Bug #858
closed
cfuse craps out with fsstress
Added by Sage Weil about 13 years ago.
Updated over 7 years ago.
Description
Then I mount it via cfuse:
# mkdir /tmp/1
# cfuse -m 10.30.16.184:6789 /tmp/1
Here 10.30.16.184:6789 is the first monitor from generated
ceph.conf
It works. Doing the first test:
# mkdir /tmp/1/d
# fsstress -d /tmp/1/d -l 1 -n 1000 -p 10
iirc fsstress is in ltp-kernel-tests at /var/lib/ltp/somethingiforget/bin/fsstress
- Assignee set to Greg Farnum
What kind of issues are you expecting to crop up here? I've so far run across:
1) an issue with cosd spinning that I haven't diagnosed but appears to be an issue with btrfs and mounting cfuse in my ceph source dir,
2) an issue with an assert on subdirs in MDCache that I added debugging for and haven't seen again, and
3) absolutely nothing so far in a run that's been going for several hours.
Greg Farnum wrote:
What kind of issues are you expecting to crop up here? I've so far run across:
1) an issue with cosd spinning that I haven't diagnosed but appears to be an issue with btrfs and mounting cfuse in my ceph source dir,
[You're running btrfs on kai?] Spinning the cpu, or blocking?
2) an issue with an assert on subdirs in MDCache that I added debugging for and haven't seen again, and
Clustered mds? Did you save the logs?
3) absolutely nothing so far in a run that's been going for several hours.
That's more like it :).
1) Spinning the cpu. I have logs and a core file but haven't looked into them deeply. (one of my spare disks I use for some testing is btrfs.)
2) Single MDS.
I've picked up a few more crashes in the MDS, but so far no apparent issues with cfuse itself:
a) an assert failure in is_leaf_subtree, because the dir isn't even a subtree -- I added debug output to help with this and have yet to reproduce it....
b) an assert failure in _unlink_local due to rstat byte counts going negative.
c) A new segfault while printing debug output that I haven't looked at yet.
I think Sam found (a) and (b) in his kernel client testing, too. :/
- Status changed from New to Resolved
I have yet to see any cfuse issues with this. I will continue running and reopen if it does, but in the meantime I will create an umbrella bug and specific bugs to deal with the server daemon bits.
- Project changed from Ceph to CephFS
- Category deleted (
11)
- Target version deleted (
v0.25.1)
Bulk updating project=ceph category=ceph-fuse issues to move to fs project so that we can remove the ceph-fuse category from the ceph project
Also available in: Atom
PDF