Tuning the ulimit -s down to 10240 solves the issue. That specific machine has a modified limit.conf. I know that with ulimit -s too high, the number of threads that can be created is limited (I'm not sure whether the reasoning applies here). For what's it's worth, tuning the stack size down solves the issue. Robin From: Microsoft Office User <robinr at muohio.edu<mailto:robinr at muohio.edu>> Date: Fri, 2 Mar 2012 13:48:17 -0500 To: "gluster-users at gluster.org<mailto:gluster-users at gluster.org>" <gluster-users at gluster.org<mailto:gluster-users at gluster.org>> Subject: Error message meaning Hi, I've several machines mounting my gluster volume just fine. However, I also have a set of machines fail to mount the same volume. Gluster-3.2.5 CentOS 5; Linux compute-1-2.local 2.6.18-274.18.1.el5 #1 SMP Thu Feb 9 12:45:44 EST 2012 x86_64 x86_64 x86_64 GNU/Linux Error message from the log: [2012-03-02 13:40:12.79630] I [glusterfsd.c:1493:main] 0-/opt/glusterfs/3.2.5/sbin/glusterfs: Started running /opt/glusterfs/3.2.5/sbin/glusterfs version 3.2.5 [2012-03-02 13:40:12.105006] E [glusterfsd-mgmt.c:740:mgmt_rpc_notify] 0-glusterfsd-mgmt: failed to connect with remote-host: Cannot allocate memory [2012-03-02 13:40:12.105300] W [glusterfsd.c:727:cleanup_and_exit] (-->/opt/glusterfs/3.2.5/sbin/glusterfs(glusterfs_mgmt_init+0x1bd) [0x40672d] (-->/opt/glusterfs/3.2.5/lib64/libgfrpc.so.0(rpc_clnt_start+0x17) [0x2b8b196bd8a7] (-->/opt/glusterfs/3.2.5/sbin/glusterfs [0x407c8e]))) 0-: received signum (1), shutting down The machine has 23 GB of free RAM. Suggestions ? Thanks. Robin -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://gluster.org/pipermail/gluster-users/attachments/20120305/df98c5a7/attachment.htm>
