Incremental repair from the get go

[Jean-Francois Gosselin <jf...@gmail.com>]

On fresh install of Cassandra what's the best approach to start using
incremental repair from the get go (I'm using LCS) ?

Run nodetool repair -inc after inserting a few rows , or we still need to
follow the migration procedure with sstablerepairedset ?

>From the documentation "... If you use the leveled compaction strategy and
perform an incremental repair for the first time, Cassandra performs
size-tiering on all SSTables because the repair/unrepaired status is
unknown. This operation can take a long time. To save time, migrate to
incremental repair one node at a time. ..."

With almost no data size-tiering should be quick ?  Basically is there a
short cut to avoid the migration via sstablerepairedset  on a fresh install
?

Thanks

JF

Re: Incremental repair from the get go

[Robert Coli <rc...@eventbrite.com>]

On Mon, Nov 2, 2015 at 3:02 PM, Maciek Sakrejda <ma...@heroku.com> wrote:

> Following up on this older question: as per the docs, one *should* still
> do full repair periodically (the docs say weekly), right? And run
> incremental more often to fill in?
>

Something that amounts to full repair once every gc_grace_seconds, unless
you never do anything that results in a tombstone. In that (very rare)
case, one should probably still occasionally (2x a year?) run repair to
cover bitrot and similar (very rare) cases.

"Something that amounts to full repair" is either a full repair or an
incremental repair that covers 100% of the new data since gc_grace_seconds.

=Rob

Re: Incremental repair from the get go

[Maciek Sakrejda <ma...@heroku.com>]

Following up on this older question: as per the docs, one *should* still do
full repair periodically (the docs say weekly), right? And run incremental
more often to fill in?

Re: Incremental repair from the get go

[Marcus Eriksson <kr...@gmail.com>]

Starting up fresh it is totally OK to just start using incremental repairs

On Thu, Sep 3, 2015 at 10:25 PM, Jean-Francois Gosselin <
jfgosselin@gmail.com> wrote:

>
> On fresh install of Cassandra what's the best approach to start using
> incremental repair from the get go (I'm using LCS) ?
>
> Run nodetool repair -inc after inserting a few rows , or we still need to
> follow the migration procedure with sstablerepairedset ?
>
> From the documentation "... If you use the leveled compaction strategy
> and perform an incremental repair for the first time, Cassandra performs
> size-tiering on all SSTables because the repair/unrepaired status is
> unknown. This operation can take a long time. To save time, migrate to
> incremental repair one node at a time. ..."
>
> With almost no data size-tiering should be quick ?  Basically is there a
> short cut to avoid the migration via sstablerepairedset  on a fresh install
> ?
>
> Thanks
>
> JF
>