Cassandra snapshot

[ng <pi...@gmail.com>]

I need to make sure that all the data in sstable before taking the snapshot.

I am thinking of
nodetool cleanup
nodetool repair
nodetool flush
nodetool snapshot

Am I missing anything else?

Thanks in advance for the responses/suggestions.

ng

Re: Cassandra snapshot

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

On Mon, Jun 2, 2014 at 12:18 PM, ng <pi...@gmail.com> wrote:

>
> I need to make sure that all the data in sstable before taking the
> snapshot.
>
> I am thinking of
> nodetool cleanup
>

Cleanup does nothing but waste i/o if you have not recently added, removed,
or replaced nodes.


> nodetool repair
>

Repair cannot actually close the consistency window while writes are
incoming, it can only shorten it.


> nodetool flush
>
nodetool snapshot
>
> Am I missing anything else?
>

Your requirement seems a bit confused. If your goal is to close the
consistency window, it's impossible without turning off writes or writing
everything at CL.ALL. If your goal is not to do that, the above "flush" is
unnecessary. Also, if I recall correctly, "snapshot" flushes as part of the
snapshot process, so it's redundant.

Most people rely on replication + backups to make the consistency window
short enough that they don't care too much about "making sure all the data
in sstable before taking snapshot."

=Rob

Re: Cassandra snapshot

[Jack Krupansky <ja...@basetechnology.com>]

You might check the doc:
http://www.datastax.com/documentation/cassandra/2.0/cassandra/operations/ops_backup_restore_c.html

-- Jack Krupansky

From: ng 
Sent: Monday, June 2, 2014 3:18 PM
To: user@cassandra.apache.org 
Subject: Cassandra snapshot


I need to make sure that all the data in sstable before taking the snapshot.

I am thinking of

nodetool cleanup
nodetool repair
nodetool flush
nodetool snapshot

Am I missing anything else?

Thanks in advance for the responses/suggestions.

ng

Re: Cassandra snapshot

[Jeremy Jongsma <je...@barchart.com>]

I wouldn't recommend doing this before regular backups for the simple
reason that for large data sets it will take a looooong time to run, and
will require that your node backup schedule be properly staggered (you
should never be running repair on all nodes at the same time.) Backups
should be treated as eventually consistent just like Cassandra itself.

That said, if you are doing a one-time backup of a node and for whatever
reason you want it as up-to-date as possible without unnecessary data, you
should also run nodetool compact.


On Mon, Jun 2, 2014 at 2:18 PM, ng <pi...@gmail.com> wrote:

>
> I need to make sure that all the data in sstable before taking the
> snapshot.
>
> I am thinking of
> nodetool cleanup
> nodetool repair
> nodetool flush
> nodetool snapshot
>
> Am I missing anything else?
>
> Thanks in advance for the responses/suggestions.
>
> ng
>