[Drbd-dev] [BUG] drbdadm_parser.c:1968:63: error: 'glob_t' has no member named 'gl_flags'

[Drbd-dev] [BUG] drbdadm_parser.c:1968:63: error: 'glob_t' has no member named 'gl_flags'

[Milan P. Stanić]

Hi,

[ Please Cc to me, I'm not subscribed to this mailing list ]

I'm trying to build drbd-tools 9.22.0 on Alpine Linux which is based on
musl libc (and not glibc).

Build fail with this error:
-----------------------------
drbdadm_parser.c: In function 'include_stmt':
drbdadm_parser.c:1968:63: error: 'glob_t' has no member named 'gl_flags'
 1968 |                         } else if (errno == ENOENT && glob_buf.gl_flags & GLOB_MAGCHAR) {
      |                                                               ^
drbdadm_parser.c:1968:75: error: 'GLOB_MAGCHAR' undeclared (first use in this function)
 1968 |                         } else if (errno == ENOENT && glob_buf.gl_flags & GLOB_MAGCHAR) {
      |                                                                           ^~~~~~~~~~~~
drbdadm_parser.c:1968:75: note: each undeclared identifier is reported only once for each function it appears in
make[1]: *** [<builtin>: drbdadm_parser.o] Error 1
make[1]: *** Waiting for unfinished jobs....
make[1]: Leaving directory '/home/mps/aports/main/drbd-utils/src/drbd-utils-9.22.0/user/v9'
make: *** [Makefile:90: tools] Error 2
>>> ERROR: drbd-utils: build failed
-----------------------------

Musl libc glob.h doesn't have gl_flags nor GLOB_MAGCHAR because it is
non standard glibc extension.

I reverted https://github.com/LINBIT/drbd-utils/commit/4a1b590090bf676cdfb10c198505b95b9eeb3120
commit and then drbd-tools builds fine.

Not sure is this revert safe to push drbd-utils in next Alpine release
(which planned for next week) or there is better fix.

TIA

-- 
Kind regards

Re: [Drbd-dev] [BUG] drbdadm_parser.c:1968:63: error: 'glob_t' has no member named 'gl_flags'

[Roland Kammerer]

On Mon, Nov 07, 2022 at 01:08:14PM +0100, Milan P. Stanić wrote:
> Hi,
> 
> [ Please Cc to me, I'm not subscribed to this mailing list ]
> 
> I'm trying to build drbd-tools 9.22.0 on Alpine Linux which is based on
> musl libc (and not glibc).
> 
> [...]
> 
> Musl libc glob.h doesn't have gl_flags nor GLOB_MAGCHAR because it is
> non standard glibc extension.

thanks for reporting, this should be fixed in
https://github.com/LINBIT/drbd-utils/commit/33d5e8f7066116bd0a706c7cdda4950895164d34

> I reverted https://github.com/LINBIT/drbd-utils/commit/4a1b590090bf676cdfb10c198505b95b9eeb3120
> commit and then drbd-tools builds fine.
> 
> Not sure is this revert safe to push drbd-utils in next Alpine release
> (which planned for next week) or there is better fix.

I think reverting for now is perfectly fine, this deserved a fix, but I
don't think it actually happened in the wild that often... Or you pick
the proposed fix, which would be a good test for us as well. Currently
I'm not sure when the next official release will be done, current gut
feeling is not enough interesting things people would wait for has have
happened.

Regards, rck

Re: [Drbd-dev] [BUG] drbdadm_parser.c:1968:63: error: 'glob_t' has no member named 'gl_flags'

[Milan P. Stanić]

Hi,

[ Please Cc to me, I'm not subscribed to this mailing list ]

On Mon, 2022-11-14 at 09:52, Roland Kammerer wrote:
> On Mon, Nov 07, 2022 at 01:08:14PM +0100, Milan P. Stanić wrote:
> > Hi,
> > 
> > [ Please Cc to me, I'm not subscribed to this mailing list ]
> > 
> > I'm trying to build drbd-tools 9.22.0 on Alpine Linux which is based on
> > musl libc (and not glibc).
> > 
> > [...]
> > 
> > Musl libc glob.h doesn't have gl_flags nor GLOB_MAGCHAR because it is
> > non standard glibc extension.
> 
> thanks for reporting, this should be fixed in
> https://github.com/LINBIT/drbd-utils/commit/33d5e8f7066116bd0a706c7cdda4950895164d34
> 
> > I reverted https://github.com/LINBIT/drbd-utils/commit/4a1b590090bf676cdfb10c198505b95b9eeb3120
> > commit and then drbd-tools builds fine.
> > 
> > Not sure is this revert safe to push drbd-utils in next Alpine release
> > (which planned for next week) or there is better fix.
> 
> I think reverting for now is perfectly fine, this deserved a fix, but I
> don't think it actually happened in the wild that often... Or you pick
> the proposed fix, which would be a good test for us as well. Currently

I picked above commit and build with it passed fine.
Thank you for fix.

> I'm not sure when the next official release will be done, current gut
> feeling is not enough interesting things people would wait for has have
> happened.

-- 
Kind regards

> 
> Regards, rck