Take the 2-minute tour ×
Server Fault is a question and answer site for system and network administrators. It's 100% free, no registration required.

Currently i am testing a openstack-swift cluster made of 1 proxy node and 6 storage nodes. Each host has ubuntu 12.04. I put 5000 containers each having a text 4.6 KB. Each storage node has a 100 GB disk mounted and formatted as a xfs partition. After the test when i check disk usage on the storage nodes i came acroos a strange behaviour.

Mount point: /srv/node/sdb1

  • 2.0M accounts
  • 2.5G containers
  • 40M objects
  • 0 tmp

As you can containers directory having only db files for indexing takes too much space. Anyone come across this kind of behaviour before? What can be the reason for containers using this much space?

share|improve this question
    
What is your df -h output? –  ewwhite Dec 13 '12 at 12:22
    
/dev/sdb1 100G 2.5G 98G 3% /srv/node/sdb1 –  fga Dec 13 '12 at 12:27

1 Answer 1

@fga

Were you able to find reason for this? You can also try out ssbench to find out what is happening with your cluster. I would hope it will be helpful to your analysis.

share|improve this answer
    
Thank you for the answer.For tests sofar i have been using tsung. But i have heard of ssbench and been planning to run another test with version grizzly soon. –  fga Nov 14 '13 at 11:33

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.