You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by Oleksandr Shulgin <ol...@zalando.de> on 2018/12/02 05:24:29 UTC

Re: Problem with restoring a snapshot using sstableloader

On Fri, 30 Nov 2018, 17:54 Oliver Herrmann <o.herrmann217@gmail.com wrote:

> When using nodetool refresh I must have write access to the data folder
> and I have to do it on every node. In our production environment the user
> that would do the restore does not have write access to the data folder.
>

OK, not entirely sure that's a reasonable setup, but do you imply that with
sstableloader you don't need to process every snapshot taken -- that is,
also visiting every node? That would only be true if your replication
factor equals to the number of nodes, IMO.

--
Alex

Re: Problem with restoring a snapshot using sstableloader

Posted by Oleksandr Shulgin <ol...@zalando.de>.
On Mon, Dec 3, 2018 at 4:24 PM Oliver Herrmann <o....@gmail.com>
wrote:

>
> You are right. The number of nodes in our cluster is equal to the
> replication factor. For that reason I think it should be sufficient to call
> sstableloader only from one node.
>

The next question is then: do you care about consistency of data restored
from one snapshot?  Is the snapshot taken after repair?  Do you still write
to those tables?

In other words, your data will be consistent after restoring from one
node's snapshot only if you were writing with consistency level ALL (or
equal to your replication factor and, transitively, to the number of nodes).

-- 
Oleksandr "Alex" Shulgin | Senior Software Engineer | Team Flux | Data
Services | Zalando SE | Tel: +49 176 127-59-707

Re: Problem with restoring a snapshot using sstableloader

Posted by Oliver Herrmann <o....@gmail.com>.
Am So., 2. Dez. 2018 um 06:24 Uhr schrieb Oleksandr Shulgin <
oleksandr.shulgin@zalando.de>:

> On Fri, 30 Nov 2018, 17:54 Oliver Herrmann <o.herrmann217@gmail.com wrote:
>
>> When using nodetool refresh I must have write access to the data folder
>> and I have to do it on every node. In our production environment the user
>> that would do the restore does not have write access to the data folder.
>>
>
> OK, not entirely sure that's a reasonable setup, but do you imply that
> with sstableloader you don't need to process every snapshot taken -- that
> is, also visiting every node? That would only be true if your replication
> factor equals to the number of nodes, IMO.
>

You are right. The number of nodes in our cluster is equal to the
replication factor. For that reason I think it should be sufficient to call
sstableloader only from one node.