You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cassandra.apache.org by Jonathan Ellis <jb...@gmail.com> on 2013/05/21 20:59:01 UTC
Time to roll 1.1.12?
CHANGES:
* Remove buggy thrift max message length option (CASSANDRA-5529)
* Add retry mechanism to OTC for non-droppable_verbs (CASSANDRA-5393)
* Use allocator information to improve memtable memory usage estimate
(CASSANDRA-5497)
* fsync leveled manifest to avoid corruption (CASSANDRA-5535)
* Fix Bound intersection computation (CASSANDRA-5551)
* Fix NPE in Pig's widerow mode (CASSANDRA-5488)
I'm feeling lucky that we're ready to stick a fork in 1.1.
--
Jonathan Ellis
Project Chair, Apache Cassandra
co-founder, http://www.datastax.com
@spyced
Re: Time to roll 1.1.12?
Posted by Jason Brown <ja...@netflix.com>.
+1
On Tue, May 21, 2013 at 1:43 PM, Brandon Williams <dr...@gmail.com> wrote:
> +1
>
> On Tue, May 21, 2013 at 1:59 PM, Jonathan Ellis <jb...@gmail.com> wrote:
>
> > CHANGES:
> > * Remove buggy thrift max message length option (CASSANDRA-5529)
> > * Add retry mechanism to OTC for non-droppable_verbs (CASSANDRA-5393)
> > * Use allocator information to improve memtable memory usage estimate
> > (CASSANDRA-5497)
> > * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
> > * Fix Bound intersection computation (CASSANDRA-5551)
> > * Fix NPE in Pig's widerow mode (CASSANDRA-5488)
> >
> > I'm feeling lucky that we're ready to stick a fork in 1.1.
> >
> > --
> > Jonathan Ellis
> > Project Chair, Apache Cassandra
> > co-founder, http://www.datastax.com
> > @spyced
> >
>
Re: Time to roll 1.1.12?
Posted by Brandon Williams <dr...@gmail.com>.
+1
On Tue, May 21, 2013 at 1:59 PM, Jonathan Ellis <jb...@gmail.com> wrote:
> CHANGES:
> * Remove buggy thrift max message length option (CASSANDRA-5529)
> * Add retry mechanism to OTC for non-droppable_verbs (CASSANDRA-5393)
> * Use allocator information to improve memtable memory usage estimate
> (CASSANDRA-5497)
> * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
> * Fix Bound intersection computation (CASSANDRA-5551)
> * Fix NPE in Pig's widerow mode (CASSANDRA-5488)
>
> I'm feeling lucky that we're ready to stick a fork in 1.1.
>
> --
> Jonathan Ellis
> Project Chair, Apache Cassandra
> co-founder, http://www.datastax.com
> @spyced
>
Re: Time to roll 1.1.12?
Posted by Omid Aladini <om...@gmail.com>.
Regarding fork sticking, it'd be great if we knew the scope of this:
https://issues.apache.org/jira/browse/CASSANDRA-5374
which I have a snapshot for, but not yet checked if I can reproduce.
On Tue, May 21, 2013 at 8:59 PM, Jonathan Ellis <jb...@gmail.com> wrote:
> CHANGES:
> * Remove buggy thrift max message length option (CASSANDRA-5529)
> * Add retry mechanism to OTC for non-droppable_verbs (CASSANDRA-5393)
> * Use allocator information to improve memtable memory usage estimate
> (CASSANDRA-5497)
> * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
> * Fix Bound intersection computation (CASSANDRA-5551)
> * Fix NPE in Pig's widerow mode (CASSANDRA-5488)
>
> I'm feeling lucky that we're ready to stick a fork in 1.1.
>
> --
> Jonathan Ellis
> Project Chair, Apache Cassandra
> co-founder, http://www.datastax.com
> @spyced
Re: Time to roll 1.1.12?
Posted by Sylvain Lebresne <sy...@datastax.com>.
+1. I'll cook up the artifacts and call for a vote.
On Wed, May 22, 2013 at 7:35 AM, Radim Kolar <hs...@filez.com> wrote:
>
> * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
>>
> you sure that this does not have performance impact? most filesystems sync
> all their data not just one file. write to .new file and then do rename.
>
manifest fsync
Posted by Radim Kolar <hs...@filez.com>.
>> should be fine, the manifest is not rewritten that often
its rewritten after each sstable flush? databases should do fsync only
in checkpoint.
Fsync scenario in WAFL is that hard checkpoint is done after predefined
number of log segments. On checkpoint fsync everything and then write
checkpoint marker to log, then remove checkpointed logs.
Re: Time to roll 1.1.12?
Posted by Marcus Eriksson <kr...@gmail.com>.
On Wed, May 22, 2013 at 7:35 AM, Radim Kolar <hs...@filez.com> wrote:
>
> * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
>>
> you sure that this does not have performance impact? most filesystems sync
> all their data not just one file. write to .new file and then do rename.
>
should be fine, the manifest is not rewritten that often
Re: Time to roll 1.1.12?
Posted by Radim Kolar <hs...@filez.com>.
> * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
you sure that this does not have performance impact? most filesystems
sync all their data not just one file. write to .new file and then do
rename.
Re: Time to roll 1.1.12?
Posted by Pavel Yaskevich <po...@gmail.com>.
+1
On Tue, May 21, 2013 at 2:10 PM, Carl Yeksigian <ca...@yeksigian.com> wrote:
> +1 on the release, and +1 on the fork sticking :)
>
>
> On Tue, May 21, 2013 at 2:59 PM, Jonathan Ellis <jb...@gmail.com> wrote:
>
> > CHANGES:
> > * Remove buggy thrift max message length option (CASSANDRA-5529)
> > * Add retry mechanism to OTC for non-droppable_verbs (CASSANDRA-5393)
> > * Use allocator information to improve memtable memory usage estimate
> > (CASSANDRA-5497)
> > * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
> > * Fix Bound intersection computation (CASSANDRA-5551)
> > * Fix NPE in Pig's widerow mode (CASSANDRA-5488)
> >
> > I'm feeling lucky that we're ready to stick a fork in 1.1.
> >
> > --
> > Jonathan Ellis
> > Project Chair, Apache Cassandra
> > co-founder, http://www.datastax.com
> > @spyced
> >
>
Re: Time to roll 1.1.12?
Posted by Carl Yeksigian <ca...@yeksigian.com>.
+1 on the release, and +1 on the fork sticking :)
On Tue, May 21, 2013 at 2:59 PM, Jonathan Ellis <jb...@gmail.com> wrote:
> CHANGES:
> * Remove buggy thrift max message length option (CASSANDRA-5529)
> * Add retry mechanism to OTC for non-droppable_verbs (CASSANDRA-5393)
> * Use allocator information to improve memtable memory usage estimate
> (CASSANDRA-5497)
> * fsync leveled manifest to avoid corruption (CASSANDRA-5535)
> * Fix Bound intersection computation (CASSANDRA-5551)
> * Fix NPE in Pig's widerow mode (CASSANDRA-5488)
>
> I'm feeling lucky that we're ready to stick a fork in 1.1.
>
> --
> Jonathan Ellis
> Project Chair, Apache Cassandra
> co-founder, http://www.datastax.com
> @spyced
>