Why are df and ls -l so slow on a GFS filesystem?
This depends on how much activity is or has been occurring. For a
df to complete all the nodes have to update any information held in cache. The same is true of any stat operation like
ls -l. All nodes in the cluster have to write all their data to disk in order for the information to be accurate.
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.