Hi Roberto,

whether you want to have lots of smaller indices or fewer larger indices is 
basically a tradeoff between different resources. The larger the indices, 
the longer searches will take when querying a small time range. The more 
indices you have, the more file handles are required and the bigger your 
Elasticsearch cluster states are (which are always transferred completely 
on Elasticsearch 1.x). Elasticsearch clusters with large indices also tend 
to take longer to restart.

See 
https://www.elastic.co/blog/performance-considerations-elasticsearch-indexing 
and https://qbox.io/blog/optimizing-elasticsearch-how-many-shards-per-index 
for some substantiated background.


Cheers,
Jochen

On Monday, 30 November 2015 14:42:46 UTC+1, [email protected] wrote:
>
> Dear, I have a Graylog 1.2 server which receives lot of messages per 
> seconds. I need to have a rotation strategy in order to mantain 6 months of 
> logs, and after that time the indexes will be deleted.
>
> I think I have to add this lines to the /etc/graylog/server/server.conf 
> file:
>
> rotation_strategy = time
>
> elasticsearch_max_time_per_index = 1d
>
> elasticsearch_max_number_of_indices = 180     # 6 months
>
> retention_strategy = delete
>
> Is this configuration OK???
>
> And also, is it a good idea to increase the 
> elasticsearch_max_time_per_index from 1d to 10d, and in this case 
> elasticsearch_max_number_of_indices = 18???
>
> Thanking in advance.
>
> Roberto.
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Graylog Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/graylog2/3fe4c4e9-b0e3-4c99-8b67-63060b610f33%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to