Per Lundqvist
2008-Jun-09 15:03 UTC
[Lustre-discuss] device entry within /proc/mounts differs from df
Hi, I recently discovered that the contents of /proc/mounts and the output from df differs for the lustre services on our 1.6 setup: # grep fs13 /proc/mounts /dev/vg /lustre/fs13_ost7 lustre ro 0 0 /dev/vg /lustre/fs13_ost8 lustre ro 0 0 # df -Pt lustre Filesystem 1024-blocks Used Available Capacity Mounted on /dev/vg0/fs13_ost7 4325506520 1876595716 2229187644 46% /lustre/fs13_ost7 /dev/vg1/fs13_ost8 4325506520 1876057960 2229725400 46% /lustre/fs13_ost8 i.e. the string /dev/vg within /proc/mounts is truncated - it should be something like /dev/vg0/fs13_ost7. I get the same result on the MDS and all the OSS:es for the same lustre setup and I haven''t found anything similar for any other /proc/mounts entry. Is this expected? Setup: Lustre 1.6.4.2 on top of LVM on Centos5. Kernel: 2.6.18-8.1.14.el5_lustre.1.6.4.2smp /Per -- Per Lundqvist National Supercomputer Centre Link?ping University, Sweden http://www.nsc.liu.se