Memory can be exhausted while accessing file with sticky bit set over nfs/ufs file systems
Last updated on APRIL 10, 2017
Applies to:Solaris Operating System - Version 10 1/13 U11 to 11.2 [Release 10.0 to 11.0]
Information in this document applies to any platform.
When large files are created, read, or written in a UFS/NFS file systems (e.g. by ufsdump or mkfile), memory of the system can be exhausted by large page cache and never been kicked out of memory even the system is running short of memory.
A large file is created by mkfile command in UFS/NFS file systems. This will create the sticky bit on the file by default (up to Solaris 11.2). Below is an example of page cache usage while a 20G created is being created by mkfile. 12G memory is being used out of total physical memory of 16G.
Another known scenario is to run ufsdump in a guest LDOM (e.g. "ufsdump 0f /mnt/file.dump /dev/rdsk/c0d1s0") where both devices are NFS mounted from files in the control domain and both files have the sticky bit set.
Sign In with your My Oracle Support account
Don't have a My Oracle Support account? Click to get started
Million Knowledge Articles and hundreds of Community platforms