New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 594374 link

Starred by 1 user

Issue metadata

Status: Verified
Owner:
Last visit > 30 days ago
Closed: Mar 2016
Cc:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 1
Type: Bug



Sign in to add a comment

servod fails to start for various boards on servo V3

Reported by jrbarnette@chromium.org, Mar 13 2016

Issue description

When the lab updated servo to R50-8041.0.0, servod for
various chromeboxes began failing.

This bug is separate from  bug 594163 , although it occurred
at the same time, and may have been caused by the same CL
As with the other bug, it will block update of servo in the
test lab.

I spot-checked for zako and panther, and saw the same
error.  Boards that may be affected include:
    zako
    panther
    guado
    tricky
    monroe
    buddy
    sumo
    mccloud
    tidus
    stumpy

Here's what I think are the relevant messages from a servo for
a zako board:
2016-03-11 12:44:11,969 - Servod - INFO - Initialized ec_ec3po_interp_connect to on
2016-03-11 12:44:11,971 - root - INFO - No device path specified for usbkm232 handler. Use the servo atmega chip to handle.
2016-03-11 12:44:13,500 - Servod - INFO - USBKM232: /dev/ttyO4
2016-03-11 12:44:14,010 - root - ERROR - Presence check response from atmega KB emu: rsp: 
Traceback (most recent call last):
  File "/usr/lib/python-exec/python2.7/servod", line 9, in <module>
    load_entry_point('servo===0.0.1-4c7d647', 'console_scripts', 'servod')()
  File "/usr/lib/python2.7/site-packages/servo/servod.py", line 512, in main
    main_function()
  File "/usr/lib/python2.7/site-packages/servo/servod.py", line 502, in main_function
    servod.hwinit(verbose=True)
  File "/usr/lib/python2.7/site-packages/servo/servo_server.py", line 772, in hwinit
    self._keyboard = self._init_keyboard_handler(self, self._board)
  File "/usr/lib/python2.7/site-packages/servo/servo_server.py", line 160, in _init_keyboard_handler
    return keyboard_handlers.USBkm232Handler(servo, self._usbkm232)
  File "/usr/lib/python2.7/site-packages/servo/keyboard_handlers.py", line 618, in __init__

 
Summary: servod fails to start for various boards on servo V3 (was: servod fails to start for chromeboxes on servo V3)
I've also found this problem on nyan_kitty.  Go figure.

The problem also occurs on rikku.

Labels: Infra-ChromeOS
Cc: dshi@chromium.org kevcheng@chromium.org
Including next week's deputies:  we can't update servo in the
lab until this is fixed.

I haven't worked on the kb emulator stuff before, but it looks like no response is received from the atmega.

Is there a link to one of the full logs? I just want to make sure that nothing else is using that PTY.
Looks like it was triggered by enabling atmega by default in
https://chromium.googlesource.com/chromiumos/third_party/hdctools/+/4c7d647133bc07eeb9e41d068450444ef9a9e39c
but I'd guess the lab servos aren't flashed yet.

I think waihong's follow-on commit above should have already fixed this, by making an unflashed atmega a warning rather than an exception.

The kb emulator error happened on chromeboxes/chromebases/chromebits when its servo has the old keyboard map. Most of the lab servos still use the old keyboard map.

It should be fixed by:
https://chromium.googlesource.com/chromiumos/third_party/hdctools/+/dc412019769f82c07b272323b9b3c333647eb212

A new servo image push with the above fix will solve this issue.

Comment 9 by autumn@chromium.org, Mar 14 2016

Owner: waihong@chromium.org
Status: Assigned (was: Available)
Labels: -Infra-ChromeOS
Status: Fixed (was: Assigned)
Labels: VerifyIn-51
Status: Verified (was: Fixed)
Bulk verified

Sign in to add a comment