The backtrace of the crash shows this:
#0 pthread_spin_lock () at ../nptl/sysdeps/x86_64/pthread_spin_lock.S:24
#1 0x00007fe5345776a5 in fd_unref (fd=0x7fe523f7205c) at fd.c:553
#2 0x00007fe53482ba18 in glfs_io_async_cbk (op_ret=<optimized out>,
op_errno=0, frame=<optimized out>, cookie=0x7fe526c67040,
iovec=iovec@entry=0x0, count=count@entry=0)
at glfs-fops.c:839
#3 0x00007fe53482beed in glfs_fsync_async_cbk (frame=<optimized out>,
cookie=<optimized out>, this=<optimized out>, op_ret=<optimized out>,
op_errno=<optimized out>,
prebuf=<optimized out>, postbuf=0x7fe5217fe890, xdata=0x0) at
glfs-fops.c:1382
#4 0x00007fe520be2eb7 in ?? () from
/usr/lib/x86_64-linux-gnu/glusterfs/3.8.5/xlator/debug/io-stats.so
#5 0x00007fe5345d118a in default_fsync_cbk (frame=0x7fe52ceef3ac,
cookie=0x560ef95398e8, this=0x8, op_ret=0, op_errno=0, prebuf=0x1,
postbuf=0x7fe5217fe890, xdata=0x0) at defaults.c:1508
#6 0x00007fe5345d118a in default_fsync_cbk (frame=0x7fe52ceef204,
cookie=0x560ef95398e8, this=0x8, op_ret=0, op_errno=0, prebuf=0x1,
postbuf=0x7fe5217fe890, xdata=0x0) at defaults.c:1508
#7 0x00007fe525f78219 in dht_fsync_cbk (frame=0x7fe52ceef2d8,
cookie=0x560ef95398e8, this=0x0, op_ret=0, op_errno=0,
prebuf=0x7fe5217fe820, postbuf=0x7fe5217fe890, xdata=0x0)
at dht-inode-read.c:873
#8 0x00007fe5261bbc7f in client3_3_fsync_cbk (req=0x7fe525f78030
<dht_fsync_cbk>, iov=0x7fe526c61040, count=8, myframe=0x7fe52ceef130) at
client-rpc-fops.c:975
#9 0x00007fe5343201f0 in rpc_clnt_handle_reply (clnt=0x18,
clnt@entry=0x7fe526fafac0, pollin=0x7fe526c3a1c0) at rpc-clnt.c:791
#10 0x00007fe53432056c in rpc_clnt_notify (trans=<optimized out>,
mydata=0x7fe526fafaf0, event=<optimized out>, data=0x7fe526c3a1c0) at
rpc-clnt.c:962
#11 0x00007fe53431c8a3 in rpc_transport_notify (this=<optimized out>,
event=<optimized out>, data=<optimized out>) at rpc-transport.c:541
#12 0x00007fe5283e8d96 in socket_event_poll_in (this=0x7fe526c69440) at
socket.c:2267
#13 0x00007fe5283eaf37 in socket_event_handler (fd=<optimized out>, idx=5,
data=0x7fe526c69440, poll_in=1, poll_out=0, poll_err=0) at socket.c:2397
#14 0x00007fe5345ab3f6 in event_dispatch_epoll_handler
(event=0x7fe5217fecc0, event_pool=0x7fe526ca2040) at event-epoll.c:571
#15 event_dispatch_epoll_worker (data=0x7fe527c0f0c0) at event-epoll.c:674
#16 0x00007fe5324140a4 in start_thread (arg=0x7fe5217ff700) at
pthread_create.c:309
#17 0x00007fe53214962d in clone () at
../sysdeps/unix/sysv/linux/x86_64/clone.S:111
The fd being unreferenced contains this:
(gdb) print *fd
$6 = {
pid = 97649,
flags = 2,
refcount = 0,
inode_list = {
next = 0x7fe523f7206c,
prev = 0x7fe523f7206c
},
inode = 0x0,
lock = {
spinlock = 1,
mutex = {
__data = {
__lock = 1,
__count = 0,
__owner = 0,
__nusers = 0,
__kind = 0,
__spins = 0,
__elision = 0,
__list = {
__prev = 0x0,
__next = 0x0
}
},
__size = "\001", '\000' <repeats 38 times>,
__align = 1
}
},
_ctx = 0x7fe52ec31c40,
xl_count = 11,
lk_ctx = 0x7fe526c126a0,
anonymous = _gf_false
}
fd->inode is NULL, explaining the cause of the crash. We also see that
fd->refcount is already 0. So I'm wondering if this couldn't be an extra
fd_unref() introduced by that patch.
The crash seems to happen immediately after a graph switch.
Xavi
