I believe browser_tests use 30 shards on the memory bots. Ryan, is single-process-mash browser_tests using 30 as well?

(pardon the earlier comment from my personal account)

It looks like both browser_tests and single_process_mash_browser_tests are running with 30 shards on the ASAN bot, but only running with 20 shards on the MSAN bot[1].  The reason the single_process_mash_browser_tests fail is because of the 1800s hard timeout we specify, which the browser_tests don't have.  I'm wondering if the right thing to do here is to bump up the timeout (and possibly have to deal with this again in the future), just remove it, or increase the number of shards.  I'm inclined to remove the timeout because it's what we're doing for browser_tests, but I can also see why we might want to have an explicit timeout.  +jamescook as well for thoughts.

[1] https://cs.chromium.org/chromium/src/testing/buildbot/chromium.memory.json?rcl=c015c9c2e2626a8b1533a790c826b9b2cdb568e9&l=4957

I'm OK with doing what browser_tests is doing.

SingleProcessMash browser_tests shouldn't take any longer than regular browser_tests. It seems like either both should have timeouts or neither should have timeouts.

I believe the reason we put a hard limit is we had some bugs that made the tests stall our for a really long time. As James says, ideally we should be doing what browser_tests do.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/cbdcf0bd4359842b286cabd40166023d3e769399

commit cbdcf0bd4359842b286cabd40166023d3e769399
Author: Ryan Cui <rcui@chromium.org>
Date: Thu Sep 13 00:08:29 2018

Remove timeouts for single_process_mash_browser_tests

To match the configuration of the regular browser_tests, and to address
timeouts seen on the MSAN builder.

Bug:  883296 
Change-Id: I6dadd77a19fb0339dd590bb101f4bcd3c4c89fde
Reviewed-on: https://chromium-review.googlesource.com/1222672
Reviewed-by: Ben Pastene <bpastene@chromium.org>
Commit-Queue: Ryan Cui <rcui@chromium.org>
Cr-Commit-Position: refs/heads/master@{#590869}
[modify] https://crrev.com/cbdcf0bd4359842b286cabd40166023d3e769399/testing/buildbot/chromium.chromiumos.json
[modify] https://crrev.com/cbdcf0bd4359842b286cabd40166023d3e769399/testing/buildbot/chromium.memory.json
[modify] https://crrev.com/cbdcf0bd4359842b286cabd40166023d3e769399/testing/buildbot/test_suites.pyl

Note there are two other related bugs,  issue 884189  and  issue 884204 .
After commit cbdcf0bd4359842b286cabd40166023d3e769399 there still were failures, e.g. https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/linux-chromeos-dbg/7829 and https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Linux%20Chromium%20OS%20ASan%20LSan%20Tests%20%281%29/29007

The first failure looks related to https://bugs.chromium.org/p/chromium/issues/detail?id=884589.  Looking into the second one.  Both are failures on the ASan LSan bot, not the MSan bot (which is showing lots of green), so marking this bug as fixed.