[
https://issues.apache.org/jira/browse/LUCENE-3349?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Uwe Schindler updated LUCENE-3349:
----------------------------------
Component/s: (was: general/website)
general/test
Description:
Today, JDK/JRE 1.7.0 GA was released by Oracle. Unfortunately they didn't fix
the Hotspot problems affecting loops to be miscompiled (LUCENE-3335,
LUCENE-3346). This can lead to Solr crashing with default configuration on
startup or sudden index corrumption depending on configuration.
We should send an email to the java-user and solr-user list describing the
problem. Also place a note in the news section of Solr, Lucene Core and
top-level website.
I propose the following text:
{quote}
*Jul 28th, 2011: WARNING: Index corruption and crashes in Apache Lucene Core /
Apache Solr with Java 7*
Oracle released *Java 7* today. Unfortunately it contains hotspot compiler
optimizations, which miscompile some loops. This can affect code of several
Apache projects. Sometimes JVMs only crash, but in several cases, results
calculated can be incorrect, leading to bugs in applications (see Hotspot bug
[http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7070134]).
*Apache Lucene Core* and *Apache Solr* are two Apache projects, which are
affected by these bugs, namely all versions released until today. Solr users
with the default configuration will have Java crashing with SIGSEGV as soon as
they start to index documents, as one affected part is the well-known Porter
stemmer (see LUCENE-3335). Other loops in Lucene may be miscompiled, too,
leading to index corruption (especially on Lucene trunk with pulsing codec;
other loops may be affected, too - LUCENE-3346).
These problems were detected only 5 days before the official Java 7 release, so
Oracle had no time to fix those bugs, affecting also many more applications. In
response to our questions, they proposed to include the fixes into service
release u2 (eventually into service release u1, see
[http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/2011-July/005971.html]).
This means you cannot use Apache Lucene/Solr with Java 7 releases before
Update 2! If you do, please don't open bug reports, it is not the committers'
fault! At least disable loop optimizations using the -XX:-UseLoopPredicate JVM
options to not risk index corruptions.
Please note: Also Java 6 users are affected, if they use one of those JVM
options, which are *not* enabled by default: -XX:+OptimizeStringConcat or
-XX:+AggressiveOpts
It is strongly recommended not to use any hotspot optimization switches in any
Java version without extensive testing!
In case you upgrade to Java 7, remember that you may have to reindex, as the
unicode version shipped with Java 7 changed and tokenization behaves
differently (e.g. lowercasing)!
{quote}
was:
Today, JDK/JRE 1.7.0 GA was released by Oracle. Unfortunately they didn't fix
the Hotspot problems affecting loops to be miscompiled (LUCENE-3335,
LUCENE-3346). This can lead to Solr crashing with default configuration on
startup or sudden index corrumption depending on configuration.
We should send an email to the java-user and solr-user list describing the
problem. Also place a note in the news section of Solr, Lucene Core and
top-level website.
I propose the following text:
{quote}
*Jul 28th, 2011: WARNING*
Oracle released *Java 7* today. Unfortunately it contains some hotspot compiler
optimizations, which miscompile some loops in Lucene's code (this affects all
versions released until today, see Hotspot bug
[http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7070134]). Solr users with
the default configuration will have Java crashing with SIGSEGV as soon as they
start to index documents, as one affected part is the well-known Porter stemmer
(see LUCENE-3335). Other loops in Lucene may be miscompiled, too, leading to
index corruption (especially on Lucene trunk with pulsing codec; other loops
may be affected, too).
These problems were detected only 5 days before the official Java 7 release, so
Oracle had no time to fix those bugs, affecting also many more applications. In
response to our questions, they proposed to include the fixes into service
release u2 (eventually into service release u1, see
[http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/2011-July/005971.html]).
This means you cannot use Lucene/Solr with Java 7 releases before Update 2! If
you do, please don't open bug reports, it is not the committers' fault! At
least disable loop optimizations using the -XX:-UseLoopPredicate JVM options to
not risk index corruptions.
Please note: Also Java 6 users are affected, if you use one of those JVM
options, which are not enabled by default: -XX:+OptimizeStringConcat or
-XX:+AggressiveOpts
It is strongly recommended not to use any hotspot optimization switches in any
Java version without extensive testing!
In the case you upgrade to Java 7, remember that you have to reindex
everything, as the unicode version shipped with Java 7 changed and tokenization
behaves differently!
{quote}
I changed the text a little bit taking Hoss' and Steven's comments into account.
> Place warning about today's released Java7 version on Lucene/Solr/Root
> webpage's news and send mail to java-user
> ----------------------------------------------------------------------------------------------------------------
>
> Key: LUCENE-3349
> URL: https://issues.apache.org/jira/browse/LUCENE-3349
> Project: Lucene - Java
> Issue Type: Sub-task
> Components: general/test
> Environment: Java7
> Reporter: Uwe Schindler
> Assignee: Uwe Schindler
>
> Today, JDK/JRE 1.7.0 GA was released by Oracle. Unfortunately they didn't fix
> the Hotspot problems affecting loops to be miscompiled (LUCENE-3335,
> LUCENE-3346). This can lead to Solr crashing with default configuration on
> startup or sudden index corrumption depending on configuration.
> We should send an email to the java-user and solr-user list describing the
> problem. Also place a note in the news section of Solr, Lucene Core and
> top-level website.
> I propose the following text:
> {quote}
> *Jul 28th, 2011: WARNING: Index corruption and crashes in Apache Lucene Core
> / Apache Solr with Java 7*
> Oracle released *Java 7* today. Unfortunately it contains hotspot compiler
> optimizations, which miscompile some loops. This can affect code of several
> Apache projects. Sometimes JVMs only crash, but in several cases, results
> calculated can be incorrect, leading to bugs in applications (see Hotspot bug
> [http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7070134]).
> *Apache Lucene Core* and *Apache Solr* are two Apache projects, which are
> affected by these bugs, namely all versions released until today. Solr users
> with the default configuration will have Java crashing with SIGSEGV as soon
> as they start to index documents, as one affected part is the well-known
> Porter stemmer (see LUCENE-3335). Other loops in Lucene may be miscompiled,
> too, leading to index corruption (especially on Lucene trunk with pulsing
> codec; other loops may be affected, too - LUCENE-3346).
> These problems were detected only 5 days before the official Java 7 release,
> so Oracle had no time to fix those bugs, affecting also many more
> applications. In response to our questions, they proposed to include the
> fixes into service release u2 (eventually into service release u1, see
> [http://mail.openjdk.java.net/pipermail/hotspot-compiler-dev/2011-July/005971.html]).
> This means you cannot use Apache Lucene/Solr with Java 7 releases before
> Update 2! If you do, please don't open bug reports, it is not the committers'
> fault! At least disable loop optimizations using the -XX:-UseLoopPredicate
> JVM options to not risk index corruptions.
> Please note: Also Java 6 users are affected, if they use one of those JVM
> options, which are *not* enabled by default: -XX:+OptimizeStringConcat or
> -XX:+AggressiveOpts
> It is strongly recommended not to use any hotspot optimization switches in
> any Java version without extensive testing!
> In case you upgrade to Java 7, remember that you may have to reindex, as the
> unicode version shipped with Java 7 changed and tokenization behaves
> differently (e.g. lowercasing)!
> {quote}
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
