Logs are present at https://pantheon.corp.google.com/storage/browser/chromiumos-test-logs/bugfiles/cros/901495/?debugUI=CLOUD

I think I saw a similar issue a few days ago on a ToT caroline build.

Is the failure reproducible on multiple sand devices?
Is it reproducible on version before or after .19?
What is the repro rate on the failing sand device?

Logs seem really strange in that the there's no record in eventlog of suspend but instead,

0 | 2018-02-12 17:31:31 | Log area cleared | 4088
1 | 2018-02-12 17:31:31 | Kernel Event | Clean Shutdown
2 | 2018-02-12 17:31:58 | System boot | 6
3 | 2018-02-12 17:31:58 | Last post code in previous boot | 0x34 | Unknown
4 | 2018-02-12 17:31:58 | Chrome OS Developer Mode
5 | 2018-02-12 17:31:58 | Memory Cache Update | Normal | Success
6 | 2018-02-12 17:31:58 | Memory Cache Update | Variable | Success
7 | 2018-02-12 17:31:58 | cr50 Update Reset
8 | 2018-02-12 17:31:58 | ACPI Enter | S5
9 | 2018-02-12 17:32:02 | System boot | 7
10 | 2018-02-12 17:32:02 | Chrome OS Developer Mode
11 | 2018-02-12 17:32:02 | Memory Cache Update | Variable | Success
12 | 2018-11-02 14:34:06 | System boot | 8
13 | 2018-11-02 14:34:06 | Chrome OS Developer Mode
14 | 2018-11-02 14:34:06 | Memory Cache Update | Variable | Success

So I suspect the 'cr50 Update Reset' forced a shutdown therefore power button was the only wake source.  Note also RTC time got lost as well?

On powerd.PREVIOUS,

[1102/213346:INFO:suspend_delay_controller.cc(223)] Notifying observers that suspend is ready
[1102/213346:INFO:suspender.cc(457)] Starting suspend
[1102/213346:INFO:main.cc(266)] Running "/usr/bin/powerd_setuid_helper --action=suspend  --suspend_wakeup_count_valid --suspend_wakeup_count=20 --suspend_to_idle"
<GARBAGE>



ec.log also shows only ~3min of uptime although during that time it does show transitions from S0 to S0ix including,

[173.884394 KB extra IRQ]
[173.965166 KB extra IRQ]

which ec code base says ( /src/platform/ec/common/keyboard_8042.c )

				/*
				 * We keep getting data, but the host keeps
				 * ignoring us.  Fine, we're done waiting.
				 * Hey, host, are you ever gonna get to this
				 * data?  Send it another interrupt in case it
				 * somehow missed the first one.
				 */
				CPRINTS("KB extra IRQ");
				lpc_keyboard_resume_irq();
				retries = 0;
				break;
			}

So it appears EC wants to announce the keyboard input but nobody is listening.  Is it because something horrible or WAI is occurring for cr50 update at the same time.

b/111832956 seems to have similar eventlog signature and repro (suspend led to reboot?)

<Bulk edit> Reminder M71 Stable is approaching. Please review this bug and assess if this is indeed a RBS. If not, please remove the RBS label. If so, please make sure any planned work will be tested in Beta and verified before the Stable date. Thanks

@Sontis, can you answer follow-up questions in #c6

Able to reproduce this issue on M70_11021.56.0 build with multiple(all) sand devices.

This issue is consistent (100%). 

<BULK EDIT> Reminder we nearinf M71 Stable. Please review this issue and update with your plan on next actions. Thanks.

Based on #9, then its not a regression in M-71 but M-70 instead.  I can certainly repro now that I have the device.  I'd suspect something to do w/ EC FW and that does align w/ a release here at least,

https://cros-goldeneye.corp.google.com/chromeos/console/firmwareQualEditDetails?firmwareQualId=542

Doesn't need to block M-71 but should definitely be correct ASAP.

YH, can you have a look?

@Sontis, could you try running on M-69, and be sure FW is before .206?

Able to reproduce this issue with M69_10895.10.0
FW: Google_Sand.9042.178.0

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/autotest/+/9dec757be52d483c082a6977aaddde643fdd6b69

commit 9dec757be52d483c082a6977aaddde643fdd6b69
Author: Todd Broch <tbroch@chromium.org>
Date: Wed Nov 21 02:27:51 2018

power_WakeSources: change _is_valid_wake_source if hammerd missing.

If DUT doesn't contain hammerd then either its image is incorrectly
configured or its not a detachable. For purposes of this test lets
assume the latter and return false.

BUG=chromium:901495
TEST=power_WakeSources on non-detachable no longer tests base wake.

Change-Id: I9e1641ac42f2c70169d263f2f87be34037dbe0fd
Reviewed-on: https://chromium-review.googlesource.com/1344349
Commit-Ready: Ravi Chandra Sadineni <ravisadineni@chromium.org>
Tested-by: Ravi Chandra Sadineni <ravisadineni@chromium.org>
Reviewed-by: Ravi Chandra Sadineni <ravisadineni@chromium.org>

[modify] https://crrev.com/9dec757be52d483c082a6977aaddde643fdd6b69/server/site_tests/power_WakeSources/power_WakeSources.py