[PATCH] Fix attempts option not being respected from resolv.conf

[PATCH] Fix attempts option not being respected from resolv.conf

[Stefan Sedich]

[-- Attachment #1: Type: text/plain, Size: 1003 bytes --]

After some extensive digging it looks like the attempts option is not being
used properly in resolv.conf, turns out there is a subtle bug and it does
not appear to read the value correctly.

Looking at the old code this did work but must have broken when the
resolv.conf reading was refactored out, below is the fix for this (first
time submitting a patch here so hopefully I am doing it right).

---
 src/network/resolvconf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/network/resolvconf.c b/src/network/resolvconf.c
index 2cf1f475..4c3e4c4b 100644
--- a/src/network/resolvconf.c
+++ b/src/network/resolvconf.c
@@ -45,8 +45,8 @@ int __get_resolv_conf(struct resolvconf *conf, char
*search, size_t search_sz)
  if (z != p) conf->ndots = x > 15 ? 15 : x;
  }
  p = strstr(line, "attempts:");
- if (p && isdigit(p[6])) {
- p += 6;
+ if (p && isdigit(p[9])) {
+ p += 9;
  unsigned long x = strtoul(p, &z, 10);
  if (z != p) conf->attempts = x > 10 ? 10 : x;
  }
-- 
2.11.0

[-- Attachment #2: Type: text/html, Size: 2037 bytes --]

Re: [PATCH] Fix attempts option not being respected from resolv.conf

[Rich Felker]

On Mon, Apr 03, 2017 at 11:15:12PM +0000, Stefan Sedich wrote:
> After some extensive digging it looks like the attempts option is not being
> used properly in resolv.conf, turns out there is a subtle bug and it does
> not appear to read the value correctly.
> 
> Looking at the old code this did work but must have broken when the
> resolv.conf reading was refactored out, below is the fix for this (first
> time submitting a patch here so hopefully I am doing it right).
> 
> ---
>  src/network/resolvconf.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/src/network/resolvconf.c b/src/network/resolvconf.c
> index 2cf1f475..4c3e4c4b 100644
> --- a/src/network/resolvconf.c
> +++ b/src/network/resolvconf.c
> @@ -45,8 +45,8 @@ int __get_resolv_conf(struct resolvconf *conf, char
> *search, size_t search_sz)
>   if (z != p) conf->ndots = x > 15 ? 15 : x;
>   }
>   p = strstr(line, "attempts:");
> - if (p && isdigit(p[6])) {
> - p += 6;
> + if (p && isdigit(p[9])) {
> + p += 9;
>   unsigned long x = strtoul(p, &z, 10);
>   if (z != p) conf->attempts = x > 10 ? 10 : x;
>   }

Looks fine, but your mailer botched the patch formatting. I've applied
it by hand and added a reference to the commit responsible for the
regression. Thanks.

Rich

Re: [PATCH] Fix attempts option not being respected from resolv.conf

[Stefan Sedich]

[-- Attachment #1: Type: text/plain, Size: 1552 bytes --]

Thanks Rich,

I have a V2 where I fixed this up as I made a mess of the first one as you
can see, thanks for getting it in :)



Stefan

On Fri, Apr 21, 2017 at 2:37 PM Rich Felker <dalias@libc.org> wrote:

> On Mon, Apr 03, 2017 at 11:15:12PM +0000, Stefan Sedich wrote:
> > After some extensive digging it looks like the attempts option is not
> being
> > used properly in resolv.conf, turns out there is a subtle bug and it does
> > not appear to read the value correctly.
> >
> > Looking at the old code this did work but must have broken when the
> > resolv.conf reading was refactored out, below is the fix for this (first
> > time submitting a patch here so hopefully I am doing it right).
> >
> > ---
> >  src/network/resolvconf.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/src/network/resolvconf.c b/src/network/resolvconf.c
> > index 2cf1f475..4c3e4c4b 100644
> > --- a/src/network/resolvconf.c
> > +++ b/src/network/resolvconf.c
> > @@ -45,8 +45,8 @@ int __get_resolv_conf(struct resolvconf *conf, char
> > *search, size_t search_sz)
> >   if (z != p) conf->ndots = x > 15 ? 15 : x;
> >   }
> >   p = strstr(line, "attempts:");
> > - if (p && isdigit(p[6])) {
> > - p += 6;
> > + if (p && isdigit(p[9])) {
> > + p += 9;
> >   unsigned long x = strtoul(p, &z, 10);
> >   if (z != p) conf->attempts = x > 10 ? 10 : x;
> >   }
>
> Looks fine, but your mailer botched the patch formatting. I've applied
> it by hand and added a reference to the commit responsible for the
> regression. Thanks.
>
> Rich
>

[-- Attachment #2: Type: text/html, Size: 2112 bytes --]