Kernel BUG at ffffffffa002f0cc - at Amazon EC2 Ubuntu 13.04 - NFS client

[Gabriel Pérez]

Failure occurred in three instances of Amazon EC2 small, happened at the same time in all three instances.

Servers each carrying a different operating time, one of them more than 48 hours and the other two only 10 hours or less.

Maybe not a bug cachefilesd, may be a bug in the "ephemeral storage" and keep the cache on this unit, but not sure. Amazon has not reported any failure on its website.

I bring all the data that I think may be useful.

Ubuntu: 13.04 updated

Linux version 3.8.0-25-generic (buildd@roseapple) (gcc version 4.7.3 (Ubuntu/Linaro 4.7.3-1ubuntu1) ) #37-Ubuntu SMP Thu Jun 6 20:47:07 UTC 2013 (Ubuntu 3.8.0-25.37-generic 3.8.13)

/etc/fstab datas:
nfs-host-server:/nfs/dir1 /exports/dir1 nfs rsize=32768,wsize=32768,vers=3,fsc,hard,intr,proto=tcp,port=2049,sync,noatime,auto 0 0
nfs-host-server:/nfs/dir2 /exports/dir2 nfs rsize=32768,wsize=32768,vers=3,fsc,hard,intr,proto=tcp,port=2049,sync,noatime,auto 0 0
# (Is the same nfs server mounted two times, the server has two HD and different security rules)

/etc/cachefilesd.conf
dir /mnt/
tag mycache
brun 10%
bcull 7%
bstop 3%
frun 10%
fcull 7%
fstop 3%

$mount
/dev/xvdb on /mnt type ext3 (rw,user_xattr,acl)

I install cachefilesd three days ago, in the tests was perfect, if it happens again try to find more data to find the cause, not currently know it can be.