These have the same symptoms, the root cause could be the same:
NoStatePrefetchBrowserTest.Png
NoStatePrefetchBrowserTest.PrefetchSimple
NoStatePrefetchBrowserTest.SSLSubresourceError
NoStatePrefetchBrowserTest.Loop
NoStatePrefetchBrowserTest.MetaTagCSP

Yesterday I wanted to post this handy command for running a test repeatedly until it fails: 

xvfb-run -s "-screen 0 1024x768x24" out/Debug/browser_tests --disable-sandbox --gtest_filter=NoStatePrefetchBrowserTest.Png --gtest_repeat=-1 --gtest_break_on_failure

(did not fail on my machine a single time running overnight)

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/9b665496e6c89667a23ae09138fd982132664b40

commit 9b665496e6c89667a23ae09138fd982132664b40
Author: mattcary <mattcary@chromium.org>
Date: Thu Oct 20 10:57:20 2016

Prerender: remove possible race in browser tests to improve flakiness.

There may be a race condition that causes flakiness in the form of timeouts in
several prerender browser tests involving waiting for prerender contents to be
created. As I have not been able to reproduce the flakiness locally, I'm not
sure this is actually a fix for the flakes. Fingers are crossed and wood is
being knocked.

BUG= 657025 

Review-Url: https://chromiumcodereview.appspot.com/2434813004
Cr-Commit-Position: refs/heads/master@{#426446}

[modify] https://crrev.com/9b665496e6c89667a23ae09138fd982132664b40/chrome/browser/prerender/prerender_test_utils.cc
[modify] https://crrev.com/9b665496e6c89667a23ae09138fd982132664b40/chrome/browser/prerender/prerender_test_utils.h

I don't see any more flakes on the dashboard, although it is hard to interpret. pasko@, can you confirm?

fakiness dashboard has recent results on the left. Link that loads faster for me:
http://test-results.appspot.com/dashboards/flakiness_dashboard.html#testType=browser_tests&tests=NoStatePrefetchBrowserTest.SSLError

timeout yesterday:
https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.linux%2FLinux_Tests%2F48989%2F%2B%2Frecipes%2Fsteps%2Fbrowser_tests_on_Ubuntu-12.04%2F0%2Fstdout

timeout yesterday on plznavigate bot:
https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.fyi%2FBrowser_Side_Navigation_Linux%2F23759%2F%2B%2Frecipes%2Fsteps%2Fbrowser_tests_on_Ubuntu-12.04%2F0%2Fstdout

It is not super common, so it it gets too hard to analyse, let's wontfix it.

What is special about these two timeout runs is that it spits this line:
[...] password_store_factory.cc(248)] Using basic (unencrypted) store for password storage. See https://chromium.googlesource.com/chromium/src/+/master/docs/linux_password_storage.md for more information about password storage options.

...which seems to correlate to browser tests timing out, not just this one. I don't see such a bug created yet, will need to look through more logs for it to be more convincing.

That password_store_factory message appears for me every time I run it locally (and it's never timed out locally). So that's probably a red herring.

There is also a low level of flakes for other SSL uses of the embedded test server when the cert is invalid. Therefore I suspect that's the root cause of the flakes (timeouts in particular). Given the very low level of flakes, I think wontfix is reasonable.

Tests looked at:
CaptivePortalBrowserTest.RedirectSSLCertError (CERT_MISMATCHED_NAME, has flakes)
ChromeNavigationBrowserTest.TransientEntryPreservedOnMultipleNavigationsDuringInterstitial (CERT_EXPIRED, has flakes)
PolicyTest.CertificateTransparencyEnforcementDisabledForUrls (CERT_OK, saw no flakes)
PolicyTest.SafeBrowsingExtendedReportingOptInAllowed (CERT_EXPIRED, has flakes)