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

Issue 715374 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner: ----
Closed: Jun 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 1
Type: Bug



Sign in to add a comment

chromeos2-row4-rack7-host4 frequently fails autoupdate_EndToEndTest.paygen* with file system errors

Project Member Reported by michae...@chromium.org, Apr 26 2017

Issue description

This host commonly runs autoupdate_EndToEndTest.paygen* tests.

The host corresponds to daisy-release so this has blocked several LKGM updates: https://uberchromegw.corp.google.com/i/chromeos/builders/daisy-release?numbuilds=100

The tests either ERROR or ABORT, with errors such as:

[0424/135414:ERROR:update_engine_client.cc(217)] Update failed, current operation is UPDATE_STATUS_IDLE, last error code is ErrorCode::kFilesystemVerifierError(47)

  or:

[0425/105454:ERROR:update_engine_client.cc(217)] Update failed, current operation is UPDATE_STATUS_IDLE, last error code is ErrorCode::kPostinstallRunnerError(5)

  And in sysinfo/var/log/update_engine/update_engine.*:

Filesystem hash verification failed
Expected 72d017b927b13be28d48a0b2d16447b3fd7d255b != 6aa98b30a233ae71e5c1c2d1160435a0349ffe80
SetImage failed.
PostInstall Failed

 and the messages file shows numerous I/O errors occurring at the same time, e.g.:

2017-04-25T10:50:41.782298-07:00 ERR kernel: [   10.520564] end_request: I/O error, dev mmcblk0, sector 4475961
2017-04-25T10:50:41.787307-07:00 ERR kernel: [   10.523214] Buffer I/O error on device mmcblk0p5, logical block 524167
2017-04-25T10:50:41.967313-07:00 ERR kernel: [   10.705602] end_request: I/O error, dev mmcblk0, sector 4476064
2017-04-25T10:50:42.137313-07:00 ERR kernel: [   10.873494] end_request: I/O error, dev mmcblk0, sector 4476065

  Other hosts don't appear to be affected with the same frequency.
 

Comment 1 by warx@chromium.org, Apr 26 2017

Dup of issue 710716?
https://codereview.chromium.org/2824283003/ may take care of this?
I don't see the similarities with issue 710716:

* This is daisy, not rialto
* These are specifically file system errors, not omaha request failures

Comment 3 by aut...@google.com, Jun 2 2017

Status: WontFix (was: Untriaged)
Sorry for the delay, looks like the host is behaving well now: http://cautotest.corp.google.com/afe/#tab_id=view_host&object_id=654 

Pls re-open if you see more issues. 

Sign in to add a comment