nfs_inode_cache usage is high compared to older kernels on RHEL 5 and RHEL 6, and possibly system hangs with pdflush / flush consuming CPU

Solution Verified - Updated -

Issue

  • nfs_inode_cache usage is high compared to older kernels.
  • pdflush or flush processes consuming 100% CPU, or soft lockups with pdflush or flush process running in nfs_flush_inode
  • nfs_inode_cache grows uncontrolled and memory pressure does not release the memory.
  • Under certain type of load the NFS performance from RHEL 6.1 client becomes very slow. The case scenario: after writing a large number of small sized files in a sequence, the write speed for large sized files becomes very slow, around 5 MB/sec or even less. When writing large sized files is not preceded by a sequence of writing large number of small sized files the speed is normal - around 200 MB/sec.
  • nfs_inode_cache getting very big and eating the total amount of RAM

Environment

  • Red Hat Enteprise Linux 5.7 (2.6.18-274)
  • Red Hat Enteprise Linux 6.1 (2.6.32-131.0.15)
  • NFS client

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In