Hi Satoru, I just give a reply to the former 2 mails together here.
On Wed, 2014-07-30 at 15:36 +0900, Satoru Takeuchi wrote:
> Hi Gui,
>
> (2014/07/30 15:03), Satoru Takeuchi wrote:
> > Hi Gui,
> >
> > (2014/07/17 11:40), Gui Hecheng wrote:
> >> o Return 0 to indicate success,
> >> when detected errors were corrected during scrubbing.
> >> P.s. This is also to facilitate scripting when return value
> >> is to be checked.
> >> o Warn the users if there are uncorrectable errors detected.
> >>
> >> Signed-off-by: Gui Hecheng <guihc.fnst@xxxxxxxxxxxxxx>
> >> ---
> >> cmds-scrub.c | 13 ++++++++-----
> >> 1 file changed, 8 insertions(+), 5 deletions(-)
> >>
> >> diff --git a/cmds-scrub.c b/cmds-scrub.c
> >> index 5265a2b..f9e2b40 100644
> >> --- a/cmds-scrub.c
> >> +++ b/cmds-scrub.c
> >> @@ -1514,14 +1514,17 @@ out:
> >> }
> >> close_file_or_dir(fdmnt, dirstream);
> >>
> >> - if (nothing_to_resume)
> >> - return 2;
> >> if (err)
> >> return 1;
> >> - if (e_correctable)
> >> + if (nothing_to_resume)
> >> + return 2;
> >> + if (e_uncorrectable) {
> >> + ERR(!do_quiet, "ERROR: There are uncorrectable errors.\n");
> >> return 3;
> >> - if (e_uncorrectable)
> >> - return 4;
> >> + }
> >> + if (e_correctable)
> >> + ERR(!do_quiet, "WARNING: errors detected during scrubbing, corrected.\n");
> >> +
> >
> > 1. ERR() messages are not necessary since start command
> > reports its status in the middle of scrub_start().
> > It includes both the number of correctable/uncorrectable errors.
Ok, it may not seems "neccessary", but these ERR()s serve as
conclusions at the end to the users. Because users care whether a scrub
is able to detect errors and if yes, whether the errors are corrected.
So it adds to user experience.
> > 2. It prints messages even if this program runs as foreground mode.
> > Checking do_print is necessary.
The do_print serves to control whether "sub procedures(such as
pthread_create/join, scrub_write_progress etc.)" failed or not in
foreground mode, so do_print should not effect conclusion outputs since
all the "sub procedures" succeeded.
> > 3. Whether correctable/uncorrectable error exists in a file system
> > is not related to the exist status of scrub. The role of this
> > program is
> >
> > a) reading all data from all devices and verify checksums, and
> > b) correct checksum errors "if possible".
> >
> > So, I consider scrub only returns error if this program itself
> > failed to finish.
>
> Sorry, I understood what you said. It's no problem to exit with
> error codes to detect whether correctable/uncorrectable errors happen
> or not. Sorry for noise.
>
> Thanks,
> Satoru
Oh, it is really kind of you to spend time reviewing. Thanks very much.
-Gui
> >
> > Thanks,
> > Satoru
> >
> >> return 0;
> >> }
> >>
> >>
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> >
>
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
