Hi ,all<br><br> We found a strange problem. Some clients worked normally while some clients couldn't access sepcial files. For exmaple, Client A couldn't create the directory xxx, but Client B could. However, if Client B created the directory, Client A could acess it and even deleted it. But Client A still couldn't create the same directory later. If I changed the directory name, Client A worked without problems. It seemed that there were some problems with special bricks in special clients. But all the bricks were online.<br><br>I saw this in the logs in the GlusterFS client after creating directory failure:<br>[2017-11-06 11:55:18.420610] W [MSGID: 109011] [dht-layout.c:186:dht_layout_search] 0-data-dht: no subvolume for hash (value) = 4148753024<br>[2017-11-06 11:55:18.457744] W [fuse-bridge.c:521:fuse_entry_cbk] 0-glusterfs-fuse: 488: MKDIR() /xxx => -1 (Input/output error)<br>The message "W [MSGID: 109011] [dht-layout.c:186:dht_layout_search] 0-data-dht: no subvolume for hash (value) = 4148753024" repeated 3 times between [2017-11-06 11:55:18.420610] and [2017-11-06 11:55:18.457731]<br><div><div><br></div><div><br></div><div style="font-size: 12px;font-family: Arial Narrow;padding:2px 0 2px 0;">------------------ 原始邮件 ------------------</div><div style="font-size: 12px;background:#efefef;padding:8px;"><div><b>发件人:</b> "Ben&amp;nbsp;Turner";<bturner@redhat.com>;</div><div><b>发送时间:</b> 2017年11月5日(星期天) 凌晨3:00</div><div><b>收件人:</b> "acfreeman"<21291285@qq.com>;<wbr></div><div><b>抄送:</b> "gluster-users"<gluster-users@gluster.org>; <wbr></div><div><b>主题:</b> Re: [Gluster-users] glusterfs segmentation fault in rdma mode</div></div><div><br></div>This looks like there could be some some problem requesting / leaking / whatever memory but without looking at the core its tought to tell for sure. Note:<br><br>/usr/lib64/libglusterfs.so.0(_gf_msg_backtrace_nomem+0x78)[0x7f95bc54e618]<br><br>Can you open up a bugzilla and get us the core file to review?<br><br>-b<br><br>----- Original Message -----<br>> From: "自由人" <21291285@qq.com><br>> To: "gluster-users" <gluster-users@gluster.org><br>> Sent: Saturday, November 4, 2017 5:27:50 AM<br>> Subject: [Gluster-users] glusterfs segmentation fault in rdma mode<br>> <br>> <br>> <br>> Hi, All,<br>> <br>> <br>> <br>> <br>> I used Infiniband to connect all GlusterFS nodes and the clients. Previously<br>> I run IP over IB and everything was OK. Now I used rdma transport mode<br>> instead. And then I ran the traffic. After I while, the glusterfs process<br>> exited because of segmentation fault.<br>> <br>> <br>> <br>> <br>> Here were the messages when I saw segmentation fault:<br>> <br>> pending frames:<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(1) op(WRITE)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> frame : type(0) op(0)<br>> <br>> patchset: git:// git.gluster.org/glusterfs.git<br>> <br>> signal received: 11<br>> <br>> time of crash:<br>> <br>> 2017-11-01 11:11:23<br>> <br>> configuration details:<br>> <br>> argp 1<br>> <br>> backtrace 1<br>> <br>> dlfcn 1<br>> <br>> libpthread 1<br>> <br>> llistxattr 1<br>> <br>> setfsid 1<br>> <br>> spinlock 1<br>> <br>> epoll.h 1<br>> <br>> xattr.h 1<br>> <br>> st_atim.tv_nsec 1<br>> <br>> package-string: glusterfs 3.11.0<br>> <br>> /usr/lib64/ libglusterfs.so.0(_gf_msg_backtrace_nomem+0x78)[0x7f95bc54e618 ]<br>> <br>> /usr/lib64/ libglusterfs.so.0(gf_print_trace+0x324)[0x7f95bc557834 ]<br>> <br>> /lib64/ libc.so.6(+0x32510)[0x7f95bace2510 ]<br>> <br>> The client OS was CentOS 7.3. The server OS was CentOS 6.5. The GlusterFS<br>> version was 3.11.0 both in clients and servers. The Infiniband card was<br>> Mellanox. The Mellanox IB driver version was v4.1-1.0.2 (27 Jun 2017) both<br>> in clients and servers.<br>> <br>> <br>> Is rdma code stable for GlusterFS? Need I upgrade the IB driver or apply a<br>> patch?<br>> <br>> Thanks!<br>> <br>> _______________________________________________<br>> Gluster-users mailing list<br>> Gluster-users@gluster.org<br>> http://lists.gluster.org/mailman/listinfo/gluster-users</div>