Created attachment 523599 [details]
/var/log/messages and "thread apply all bt full" from cfuse process
Description of problem:
During Ceph 0.34 FUSE client testing we caused oddities in current/released RHEL 6.1 kernels. I'm attaching a simple repro, sysrq+w output from two hangs and one after I killed off the cfuse process the second time, and a "thread apply all bt full" from the cfuse userspace process. I can often, but not always, recover by kill -9'ing cfuse and umount'ing the cfuse mountpoint.
Version-Release number of selected component (if applicable):
ceph & ceph-fuse 0.34
100% in my environment
Steps to Reproduce:
1. Set up a Ceph server, set up a client keyring for client.yourhostname
2. cfuse /mnt --name=client.yourhostname
dd if=/dev/zero of=/mnt/file bs=1M count=1
Files created/removed, then mountpoint hangs and causes other odd kernel lockups.
Since RHEL 6.2 External Beta has begun, and this bug remains
unresolved, it has been rejected as it is not proposed as
exception or blocker.
Red Hat invites you to ask your support representative to
propose this request, if appropriate and relevant, in the
next release of Red Hat Enterprise Linux.
I got assigned to this bug, but it's a very old bug.
Lans, can you let me know if you can still reproduce it on newer versions of RHEL6? If so, can you provide kernel stack traces of the hung processes? in the messages file attached, no stack traces from kernel side are available.
Wow, this is a blast from the past. Go ahead and close this as unreproducible, and thanks for following up!
Thanks for the follow up Lans, feel free to re-open it if you step on this bug again.