Run compilebench on the mountpoint, for example: ./compilebench -D /mnt/replicate/ -i 25 -r 75 Do some graph changes repeatedly on the servers for i in `seq 10`; do gluster volume set quick-read on; gluster volume set quick-read off; done glusterfs client crashes. And on the subsequent mounts, mkdir and rmdir crashes the client. ========================== [2012-12-18 10:45:03.903009] W [dht-diskusage.c:45:dht_du_info_cbk] 0-rep-dht: failed to get disk info from rep-replicate-1 [2012-12-18 10:45:03.903019] W [dht-layout.c:179:dht_layout_search] 0-rep-dht: no subvolume for hash (value) = 3441719530 [2012-12-18 10:45:03.903032] W [fuse-bridge.c:417:fuse_entry_cbk] 0-glusterfs-fuse: 33: MKDIR() /jon => -1 (Invalid argument) [2012-12-18 10:45:06.154077] W [common-utils.c:2296:gf_ports_reserved] 0-glusterfs-socket: is not a valid port identifier [2012-12-18 10:45:06.154301] E [socket.c:2695:socket_connect] 0-rep-client-1: connection attempt failed (Connection refused) [2012-12-18 10:45:06.158675] W [common-utils.c:2296:gf_ports_reserved] 0-glusterfs-socket: is not a valid port identifier [2012-12-18 10:45:06.158860] E [socket.c:2695:socket_connect] 0-rep-client-2: connection attempt failed (Connection refused) [2012-12-18 10:45:06.163262] W [common-utils.c:2296:gf_ports_reserved] 0-glusterfs-socket: is not a valid port identifier [2012-12-18 10:45:06.163478] E [socket.c:2695:socket_connect] 0-rep-client-3: connection attempt failed (Connection refused) [2012-12-18 10:45:09.168510] W [common-utils.c:2296:gf_ports_reserved] 0-glusterfs-socket: is not a valid port identifier [2012-12-18 10:45:09.168784] E [socket.c:2695:socket_connect] 0-rep-client-1: connection attempt failed (Connection refused) [2012-12-18 10:45:09.173340] W [common-utils.c:2296:gf_ports_reserved] 0-glusterfs-socket: is not a valid port identifier [2012-12-18 10:45:09.173687] E [socket.c:2695:socket_connect] 0-rep-client-2: connection attempt failed (Connection refused) [2012-12-18 10:45:09.178247] W [common-utils.c:2296:gf_ports_reserved] 0-glusterfs-socket: is not a valid port identifier [2012-12-18 10:45:09.178532] E [socket.c:2695:socket_connect] 0-rep-client-3: connection attempt failed (Connection refused) [2012-12-18 10:45:10.822603] I [afr-common.c:3874:afr_local_init] 0-rep-replicate-1: no subvolumes up pending frames: frame : type(1) op(LOOKUP) frame : type(0) op(0) patchset: git://git.gluster.com/glusterfs.git signal received: 11 time of crash: 2012-12-18 10:45:10configuration details: argp 1 backtrace 1 dlfcn 1 fdatasync 1 libpthread 1 llistxattr 1 setfsid 1 spinlock 1 epoll.h 1 xattr.h 1 st_atim.tv_nsec 1 package-string: glusterfs 3.4.0qa5 /lib64/libc.so.6[0x358ba32900] /usr/lib64/glusterfs/3.4.0qa5/xlator/cluster/distribute.so(dht_inode_ctx_time_update+0x4e)[0x7fc7821aec9e] /usr/lib64/glusterfs/3.4.0qa5/xlator/cluster/distribute.so(dht_revalidate_cbk+0x1b1)[0x7fc7821c9291] /usr/lib64/glusterfs/3.4.0qa5/xlator/cluster/replicate.so(afr_lookup+0x20c)[0x7fc7824373fc] /usr/lib64/glusterfs/3.4.0qa5/xlator/cluster/distribute.so(dht_lookup+0x275)[0x7fc7821c7635] /usr/lib64/libglusterfs.so.0(default_lookup+0x6d)[0x7fc786e4ec3d]
Created attachment 665443 [details] Core
looks similar to bug 881013. Can you please work on this?
Patch is on review http://review.gluster.com/#change,4338.
Tried with TC: https://tcms.engineering.redhat.com/case/243282/?from_plan=7656 unable to reproduce.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-1262.html