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

Issue 850731 link

Starred by 3 users

Issue metadata

Status: Fixed
Owner:
Closed: Jun 2018
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 3
Type: Bug

Blocking:
issue 844865



Sign in to add a comment

install-build-deps-android wasn't run on (at least) the main waterfall linux msan bot

Project Member Reported by thakis@chromium.org, Jun 7 2018

Issue description

In https://chromium-review.googlesource.com/c/chromium/src/+/1089430 dpranke says: "It was my understanding that we ran both install-build-deps and install-build-deps-android as part of the image that we use [...]. Did you happen to file a bug I can look into?"

On (at least) the MSan build this didn't happen, see https://chromium-review.googlesource.com/c/chromium/src/+/1089211 / https://ci.chromium.org/buildbot/chromium.memory/Linux%20MSan%20Builder/12405 / https://logs.chromium.org/v/?s=chromium%2Fbb%2Fchromium.memory%2FLinux_MSan_Builder%2F12405%2F%2B%2Frecipes%2Fsteps%2Fcompile%2F0%2Fstdout :

FAILED: gen/headless/devtools_bindings_test.js 
python ../../third_party/closure_compiler/js_binary.py --compiler ../../third_party/closure_compiler/compiler/compiler.jar --output gen/headless/devtools_bindings_test.js --deps gen/headless/js_devtools_bindings_lib.js_library --sources ../../headless/test/bindings_test.js --config --flags jscomp_error=checkTypes dependency_mode=STRICT compilation_level=SIMPLE language_out=ES5_STRICT entry_point=chromium.BindingsTest --externs ../../third_party/closure_compiler/externs/chrome.js ../../third_party/closure_compiler/externs/polymer-1.0.js ../../headless/lib/tab_socket_externs.js
Exception in thread "main" java.lang.UnsupportedClassVersionError: com/google/javascript/jscomp/CommandLineRunner : Unsupported major.minor version 52.0
	at java.lang.ClassLoader.defineClass1(Native Method)
	at java.lang.ClassLoader.defineClass(ClassLoader.java:803)
	at java.security.SecureClassLoader.defineClass(SecureClassLoader.java:142)
	at java.net.URLClassLoader.defineClass(URLClassLoader.java:442)
	at java.net.URLClassLoader.access$100(URLClassLoader.java:64)
	at java.net.URLClassLoader$1.run(URLClassLoader.java:354)
	at java.net.URLClassLoader$1.run(URLClassLoader.java:348)
	at java.security.AccessController.doPrivileged(Native Method)
	at java.net.URLClassLoader.findClass(URLClassLoader.java:347)
	at java.lang.ClassLoader.loadClass(ClassLoader.java:425)
	at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:308)
	at java.lang.ClassLoader.loadClass(ClassLoader.java:358)
	at sun.launcher.LauncherHelper.checkAndLoadMain(LauncherHelper.java:482)

(or, it happened long ago; in any case the java binary there is out of date)
 
Blocking: 844865
Components: Infra
Labels: Infra-Troopers
Since this is a bot config thing (probably need to rerun install-build-deps-android.sh on that bot?), can troopers help with this?
Cc: friedman@chromium.org smut@chromium.org
Components: -Infra Infra>Labs
I think this is probably best handled by labs, actually.

Can we (re-)run install-build-deps-android.sh on all of our current linux bots?

Also, smut/friedman - can you confirm that we run this when we build the trusty and xenial images?
Owner: friedman@chromium.org
Status: Assigned (was: Untriaged)
install-build-deps{,-android}.sh are run on bot deployment, but not after that.

We have avoided running install-build-deps{,-android}.sh on bots after deployment for fear that doing so might break them.  We've debated it numerous times but never decided to actually re-run them.  Instead, changes to those scripts are mirrored in puppet and allowed to run via puppet for ~24 hours and then removed.

Example would be adding a new required package, you'd add it to puppet, puppet would ensure it's installed, then you'd revert the change after it's been allowed to bake in for 24 hours.

Which package is needed?
Nevermind, it looks like you need java installed... probably java 8?  Which master/builder does this need to be installed on?
yeah, looks like it's missing java8.

Let's put it on every bot on master.chromium.memory.
Status: Fixed (was: Assigned)
Oh! We have a master that tracks chromium memory usage?! NICE!

Fix is rolling out now and will be complete within 1 hr.
We should make sure java is installed on *all* the linux machines, not just machines on chromium.memory.
java is installed, just the java that that version of ubuntu specifies (7).  Java 8 is a special package in our local repo that we grabbed from Xenial and are forcing upon Trusty.

Do you want java8 on all linux machines?
Yes, java8 please. Sorry for not being clearer.
Project Member

Comment 11 by bugdroid1@chromium.org, Jun 9 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/1c149502277c1441eec693c3ec160462e150000b

commit 1c149502277c1441eec693c3ec160462e150000b
Author: Nico Weber <thakis@chromium.org>
Date: Sat Jun 09 04:10:45 2018

Run headless tests on memory waterfall.

Relands the last missing bit of https://chromium-review.googlesource.com/1088754
This got reverted the last time because the msan bot had a too-old version of
Java.  This should now be fixed.

TBR=kbr

Bug: 843511, 850731 , 844865 , 843732 , 843734 
Change-Id: I2359e6e721fa603117dd2a0a2f11f459ee2a2ddf
Reviewed-on: https://chromium-review.googlesource.com/1094036
Reviewed-by: Nico Weber <thakis@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#565841}
[modify] https://crrev.com/1c149502277c1441eec693c3ec160462e150000b/testing/buildbot/chromium.fyi.json
[modify] https://crrev.com/1c149502277c1441eec693c3ec160462e150000b/testing/buildbot/chromium.memory.json
[modify] https://crrev.com/1c149502277c1441eec693c3ec160462e150000b/testing/buildbot/test_suite_exceptions.pyl
[modify] https://crrev.com/1c149502277c1441eec693c3ec160462e150000b/testing/buildbot/test_suites.pyl

https://ci.chromium.org/buildbot/chromium.memory/Linux%20MSan%20Builder/12479 was happy now, so the msan bot is good. Don't know if you want to reopen this for comment 10, but I'm all set. Thanks!
Cc: kvitekp@chromium.org
 Issue 844865  has been merged into this issue.

Sign in to add a comment