Re: infinite osd operation RETRY loop

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

 



On Wed, Jan 2, 2019 at 5:16 PM Brad Hubbard <bhubbard@xxxxxxxxxx> wrote:
>
> On Thu, Jan 3, 2019 at 9:50 AM Gregory Farnum <gfarnum@xxxxxxxxxx> wrote:
> >
> > Well, there's an Op::out_bl, Op::outbl, and ObjectOperation::out_bl
> > and I'm having some trouble keeping them straight as I skim through
> > the code.
> >
> > But I do see a lot of stuff like
> > o->out_bl.swap(op.out_bl);
> > m->claim_data(*op->outbl);
> > vector<bufferlist*>::iterator pb = op->out_bl.begin();
> >
> > I'm still not clear on why having the previous result's data in there
> > doesn't work, but a new bit of "random" data does. But given that this
> > is just a client messing itself up I think we should just create a
> > ticket so we can track it, add a doc string somewhere saying
> > explicitly that output buffer lists need to be new/clean, and get on
> > with our lives. Unless somebody wants to walk through what's going on
> > in librados, which might be a good exercise?
>
> Challenge accepted.

I was playing around with this behavior in
94cb394d62936171196cd434b8306504bd9db796



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

  Powered by Linux