Re: [PATCH v4 0/2] ceph: add debugfs entries signifying new mount syntax support

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, 2021-10-01 at 16:18 -0400, Patrick Donnelly wrote:
> On Fri, Oct 1, 2021 at 12:24 PM Jeff Layton <jlayton@xxxxxxxxxx> wrote:
> > Note that there is a non-zero chance that this will break teuthology in
> > some wa. In particular, looking at qa/tasks/cephfs/kernel_mount.py, it
> > does this in _get_global_id:
> > 
> >             pyscript = dedent("""
> >                 import glob
> >                 import os
> >                 import json
> > 
> >                 def get_id_to_dir():
> >                     result = {}
> >                     for dir in glob.glob("/sys/kernel/debug/ceph/*"):
> >                         mds_sessions_lines = open(os.path.join(dir, "mds_sessions")).readlines()
> >                         global_id = mds_sessions_lines[0].split()[1].strip('"')
> >                         client_id = mds_sessions_lines[1].split()[1].strip('"')
> >                         result[client_id] = global_id
> >                     return result
> >                 print(json.dumps(get_id_to_dir()))
> >             """)
> > 
> > 
> > What happens when this hits the "meta" directory? Is that a problem?
> > 
> > We may need to fix up some places like this. Maybe the open there needs
> > some error handling? Or we could just skip directories called "meta".
> 
> Yes, this will likely break all the kernel tests. It must be fixed
> before this can be merged into testing.
> 

Ok, I'll drop these patches for now. Let me know when it's clear to
merge them again, and I'll do so.

Thanks,
-- 
Jeff Layton <jlayton@xxxxxxxxxx>




[Index of Archives]     [CEPH Users]     [Ceph Large]     [Ceph Dev]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux