This test is trying to allocate 1,000,000,000,000 bytes to verify that it fails predictably: https://cs.chromium.org/chromium/src/components/cronet/native/buffer_test.cc?l=72

I guess ASAN changes malloc to check on outrageously large allocations. 
I'll disable this test on ASAN configurations unless somebody has a better idea.
 

I   64.953s Main  FINISHED TRY #3/3
I   64.953s Main  1 failed tests remain.
C   64.954s Main  ********************************************************************************
C   64.954s Main  Detailed Logs
C   64.954s Main  ********************************************************************************
C   64.954s Main  [CRASH] BufferTest.TestInitWithHugeAllocFails:
C   64.954s Main  [ RUN      ] BufferTest.TestInitWithHugeAllocFails
C   64.954s Main  ==7102==WARNING: AddressSanitizer failed to allocate 0xd4a51000 bytes
C   64.954s Main  ==7102==AddressSanitizer's allocator is terminating the process instead of returning 0
C   64.954s Main  ==7102==If you don't like this behavior set allocator_may_return_null=1
C   64.954s Main  ==7102==AddressSanitizer CHECK failed: /b/build/slave/linux_upload_clang/build/src/third_party/llvm/compiler-rt/lib/sanitizer_common/sanitizer_allocator.cc:225 "((0)) != (0)" (0x0, 0x0)
C   64.954s Main  
C   64.954s Main  [ CRASHED      ]
C   64.954s Main  Searching for native crashes in: /tmp/tmpYFnKtW
C   64.954s Main  Unknown Android release, consider passing --packed-lib.
C   64.954s Main  Reading Android symbols from: /b/build/slave/Android_Cronet_Builder_Asan/build/src
C   64.954s Main  Searching for Chrome symbols from within: /b/build/slave/Android_Cronet_Builder_Asan/build/src/out/Release/lib.unstripped:/b/build/slave/Android_Cronet_Builder_Asan/build/src/out/Release/lib:/b/build/slave/Android_Cronet_Builder_Asan/build/src/out/Release
C   64.954s Main  
C   64.954s Main  Stack Trace:
C   64.954s Main    RELADDR   FUNCTION   FILE:LINE
C   64.954s Main    4005b05d  <UNKNOWN>  /b/build/slave/Android_Cronet_Builder_Asan/build/src/ (/system/bin/app_process32
C   64.954s Main  ********************************************************************************
C   64.954s Main  Summary
C   64.954s Main  ********************************************************************************
C   64.955s Main  [==========] 79 tests ran.
C   64.955s Main  [  PASSED  ] 78 tests.
C   64.955s Main  [  FAILED  ] 1 test, listed below:
C   64.955s Main  [  FAILED  ] BufferTest.TestInitWithHugeAllocFails (CRASHED)
C   64.955s Main  
C   64.955s Main  1 FAILED TEST

I just realized that I took the wrong bug. 

I don't know what asan_symbolize.py failure is.

SymbolInformationForSet is removed in r526304 but src/build/android/asan_symbolize.py still references to that function.

+hzl@: could you take a look?

Please take a look at this ASAP, this is an ongoing waterfall failure.

It looks like https://uberchromegw.corp.google.com/i/chromium.android/builders/Android%20Cronet%20Builder%20Asan is stuck (there are no builds since Jan 31 11:08) and one build is pending for about 1.5 hours.

The fix for crash  issue 807590  has landed, so the bot should be fine after it is picked up.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f2408e016d6b5a1f95f1e8f9add2a15653fc8421

commit f2408e016d6b5a1f95f1e8f9add2a15653fc8421
Author: Zhiling Huang <hzl@chromium.org>
Date: Thu Feb 01 04:15:27 2018

Include back SymbolInformationForSet into symbol.py.

Bug:  807591 
Change-Id: Id3764225bcf94ad2f87b32df9ea874272dd97c5d
Reviewed-on: https://chromium-review.googlesource.com/896103
Reviewed-by: Maria Khomenko <mariakhomenko@chromium.org>
Reviewed-by: agrieve <agrieve@chromium.org>
Commit-Queue: Zhiling Huang <hzl@chromium.org>
Cr-Commit-Position: refs/heads/master@{#533551}
[modify] https://crrev.com/f2408e016d6b5a1f95f1e8f9add2a15653fc8421/third_party/android_platform/development/scripts/symbol.py

The latest builds are green. 
https://ci.chromium.org/buildbot/chromium.android/Android%20Cronet%20Builder%20Asan/