Hi, I tried to use Gluster 1.3.12. The installation works fine, the remote files can be create and accessed normally. But when I try to loop mount a file(8G image), it will hang the system and makes it reboot. (loop mount a local file is ok) Gluster version is 1.3.12 Fuse, tried 2.7.4 and Gluster patched 2.7.4 Kernel is 2.6.18-8.1.15.1.16.el5xen (using Oracle Virtual Server, which is essentially Xen 3.1.4) I specified -d disable when launching glusterfs on client Mount command is: mount -o loop /path/to/file /mnt Last few lines of gluster log on client. 2009-04-22 05:08:28 D [fuse-bridge.c:529:fuse_lookup] glusterfs-fuse: 46: LOOKUP /dkt_disk1(14516228) 2009-04-22 05:08:28 D [fuse-bridge.c:384:fuse_entry_cbk] glusterfs-fuse: 46: (34) /dkt_disk1 => 14516228 2009-04-22 05:08:28 D [inode.c:367:__active_inode] fuse/inode: activating inode(14516228), lru=0/1024 2009-04-22 05:08:28 D [fuse-bridge.c:1553:fuse_open] glusterfs-fuse: 47: OPEN /dkt_disk1 2009-04-22 05:08:28 D [fuse-bridge.c:687:fuse_fd_cbk] glusterfs-fuse: 47: (12) /dkt_disk1 => 0x99ec2e8 last few lines of gluster log on server (2 servers, in unify) 2009-04-22 21:40:17 D [inode.c:367:__active_inode] brick-ns/inode: activating inode(14516227), lru=1/1024 2009-04-22 21:40:17 D [inode.c:367:__active_inode] brick/inode: activating inode(14516226), lru=0/1024 2009-04-22 21:40:17 D [inode.c:367:__active_inode] brick-ns/inode: activating inode(14516228), lru=1/1024 2009-04-22 21:40:23 D [inode.c:367:__active_inode] brick-ns/inode: activating inode(14516228), lru=1/1024 2009-04-22 21:40:23 D [inode.c:367:__active_inode] brick-ns/inode: activating inode(14516228), lru=1/1024 Any suggestion how to solve this issue? Thanks, Frank -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://zresearch.com/pipermail/gluster-users/attachments/20090422/66c2aa3f/attachment.htm>
