glusterfs的挂载点经常crash

guoli0813

HI，各位兄弟们，
  大家有没有将日志直接写入glusterfs的情况，举个例子：

将glusterfs进行挂载：
mount -t glusterfs db-host01.inc.mtime.com:/gv1 /gfs

业务模块的日志路径是/gfs /log-service/logs/2015-02-27/20150227-1000-1425002400.log 类似这种方式。在使用过程中，/gfs挂载点经常崩溃，查看/var/log/glusterfs/gfs.log，错误日志如下：

2015-02-27 02:00:00.011261] I [dht-common.c:1822:dht_lookup_cbk] 0-gv1-dht: Entry /log-service/logs/2015-02-27/20150227-1000-1425002400.log missing on subvol gv1-stripe-0
[2015-02-27 02:30:00.011621] I [dht-common.c:1822:dht_lookup_cbk] 0-gv1-dht: Entry /log-service/logs/2015-02-27/20150227-1030-1425004200.log missing on subvol gv1-stripe-0
[2015-02-27 02:30:00.027267] E [stripe.c:3669:stripe_writev] 0-gv1-stripe-0: fctx->xl_array[1] is NULL
pending frames:
frame : type(0) op(0)
frame : type(1) op(OPEN)
frame : type(0) op(0)
patchset: git://git.gluster.com/glusterfs.git
signal received: 11
time of crash:
2015-02-27 02:30:00
configuration details:
argp 1
backtrace 1
dlfcn 1
libpthread 1
llistxattr 1
setfsid 1
spinlock 1
epoll.h 1
xattr.h 1
st_atim.tv_nsec 1
package-string: glusterfs 3.6.1
/usr/lib64/libglusterfs.so.0(_gf_msg_backtrace_nomem+0xb6)[0x7f28ffcf5016]
/usr/lib64/libglusterfs.so.0(gf_print_trace+0x33f)[0x7f28ffd0fa8f]
/lib64/libc.so.6(+0x326a0)[0x7f28fed3e6a0]
/usr/lib64/glusterfs/3.6.1/xlator/cluster/distribute.so(dht_writev_cbk+0x26[0x7f28eeaa2ce8]
/usr/lib64/glusterfs/3.6.1/xlator/cluster/stripe.so(stripe_writev+0x87f)[0x7f28eecdd84f]
/usr/lib64/glusterfs/3.6.1/xlator/cluster/distribute.so(dht_writev+0x239)[0x7f28eeaa3099]
/usr/lib64/glusterfs/3.6.1/xlator/performance/write-behind.so(wb_fulfill_head+0x2e4)[0x7f28ee85dee4]
/usr/lib64/glusterfs/3.6.1/xlator/performance/write-behind.so(wb_fulfill+0x14b)[0x7f28ee85e07b]
/usr/lib64/glusterfs/3.6.1/xlator/performance/write-behind.so(wb_process_queue+0x92)[0x7f28ee85fa92]
/usr/lib64/glusterfs/3.6.1/xlator/performance/write-behind.so(wb_writev+0x2a[0x7f28ee860398]
/usr/lib64/glusterfs/3.6.1/xlator/performance/read-ahead.so(ra_writev+0x1ec)[0x7f28ee65057c]
/usr/lib64/glusterfs/3.6.1/xlator/performance/io-cache.so(ioc_writev+0x1b6)[0x7f28ee4421c6]
/usr/lib64/glusterfs/3.6.1/xlator/performance/quick-read.so(qr_writev+0x170)[0x7f28ee236000]


请问大家有碰到的没，或者给个解决思路

guoli0813

我是过来结束帖子的，经过很长时间的查找，发现glusterfs3.6.2版本中的条带式，不稳定，所以我更改了卷的类型。改为分布式复制方式，经过这两个case测试，没有出现类似问题