Hide Forgot
what kernel version did you try this in? Can you try mounting glusterfs with -o direct-io-mode=off and see if this losetup works well? This must just be a documentation update in the FAQ section. Avati
Am I doing something wrong? [root@jacobgfs31-s1 ~]# mount -t glusterfs -o direct-io-mode=off localhost:/test /test /usr/sbin/glusterfs: unrecognized option `--disable-direct-io-mode' Try `glusterfs --help' or `glusterfs --usage' for more information. Mount failed. Please check the log file for more details. [root@jacobgfs31-s1 ~]# mount -t glusterfs -o direct-io-mode=disable localhost:/test /test /usr/sbin/glusterfs: unrecognized option `--disable-direct-io-mode' Try `glusterfs --help' or `glusterfs --usage' for more information. Mount failed. Please check the log file for more details. Is there a bug with disabling direct-io-mode in 3.1.1?
Kernel version is: 2.6.18-194.17.1.el5.centos.plus This is CentOS 5.5 64-bit
Try this instead - sh# glusterfs -s localhost --volfile-id volname --direct-io-mode=off /test
That worked.
(In reply to comment #5) > That worked. if you have a recent enough kernel (2.6.27+) it would have worked by default
IF I run the following command inside a Gluster volume mounted using the Gluster native protocol: mount -t glusterfs localhost:/volname /test cd /test dd if=/dev/zero of=./whatever bs=1024 count=1 seek=$((1*1024*1024)) losetup /dev/loop0 /test/whatever I get: ioctl: LOOP_SET_FD: Invalid argument The same works using NFS. A user Verio reported this.