Filed feedback immediately after I recovered and signed in.

https://feedback.corp.google.com/product/208/neutron?lView=rd&lRSort=1&lROrder=2&lRFilter=1&lReportSearch=user:abodenha@google.com&lReport=18068970329

Here's eventlog:

238 | 2016-11-29 17:45:20 | ACPI Enter | S3
239 | 2016-11-30 08:58:41 | ACPI Wake | S3
240 | 2016-11-30 08:58:41 | Wake Source | GPIO | 27
241 | 2016-11-30 08:58:41 | EC Event | Lid Open
242 | 2016-11-30 08:58:41 | EC Event | Battery Status Request

And here are the relevant bits from powerd. I don't see anything interesting there:

[1129/140253:INFO:daemon.cc(484)] Lid closed
...
[1129/140253:INFO:state_controller.cc(846)] Turning panel off after entering docked mode
...
[1129/171321:INFO:state_controller.cc(89)] Dimming screen after 1m
[1129/171331:INFO:state_controller.cc(89)] Turning screen off after 1m10s
[1129/171331:INFO:display_power_setter.cc(80)] Asking Chrome to turn all displays off
[1129/171332:INFO:daemon.cc(1337)] Chrome is using presentation display mode
...
[1129/171437:INFO:daemon.cc(760)] Audio is active
...
[1129/171448:INFO:daemon.cc(760)] Audio is inactive
...
[1129/174518:INFO:state_controller.cc(872)] Ready to perform idle action (suspend) after 30m30s
...
[1129/174519:INFO:suspender.cc(466)] Starting suspend
[1129/174519:INFO:main.cc(259)] Running "/usr/bin/powerd_setuid_helper --action=suspend  --suspend_wakeup_count_valid --suspend_wakeup_count=1055"
[1130/085909:INFO:daemon.cc(687)] powerd_suspend returned 0
[1130/085909:INFO:suspender.cc(424)] Finishing request 234094598 successfully
[1130/085909:INFO:state_controller.cc(96)] Undimming screen
[1130/085909:INFO:internal_backlight_controller.cc(686)] Setting resume brightness to 636 (80%)
[1130/085909:INFO:state_controller.cc(96)] Turning screen on
[1130/085909:INFO:internal_backlight_controller.cc(686)] Setting resume brightness to 636 (80%)
[1130/085909:INFO:display_power_setter.cc(80)] Asking Chrome to turn internal display off and external displays on
[1130/085911:INFO:daemon.cc(493)] Lid opened
[1130/085911:INFO:state_controller.cc(846)] Turning panel on after leaving docked mode
[1130/085911:INFO:display_power_setter.cc(80)] Asking Chrome to turn all displays on
[1130/085913:INFO:internal_backlight_controller.cc(670)] Setting brightness to 636 (80%) over 0 ms

Albert, can you attach the relevant Chrome user log from file:///home/chronos/user/log?

Are you using a dock or do you have keyboard/mouse/display connected directly to the device?

Using a dock. One of the plugable USB-C ones.
Attaching the logs from today. 090950 is probably the one you want.

Deleted: chrome_20161129-090950 295 KB

chrome_20161129-090950 295 KB View Download

Deleted: chrome_20161130-090236 66.5 KB

chrome_20161130-090236 66.5 KB View Download

There's a lot of unhappiness in the first log:

[18904:18904:1129/143630:VERBOSE1:screen_locker.cc(213)] Authentication success: 0.005533 second(s)
[18904:18904:1129/143630:VERBOSE1:screen_locker.cc(281)] Hiding the lock screen.
[18904:18904:1129/143630:VERBOSE1:lock_state_controller.cc(445)] StartUnlockAnimationBeforeUIDestroyed
[18904:18904:1129/143630:VERBOSE1:screen_locker.cc(472)] Deleting ScreenLocker 0x1877b55375a0
[18904:18904:1129/143630:VERBOSE1:screen_locker.cc(485)] Destroying ScreenLocker 0x1877b55375a0
[18904:18904:1129/143630:VERBOSE1:screen_locker.cc(492)] Moving wallpaper to unlocked container
[18904:18904:1129/143630:VERBOSE1:screen_locker.cc(497)] Emitting SCREEN_LOCK_STATE_CHANGED with state=0
[18904:18904:1129/143630:VERBOSE1:lock_state_controller.cc(228)] OnLockStateChanged called with locked: 0, shutting_down_: 0, system_is_locked_: 1, lock_fail_timer_.IsRunning(): 0
[18904:18904:1129/143630:VERBOSE1:lock_state_controller.cc(453)] StartUnlockAnimationAfterUIDestroyed
[18904:18904:1129/143630:VERBOSE1:screen_locker.cc(503)] Calling session manager's HandleLockScreenDismissed D-Bus method
[18904:18926:1129/145733:WARNING:shared_memory_posix.cc(430)] Shared memory creation failed; out of file descriptors
[18904:18926:1129/145733:WARNING:shared_memory_posix.cc(430)] Shared memory creation failed; out of file descriptors
[18904:18926:1129/145733:WARNING:shared_memory_posix.cc(430)] Shared memory creation failed; out of file descriptors
...
[18904:18926:1129/145733:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.xLaVvp failed: Too many open files
[18904:18926:1129/145733:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.X5UBV9 failed: Too many open files
[18904:18926:1129/145733:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.eG3mlU failed: Too many open files
[18904:18926:1129/145733:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.2evcLE failed: Too many open files
[18904:18926:1129/145733:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.AAT5ap failed: Too many open files
[18904:18926:1129/145733:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.34W2A9 failed: Too many open files
[18904:18926:1129/145733:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.sIf40T failed: Too many open files
[18904:18920:1129/145733:ERROR:accelerometer_reader.cc(466)] Accelerometer trigger failure: -1: Too many open files
...

[18904:18926:1130/085914:ERROR:socket_posix.cc(82)] CreatePlatformSocket() returned an error, errno=24: Too many open files
[18904:18926:1130/085914:ERROR:socket_posix.cc(82)] CreatePlatformSocket() returned an error, errno=24: Too many open files
[18904:18926:1130/085914:ERROR:socket_posix.cc(82)] CreatePlatformSocket() returned an error, errno=24: Too many open files
[18904:18923:1130/085914:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.PuZEip failed: Too many open files
[18904:18923:1130/085914:ERROR:child_process_host_impl.cc(228)] NOTREACHED() hit.
[18904:18926:1130/085915:ERROR:socket_posix.cc(82)] CreatePlatformSocket() returned an error, errno=24: Too many open files
[18904:18926:1130/085915:ERROR:socket_posix.cc(82)] CreatePlatformSocket() returned an error, errno=24: Too many open files
[18904:18926:1130/085915:ERROR:socket_posix.cc(82)] CreatePlatformSocket() returned an error, errno=24: Too many open files
[18904:18904:1130/085915:WARNING:shared_memory_posix.cc(430)] Shared memory creation failed; out of file descriptors
[18904:18926:1130/085915:ERROR:gpu_memory_buffer_impl_ozone_native_pixmap.cc(59)] dup: Too many open files
[18904:18926:1130/085915:ERROR:socket_posix.cc(82)] CreatePlatformSocket() returned an error, errno=24: Too many open files
[18904:18926:1130/085915:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.0UtHGi failed: Too many open files
[18904:18926:1130/085915:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.os7zqK failed: Too many open files
[18904:18926:1130/085915:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.s2Rzac failed: Too many open files
[18904:18926:1130/085915:WARNING:crash_handler_host_linux.cc(324)] Could not translate tid, attempt = 1 retry ...
[18904:18926:1130/085915:WARNING:shared_memory_posix.cc(430)] Shared memory creation failed; out of file descriptors
[18904:18923:1130/085915:ERROR:shared_memory_posix.cc(290)] Creating shared memory in /dev/shm/.com.google.Chrome.jIXeqn failed: Too many open files
[18904:18923:1130/085915:ERROR:child_process_host_impl.cc(228)] NOTREACHED() hit.
[18904:18904:1130/085915:ERROR:buffer_queue.cc(234)] Failed to allocate backing image surface
...

[18904:18925:1130/085957:ERROR:simple_backend_impl.cc(584)] Simple Cache Backend: wrong file structure on disk: /home/chronos/u-584468ac1ee19cfc950ccb50344f9b2a0fcc0684/Service Worker/CacheStor
age/4b3c19c3f2722f26e3232756cd5969a07cd830d1/cc4c6d92-2adb-44d3-abc9-57196cf26ab5
[18904:18926:1130/085957:ERROR:cache_creator.cc(134)] Unable to create cache
[18904:18925:1130/085957:ERROR:simple_backend_impl.cc(584)] Simple Cache Backend: wrong file structure on disk: /home/chronos/u-584468ac1ee19cfc950ccb50344f9b2a0fcc0684/Service Worker/CacheStor
age/4b3c19c3f2722f26e3232756cd5969a07cd830d1/eb922d0a-0bfc-4a0a-8eb0-dd33d029716f
[18904:18926:1130/085957:ERROR:cache_creator.cc(134)] Unable to create cache
[18904:18925:1130/085957:ERROR:simple_backend_impl.cc(584)] Simple Cache Backend: wrong file structure on disk: /home/chronos/u-584468ac1ee19cfc950ccb50344f9b2a0fcc0684/Service Worker/CacheStor
age/4b3c19c3f2722f26e3232756cd5969a07cd830d1/cc4c6d92-2adb-44d3-abc9-57196cf26ab5
[18904:18926:1130/085957:ERROR:cache_creator.cc(134)] Unable to create cache
...

etc. etc.

I could believe that not being able to wake via mouse/keyboard is due to a bad interaction with the dock that you're using (or does it usually work with that dock?), but the brokenness that you saw after opening the lid is probably due to running out of FDs... which looks similar to what's going on in issue 661306 (is this the same device?).

Yeah, same device. That's at least 4 different issues that appear to stem from something screwing up FD handling. I suspect a single root cause but no one has found a smoking gun yet. Thanks for digging into it.

Mouse and keyboard fail could very easily be the dock. I'm going to take a closer look next time it happens. I suspect the dock is powering down when we suspend which would leave mouse and keyboard unable to trigger a wake.

Benson and/or Sameer probably know whether it's expected for USB input devices connected to a Type C dock to be able to wake a samus from S3 or not. Cc-ing Alec in the off chance that he remembers, too. :-)

USB HID input devices anywhere in the USB topology should always be a wakeup source if they advertise as wakeup capable. The fact that it's a Type-C hub should not make a difference.

In practice, I had good luck with waking up my Cave in docked mode when used with the LG 27UD88-W USB-C docking station monitor.

It's possible the hub has a bug that breaks remote wakeup. Which hub was in in particular albert?

Yeah. Mouse is powered off. Keyboard has power but isn't in a state where it can process keys*. No way mouse and keyboard will trigger a wake if the dock is putting them into that state.

The dock is http://plugable.com/products/ud-ultcdl/

*It's a Kinesis keyboard. Their handling of USB is an epic fail and always has been. I can tell it's powered because it'll beep when tapping caps lock, but takes no other action. It's a state I'm pretty familiar with.

Albert : Have you tried this keyboard directly attached to your laptop while in Docked mode? Does it wake then?

I want to separate to see if the issue is with the keyboard or with the dock.

In terms of the dock, I had one of these docks, but it literally caught on fire and let out acrid smoke so i sent it back to Plugable, so I won't be able to test this.

Also, Plugable has discontinued the original version of the dock due to firmware issues and has been sending replacements to users (except for CHromebook users) : http://plugable.com/2016/10/19/hardware-update-for-plugable-ud-ultcdl-triple-display-docks-purchased-before-july-18-2016/

If we determine it's a hub problem, I'll pile it onto the list of other issues with Plugable products.

Holy crap! Other than the suspend issues, I've found it a lot more reliable than the other docks I've played with, but it sounds like I may have just gotten lucky.  You're welcome to borrow this one to poke at if you'd like. Let me know and I'll drop it by your desk.

I got this one well after July 18 so it should be the updated version.

I haven't tried direct connect on the keyboard or mouse. Annoyance level of multiple cables exceeds the annoyance level of having to open the lid :-), but I'll direct connect today and let you know how it goes.

Sure. I can take a quick look at your dock. any time that is convenient for you. 

I've also hit "resume from sleep" problems with this exact same hub.

# Sequence of events
- plugged into usb-c hub, with 3 external monitors
- using internal keyboard, with external mouse
- put machine to sleep with [search][shift][l]
- disconnect dock, close laptop's lid

- reconnect dock
- open lid to wake up the machine

## Results
- all displays (external & internal) remain off
- no reaction to keyboard input
- the dock already had a usb stick, and judging by the lights on the usb stick, it does look as if chromeos is remounting the stick

I'm forced to do a hard shutdown and reboot. Not good!

BTW, I'm hitting this behaviour on the dev channel (switched to this for android apps), but did NOT get this behaviour on the stable channel prior to switching.

# Version Info
## Machine
Lenovo Thinkpad 13 (SENTRY)

## ChromeOS
Version 56.0.2924.12 dev (64-bit)
Platform 9000.15.0 (Official Build) dev-channel sentry
ARC Version 3535779
Firmware Google_Sentry.7820.156.0

is.rafique@ that sounds different than what I ran into.

Can you file a new bug?  Also file a feedback report (alt+shift+i) right after the problem happens and include the contents of chrome://crashes in the bug.

Hi Albert,

I was not able to repro this with your plugable hub. Any input device i have plugged in consistently wakes my Chromebook Pixel 2015. Maybe I should try docked mode...

Not able to reproduce it while in docked mode either. Maybe this is some interesting combination of exactly your monitor and your keyboard and whatever peripherals you have hanging off of your dock, Albert?

Is this easily reproducible on your Samus system? Could I bring this back to you and see?

I just discovered I can repro by manually causing suspend with shift+search+l, so yes it's easy to repro.

Glad to demo if you'd like though at this point it seems pretty clearly a buggy peripheral issue rather than a Chrome OS issue.

@abodenha (comment 15)

How is it not the same issue?

The exact same dock. And in both cases, it gets stuck part way when resuming from sleep mode.

In your case, you started in docked mode, and ended up with partially restored displays. In my case, I wasn't in docked mode, and it hangs before any screen is activated.

Seems to fit the title exactly: "Failed to wake from sleep in docked mode"

> it seems pretty clearly a buggy peripheral issue rather than a Chrome OS issue.

You didn't read my original post fully. As I said, with the same dock, on the Chrome OS stable channel, this problem did **NOT** occur.

So how do you conclude that this isn't a chrome os issue?

Anyway, you have enough information from my previous post to reproduce this on your end, right?

is.rafique@ The end state is different.  You're also dealing with 3 displays vs my one and you unplugged after suspend while I did not.  The cases might sound similar, but under the hood those are some pretty big differences.

In my case there were 2 issues:
1: My keyboard and mouse will not wake the device from suspend. This appears to be a peripheral issue (which is what I meant in comment 18)
2: My displays activated but the device didn't fully exit docked mode. This happened exactly once and appeared to be caused by bug 660960 (restricted for security reasons, but something to do with file descriptors is misbehaving)

I'm about to close this one out because all of the issues in the initial report are addressed, but I DON'T want to lose track of what you're seeing. THAT needs additional investigation which is why I asked you to file a new bug.

> bug 660960 (restricted for security reasons, but something to do with file descriptors is misbehaving)

Ah, so thats why I couldn't view it. I saw a reference to it above, but couldn't follow it up.

> THAT needs additional investigation which is why I asked you to file a new bug.

Ok, I'll file a new bug (here it is: 673040). If there are any additional forms I need to sign in triplicate to satisfy bureaucracy, do let me know ...

Closing based on reasons in comment #20 and remaining issues now tracked in bug 673040