Hi, This bug is fixed in the latest tla archive (mainline--2.5) Thanks and regards, Amar On 9/27/07, Alexey Filin <alexey.filin@xxxxxxxxx> wrote: > > Hi Amar, > > On 9/26/07, Amar S. Tumballi <amar@xxxxxxxxxxxxx> wrote: > > > > Hi Alexey, > > These trace log helped a lot.. I got the problem. Will be fixing the > bug > > soon. > > > > The issue is that, in older kernels fuse module sends 'creat()' call as > > 'mknod();open()' one after another. And sadly I had not implemented > striping > > option in mknod() fop. hence, the file is created in only one > node. With > > people using newer kernel version, this problem is not there. > > > > Thanks and Regards, > > Amar > > > yes, I looked at it myself, mknod() sets callback > > stripe_common_inode_cbk() > > which calls always > > dict_set (inode->ctx, this->name, data_from_int8 (1)); // not stripped > > looking forward to the bugfix :) > > regards, > > PS Debug prints work in callbacks. It was too late and I made a mistake in > sources during debugging > _______________________________________________ > Gluster-devel mailing list > Gluster-devel@xxxxxxxxxx > http://lists.nongnu.org/mailman/listinfo/gluster-devel > -- Amar Tumballi Gluster/GlusterFS Hacker [bulde on #gluster/irc.gnu.org] http://www.zresearch.com - Commoditizing Supercomputing and Superstorage!
