Mailing-List: contact reviews-help@impala.incubator.apache.org; run by ezmlm
Precedence: bulk
Message-Id: <201705122056.v4CKuCdo000360@ip-10-146-233-104.ec2.internal>
Date: Fri, 12 May 2017 20:56:12 +0000
From: "Lars Volker (Code Review)" <gerrit@cloudera.org>
To: impala-cr@cloudera.com, reviews@impala.incubator.apache.org
CC: Matthew Jacobs <mj@cloudera.com>
Reply-To: lv@cloudera.com
Subject: =?UTF-8?Q?=5Bnative-toolchain-CR=5D_IMPALA-5208=2C_IMPALA-5187=3A_Fixes_for_Breakpad_=23681=2C_=23728=0A?=
In-Reply-To: <gerrit.1494538118000.If080919f0d241eac90edd8d23bf6007e09157e23@gerrit.cloudera.org>
References: <gerrit.1494538118000.If080919f0d241eac90edd8d23bf6007e09157e23@gerrit.cloudera.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Content-Disposition: inline
User-Agent: Gerrit/2.12.7
archived-at: Fri, 12 May 2017 20:56:18 -0000

Lars Volker has submitted this change and it was merged.

Change subject: IMPALA-5208, IMPALA-5187: Fixes for Breakpad #681, #728
......................................................................


IMPALA-5208, IMPALA-5187: Fixes for Breakpad #681, #728

IMPALA-5208 depends on Breakpad #681:
Breakpad had an issue [1] with random GUID generation on Linux, that
lead to processes started within the same second to write minidumps to
the same filename. Only one would win and overwrite minidumps of the
other processes.

IMPALA-5187 depends on Breakpad #728:
When writing a minidump on Linux, Breakpad called clone() in
linux/handler/exception_handler.cc with the CLONE_FILES flag. If the
parent process died while the child waited for the continuation signal,
the write side of the pipe 'fdes' stayed open in the child. The child
would not receive a SIGPIPE and would wait forever.

This change bumps the Breakpad version in the toolchain to include fixes
for the upstream bugs. It will allow us to remove the workaround we
introduced to address IMPALA-3794. I tested this by reverting the
workaround, removing any startup delay from the cluster startup scripts,
and then verified that the test_breakpad.py passes.

This change also removes the patch to increase the maximum number of
threads for minidump_stackwalk, which has been merged upstream.

[1] https://bugs.chromium.org/p/google-breakpad/issues/detail?id=681

Change-Id: If080919f0d241eac90edd8d23bf6007e09157e23
---
M buildall.sh
A source/breakpad/breakpad-ffe3e478657dc7126fca6329dfcedc49f4c726d9-patches/0001-Add-basic-support-for-dwz-dwarf-extension.patch
2 files changed, 78 insertions(+), 1 deletion(-)

Approvals:
  Lars Volker: Looks good to me, approved; Verified


-- 
To view, visit http://gerrit.cloudera.org:8080/6866
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: merged
Gerrit-Change-Id: If080919f0d241eac90edd8d23bf6007e09157e23
Gerrit-PatchSet: 2
Gerrit-Project: native-toolchain
Gerrit-Branch: master
Gerrit-Owner: Lars Volker <lv@cloudera.com>
Gerrit-Reviewer: Lars Volker <lv@cloudera.com>
Gerrit-Reviewer: Matthew Jacobs <mj@cloudera.com>