<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Hello there<br>
<br>
Our gluster client was crashed 2 times last night. The gluster
partition was unavailable for a long time and i have to remount it
manually. On the gluster log, i saw things like these:<br>
<br>
<i>pending frames:<br>
frame : type(1) op(LOOKUP)<br>
frame : type(1) op(LOOKUP)<br>
<br>
patchset: git://git.gluster.com/glusterfs.git<br>
signal received: 11<br>
time of crash: 2012-06-28 00:19:24<br>
configuration details:<br>
argp 1<br>
backtrace 1<br>
dlfcn 1<br>
fdatasync 1<br>
libpthread 1<br>
llistxattr 1<br>
setfsid 1<br>
spinlock 1<br>
epoll.h 1<br>
xattr.h 1<br>
st_atim.tv_nsec 1<br>
package-string: glusterfs 3.2.5<br>
/lib64/libc.so.6[0x3cf0c32900]</i><br>
<br>
<br>
Here is my "gluster volume info" output: <br>
<br>
<i>Volume Name: xxxxxx<br>
Type: Distribute<br>
Status: Started<br>
Number of Bricks: 2<br>
Transport-type: tcp<br>
Bricks:<br>
Brick1: Gnode1:/mnt/store<br>
Brick2: Gnode2:/mnt/store2<br>
Options Reconfigured:<br>
nfs.addr-namelookup: off<br>
nfs.rpc-auth-allow:
10.0.0.245,10.0.0.244,10.0.0.247,10.0.0.54,10.0.0.55<br>
auth.allow: 10.*<br>
cluster.min-free-disk: 5<br>
performance.io-thread-count: 64<br>
performance.cache-size: 512MB<br>
nfs.disable: off<br>
performance.write-behind-window-size: 4MB<br>
cluster.data-self-heal: off<br>
performance.stat-prefetch: off</i><br>
<br>
I googled around and found some people have the same problem with us
but haven't found solution or any clues of what happened yet. Is it
a bug ? and if it's could you please tell me how can i overcome it ?
<br>
<br>
Many thanks <br>
<br>
<br>
</body>
</html>