Description of problem: Recently,I upgrade from version 3.6.4 to version 3.7.13 in four CentOS-7-x86_64-1511 nodes.I create striped replicated volume with command 'gluster volume create vol0 stripe 2 replica 2 172.16.52.{115,108,100,117}:/data/gluster/brick0/brick'. When I copy directories and files into volume by using fuse it didn't work correct, cannot create regular file or cannot create directory happened. Glusterfs version 3.6.4 work fine with the same configuration. Glusterfs version 3.7.13 work fine when using distributed replicated volume. Version-Release number of selected component (if applicable): I install glusterfs with source code glusterfs-3.7.13.tar.gz How reproducible: Steps to Reproduce: 1.Need four linux nodes(my nodes os:CentOS-7-x86_64-1511). 2.Install glusterfs with source code glusterfs-3.7.13.tar.gz. 3.Create striped replicated volume with command like 'gluster volume create vol0 stripe 2 replica 2 172.16.52.{115,108,100,117}:/data/gluster/brick0/brick'. 4.Start volume: gluster volume start vol0. 5.Mount glustergs volume: mount -t glusterfs localhost:/vol0 /mnt/vol0 6.Copy some not empty directory to volume: cp -r /home/admin/Downloads /mnt/vol0/ Actual results: Expected results: Additional info: 1,glusterfs version 3.6.4 work fine with the same configuration. 2,glusterfs version 3.7.13 work fine when using distributed replicated volume. 3,my steps: 1),install glusterfs with source code glusterfs-3.7.13.tar.gz in all 4 nodes. 2),in node 172.16.52.115,executed next command: [root@localhost vol0]#gluster peer probe 172.16.52.108 [root@localhost vol0]#gluster peer probe 172.16.52.100 [root@localhost vol0]#gluster peer probe 172.16.52.117 [root@localhost vol0]#gluster volume create vol0 stripe 2 replica 2 172.16.52.{115,108,100,117}:/data/gluster/brick0/brick [root@localhost vol0]#gluster volume start vol0 [root@localhost vol0]#mount -t glusterfs localhost:/vol0 /mnt/vol0 [root@localhost vol0]#cp -r /home/admin/Downloads ./ ...... cp: cannot create regular file ‘./Downloads/install/glusterfs-3.7.13.tar.gz’: No such file or directory cp: cannot create directory ‘./Downloads/install/glusterfs-3.7.13’: No data available ...... [root@localhost vol0]# ls ls: cannot access Downloads: No data available Downloads [root@localhost vol0]# rm -rf Downloads rm: cannot remove ‘Downloads’: No data available [root@localhost vol0]# gluster volume info Volume Name: vol0 Type: Striped-Replicate Volume ID: e17a8660-2eb4-45cb-9cdf-8b1785893f00 Status: Started Number of Bricks: 1 x 2 x 2 = 4 Transport-type: tcp Bricks: Brick1: 172.16.52.115:/data/gluster/brick0/brick Brick2: 172.16.52.108:/data/gluster/brick0/brick Brick3: 172.16.52.100:/data/gluster/brick0/brick Brick4: 172.16.52.117:/data/gluster/brick0/brick [root@localhost vol0]# gluster volume heal vol0 info Brick 172.16.52.115:/data/gluster/brick0/brick Status: Connected Number of entries: 0 Brick 172.16.52.108:/data/gluster/brick0/brick Status: Connected Number of entries: 0 Brick 172.16.52.100:/data/gluster/brick0/brick Status: Connected Number of entries: 0 Brick 172.16.52.117:/data/gluster/brick0/brick Status: Connected Number of entries: 0 [root@localhost vol0]# gluster volume status Status of volume: vol0 Gluster process TCP Port RDMA Port Online Pid ------------------------------------------------------------------------------ Brick 172.16.52.115:/data/gluster/brick0/br ick 49152 0 Y 2698 Brick 172.16.52.108:/data/gluster/brick0/br ick 49152 0 Y 2622 Brick 172.16.52.100:/data/gluster/brick0/br ick 49152 0 Y 2300 Brick 172.16.52.117:/data/gluster/brick0/br ick 49152 0 Y 2466 NFS Server on localhost 2049 0 Y 2720 Self-heal Daemon on localhost N/A N/A Y 2727 NFS Server on 172.16.52.108 2049 0 Y 2644 Self-heal Daemon on 172.16.52.108 N/A N/A Y 2651 NFS Server on 172.16.52.100 2049 0 Y 2323 Self-heal Daemon on 172.16.52.100 N/A N/A Y 2329 NFS Server on 172.16.52.117 2049 0 Y 2488 Self-heal Daemon on 172.16.52.117 N/A N/A Y 2495 Task Status of Volume vol0 ------------------------------------------------------------------------------ There are no active volume tasks [root@localhost vol0]# cat /var/log/glusterfs/mnt-vol0.log | grep " E " [2016-08-20 08:11:28.347465] E [MSGID: 114031] [client-rpc-fops.c:321:client3_3_mkdir_cbk] 0-vol0-client-3: remote operation failed. Path: /Downloads [No data available] [2016-08-20 08:11:28.347488] E [MSGID: 114031] [client-rpc-fops.c:321:client3_3_mkdir_cbk] 0-vol0-client-2: remote operation failed. Path: /Downloads [No data available]
The stripe type volume is not maintained anymore. Instead you can try out Sharded volume which have the same behaviour. Please try it on sharded volume and check whether it is reproducing.
(In reply to Jiffin from comment #1) > The stripe type volume is not maintained anymore. > Instead you can try out Sharded volume which have the same behaviour. > Please try it on sharded volume and check whether it is reproducing. But how to create Sharded volume? There is no mention in gluster volume create command: volume create <NEW-VOLNAME> [stripe <COUNT>] [replica <COUNT> [arbiter <COUNT>]] [disperse [<COUNT>]] [disperse-data <COUNT>] [redundancy <COUNT>] [transport <tcp|rdma|tcp,rdma>] <NEW-BRICK>?<vg_name>... [force]
You'd need to create a volume and then execute gluster volume set <volname> features.shard on
This bug is getting closed because GlusteFS-3.7 has reached its end-of-life. Note: This bug is being closed using a script. No verification has been performed to check if it still exists on newer releases of GlusterFS. If this bug still exists in newer GlusterFS releases, please reopen this bug against the newer release.