Opscenter space usage

(2 posts) (2 voices)

• Started 1 year ago by apetcu
• Latest reply from mbulman
• This topic is not a support question

Tags:

No tags yet.

1. 

   apetcu
   Member

   What is "normal" Opscenter space usage?

   On my cluster (5 nodes) the directory /var/lib/cassandra/data/OpsCenter/ seems unjustifiably large. On all the nodes in the cluster, there is between 1 and 5 GB of data in this folder, after about a couple of weeks of operation. The culprits seem to be metrics-g-*-Data.db files taking up most of the space.

   Is this normal behaviour? At this rate, space usage is unsustainable, we need to constantly add storage to our cluster nodes.
   Is this a known issue, or do I have a problem with my installation?

   thanks

   Posted 1 year ago #
2. 

   mbulman
   Key Master

   This is the expected size for your cluster, but also a known issue. There will be changes in the next version of OpsCenter (July 11) in the way we store data that will reduce this size. This version will also stop collecting metrics for the system and OpsCenter keyspaces by default, which will cause less data to be stored.

   OpsCenter 1.2.1, slated for late July/early August, will also have more improvements to reduce data size. It will also expose an option to expire metrics. This should help significantly for people concerned with disk space who are willing to lose some historical data after a certain amount of time.

   If you notice you have a large number of SSTables for the metrics column family, running a compaction on each node (only if you can sacrifice the performance impact) should reduce the total disk usage.

   Posted 1 year ago #

RSS feed for this topic

Reply

You must log in to post.