Writing to NFS is working as intended for the most part, however the Windows client was found to be re-mounting /exporting the filesystem on what looks like a 15 minute interval. Exactly 15 minutes after the first export/write the Win client requests a re-export of share.
- Application continually streams HLS/HDS video content, so the remount / re-export breaks streaming functions. The streaming recovers but the interruption causes endpoint failures.
Configuration:
NFS server:
CentOS release 6.3 (Final) NFS Version 3.2.29 nfs-utils-1.2.3-36.el6.x86_64 nfs-utils-lib-1.1.5-6.el6.x86_64
NFS Exports: /export/somedir *(rw,sync,no_root_squash,insecure)
Windows NFS client settings:
Windows server 2008 r2 NFS support setup via add roles
Client Settings (all variations have been tried here I believe with no avail):
Protocol : TCP + UDP
Mounts: Use Soft Mounts, retry 5
File Security : Default Security flavor: sys
Client NFS access: anonymous, no AD or user mapping
The fact that this happens on a 15 minute interval immediately points to an intended software configuration like auth caching or export timeout, and at this point appears to be the Windows client forced to re-request the export.
Any suggestions, solutions are extremely appreciated.