We have a NetSUS instance (Ubuntu VM) with a CIFS share attached to host our updates. The VM seems to be running out of space, but it's a bit of a puzzle to me. When I run $ df -h I get the following:
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/NetSUS--vg-root 199G 191G 0 100% /
udev 488M 4.0K 488M 1% /dev
tmpfs 100M 376K 100M 1% /run
none 5.0M 0 5.0M 0% /run/lock
none 497M 0 497M 0% /run/shm
overflow 1.0M 0 1.0M 0% /tmp
//CIFSShare/osx_updates 1.0T 491G 534G 48% /csc/updates
/dev/sda1 228M 181M 35M 84% /boot
Showing that the NetSUS volume is out of space. However, running a command to find out where the space is shows me that the space is in fact on the CIFS share, not on the VM drive itself (CIFS share is mounted via fstab to /CSC):
du -sk * | sort -n
0 initrd.img
0 initrd.img.old
0 proc
0 sys
0 tmp
0 vmlinuz
0 vmlinuz.old
4 dev
4 lib64
4 mnt
4 opt
4 selinux
12 media
12 root
16 lost+found
40 srv
60 home
376 run
6340 etc
8748 sbin
8852 bin
185321 boot
409668 var
1128316 usr
1300496 lib
166390811 csc
I've also run a find command and confirmed that practically all files over 50MB are found on the CIFS share.
The hard drive limit in VSphere is set to 203GB, so it seems like the CSC/CIFS share is being read as a physical part of the volume... We don't have this behavior with our NFS shares (NFS is not an option in this case).
Is this a CIFS/Linux/Vsphere idiosyncrasy?
Is enlarging the Hard drive setting of the VM in Vsphere the only option or am I missing something here?
