.h (right):

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.h:21: class LowDiskNotification
{
On 2016/06/27 07:46:39, satorux1 wrote:
> class comment is missing.
> 
> https://google.github.io/styleguide/cppguide.html#Class_Comments

Done.

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.h:40: void
SetNotificationIntervalForTest(base::TimeDelta interval);
On 2016/06/27 07:46:40, satorux1 wrote:
> function comments are missing.
> 
> https://google.github.io/styleguide/cppguide.html#Function_Comments

Done.

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification_unittest.cc (right):

https://codereview.chromium.org/2082363004/diff/20001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification_unittest.cc:39:
low_disk_notification_ = new LowDiskNotification();
On 2016/06/27 08:14:02, yoshiki wrote:
> nit: using unique_ptr is better for consistency?

Done.

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
File chromeos/dbus/cryptohome_client.cc (right):

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
chromeos/dbus/cryptohome_client.cc:1174:
low_disk_space_handler_.Run(disk_free_bytes);
On 2016/06/27 07:46:40, satorux1 wrote:
> Check if low_disk_space_handler_.is_null() before Running?

Done.

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
File chromeos/dbus/cryptohome_client.h (right):

https://codereview.chromium.org/2082363004/diff/20001/chromeos/dbus/cryptohom...
chromeos/dbus/cryptohome_client.h:120: virtual void ResetLowDiskSpaceHandler() =
0;
On 2016/06/27 07:46:40, satorux1 wrote:
> I think resetting is possible by something like
> SetLowDiskSpaceHandler(LowDiskSpaceHandler()). Having one less method would be
> better?

Done.

[dspaid, 2016-06-28 01:46:43]

Description was changed from

==========
Show notifications on low disk space.

BUG=622590
==========

to

==========
Show notifications on low disk space.

The notifications are shown at most once per hour, or whenever the severity of
the notification changes.

BUG=622590
TEST=./out/Release/unit_tests --gtest_filter=LowDiskNotificationTest.*
==========

[yoshiki, 2016-06-28 02:05:53]

Thanks, lgtm

[dspaid, 2016-06-28 04:16:15]

PTAL

The spec has been updated to only show MEDIUM severity notifications once per
session (unless severity level changes), so that log has been updated.

[satorux1, 2016-06-28 04:16:53]

lgtm

[dspaid, 2016-06-28 04:27:40]

dspaid@chromium.org changed reviewers:
+ oshima@chromium.org

[dspaid, 2016-06-28 04:27:41]

+oshima for ui/chromeos/ changes.

[satorux1, 2016-06-28 04:42:34]

oops. my lgtm was for another patch...

[satorux1, 2016-06-28 05:53:27]

lgtm with a request.

https://codereview.chromium.org/2082363004/diff/60001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification.h (right):

https://codereview.chromium.org/2082363004/diff/60001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.h:46: Severity severity);
nit: adds a blank line between functions?

[satorux1, 2016-06-28 05:56:42]

Maybe src/third_party/cros_system_api  needs to be updated?


../../chromeos/dbus/cryptohome_client.cc:937:41: error: no member named
'kSignalLowDiskSpace' in namespace 'cryptohome'
                            cryptohome::kSignalLowDiskSpace,

[dspaid, 2016-06-28 06:29:53]

system_api changes pending cq at
https://chromium-review.googlesource.com/#/c/356223/1

https://codereview.chromium.org/2082363004/diff/60001/chrome/browser/chromeos...
File chrome/browser/chromeos/ui/low_disk_notification.h (right):

https://codereview.chromium.org/2082363004/diff/60001/chrome/browser/chromeos...
chrome/browser/chromeos/ui/low_disk_notification.h:46: Severity severity);
On 2016/06/28 05:53:27, satorux1 wrote:
> nit: adds a blank line between functions?

Done.

[oshima, 2016-06-29 15:26:26]

can you look into build failures?

[dspaid, 2016-06-29 20:59:16]

On 2016/06/29 15:26:26, oshima wrote:
> can you look into build failures?

They're because the new dbus constant in third_party/cros_system_api isn't being
seen yet (submitted Tuesday).  Is there anything that needs to be done to get
that synced or is it just a matter of time?

[satorux1, 2016-06-29 21:00:58]

Did you update src/DEPS for  src/third_party/cros_system_api?

[dspaid, 2016-06-29 22:44:47]

On 2016/06/29 21:00:58, satorux1 wrote:
> Did you update src/DEPS for  src/third_party/cros_system_api?

Thanks, didn't know about that.  Sent you
https://codereview.chromium.org/2107363002/

[dspaid1, 2016-06-30 04:34:19]

Tests should be passing now (with the exception of a flaky chromevox test that
has been particularly unstable over the past few days).
The red tests at the end of the list are weren't run (not sure why) but pass
manually.

[dspaid, 2016-06-30 05:59:50]

On 2016/06/30 04:34:19, dspaid1 wrote:
> Tests should be passing now (with the exception of a flaky chromevox test that
> has been particularly unstable over the past few days).
> The red tests at the end of the list are weren't run (not sure why) but pass
> manually.

Sorry, spoke too soon.  Some test errors crept in.  Fixing...

[dspaid, 2016-06-30 07:30:01]

On 2016/06/30 05:59:50, dspaid wrote:
> On 2016/06/30 04:34:19, dspaid1 wrote:
> > Tests should be passing now (with the exception of a flaky chromevox test
that
> > has been particularly unstable over the past few days).
> > The red tests at the end of the list are weren't run (not sure why) but pass
> > manually.
> 
> Sorry, spoke too soon.  Some test errors crept in.  Fixing...

Ok, all the tests should be green now.

[oshima, 2016-06-30 12:43:02]

Note that we're migrating vector icons, except for multi colored one.

Looks like they're multi color, so lgtm.

[dspaid, 2016-06-30 23:33:25]

The CQ bit was checked by dspaid@chromium.org

[dspaid, 2016-06-30 23:33:26]

The patchset sent to the CQ was uploaded after l-g-t-m from
yoshiki@chromium.org, satorux@chromium.org
Link to the patchset: https://codereview.chromium.org/2082363004/#ps100001
(title: "Switch to thread checker instead of checking for UI thread explicitly")

[commit-bot: I haz the power, 2016-06-30 23:34:05]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-06-30 23:39:37]

Description was changed from

==========
Show notifications on low disk space.

The notifications are shown at most once per hour, or whenever the severity of
the notification changes.

BUG=622590
TEST=./out/Release/unit_tests --gtest_filter=LowDiskNotificationTest.*
==========

to

==========
Show notifications on low disk space.

The notifications are shown at most once per hour, or whenever the severity of
the notification changes.

BUG=622590
TEST=./out/Release/unit_tests --gtest_filter=LowDiskNotificationTest.*
==========

[commit-bot: I haz the power, 2016-06-30 23:39:39]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2016-06-30 23:40:07]

CQ bit was unchecked.

[commit-bot: I haz the power, 2016-06-30 23:42:23]

Description was changed from

==========
Show notifications on low disk space.

The notifications are shown at most once per hour, or whenever the severity of
the notification changes.

BUG=622590
TEST=./out/Release/unit_tests --gtest_filter=LowDiskNotificationTest.*
==========

to

==========
Show notifications on low disk space.

The notifications are shown at most once per hour, or whenever the severity of
the notification changes.

BUG=622590
TEST=./out/Release/unit_tests --gtest_filter=LowDiskNotificationTest.*

Committed: https://crrev.com/0d43514effc1b793aef2794815bc6f5b5e2811db
Cr-Commit-Position: refs/heads/master@{#403342}
==========

[commit-bot: I haz the power, 2016-06-30 23:42:25]

Patchset 6 (id:??) landed as
https://crrev.com/0d43514effc1b793aef2794815bc6f5b5e2811db
Cr-Commit-Position: refs/heads/master@{#403342}

[James Cook, 2016-07-01 04:29:34]

On 2016/06/30 23:42:25, commit-bot: I haz the power wrote:
> Patchset 6 (id:??) landed as
> https://crrev.com/0d43514effc1b793aef2794815bc6f5b5e2811db
> Cr-Commit-Position: refs/heads/master@{#403342}

Appears to have an uninitialized memory read. See
https://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20ChromeOS%20...

Not reverting yet - can you take a look?

[dspaid, 2016-07-01 06:25:14]

Patchset #7 (id:120001) has been deleted