[sheepdog] [PATCH v2 1/3] journal: ignore invalid journal files

[Hitoshi Mitake]

At Thu, 11 Apr 2013 13:31:39 +0800,
Liu Yuan wrote:
> 
> On 04/10/2013 04:25 PM, Hitoshi Mitake wrote:
> > From: Hitoshi Mitake <mitake.hitoshi at gmail.com>
> > 
> > Even if journal files are broken, we can do nothing for recoverying
> > it. It is better to inform it as log messages and continue the
> > initialization process.
> > 
> > Signed-off-by: Hitoshi Mitake <mitake.hitoshi at lab.ntt.co.jp>
> > ---
> > 
> > v2: panic when do_recover() fails
> > 
> >  sheep/journal.c |   16 +++++++---------
> >  1 files changed, 7 insertions(+), 9 deletions(-)
> > 
> > diff --git a/sheep/journal.c b/sheep/journal.c
> > index 2e1d21f..59608ac 100644
> > --- a/sheep/journal.c
> > +++ b/sheep/journal.c
> > @@ -245,31 +245,29 @@ skip:
> >   * we actually only recover one jfile, the other would be empty. This process
> >   * is fast with buffered IO that only take several secends at most.
> >   */
> > -static int check_recover_journal_file(const char *p)
> > +static void check_recover_journal_file(const char *p)
> >  {
> >  	int old = 0, new = 0;
> >  
> >  	if (get_old_new_jfile(p, &old, &new) < 0)
> > -		return -1;
> > +		return;
> >  
> >  	/* No journal file found */
> >  	if (old == 0)
> > -		return 0;
> > +		return;
> >  
> >  	if (do_recover(old) < 0)
> > -		return -1;
> > +		panic("recoverying from journal file (old) failed");
> >  	if (do_recover(new) < 0)
> > -		return -1;
> > -
> > -	return 0;
> > +		panic("recoverying from journal file (new) failed");
> >  }
> >  
> >  int journal_file_init(const char *path, size_t size, bool skip)
> >  {
> >  	int fd;
> >  
> > -	if (!skip && check_recover_journal_file(path) < 0)
> > -		return -1;
> > +	if (!skip)
> > +		check_recover_journal_file(path);
> >  
> >  	jfile_size = (size * 1024 * 1024) / 2;
> >  
> > 
> 
> We need to handle empty jfile too (proceed when meeting empty jfile, not
> exist as now). Empty jfile could exist when sheep crash just between
> ftruncate() and prealloc() in commit_data().

I couldn't assume such a case. I'll do it in v2.

Thanks,
Hitoshi