Re: [Makefile:1063: 828-ncurses] Error 1

On Sat, 2025-06-07 at 14:33 +0200, Marcin Dulak wrote:
> On Sat, Jun 7, 2025 at 4:30 AM Bruce Dubbs
> <[email protected]> wrote:
> > 
> > On 6/6/25 6:09 PM, Marcin Dulak ([email protected] via alfs-
> > discuss Mailing
> > List) wrote:
> > 
> > > The pipeline often fails unable to download various patch files.
> > > For example this one
> > > https://github.com/marcindulak/jhalfs-ci/actions/runs/15370571417/job/43249083644
> > > (you can't see the logs, so I paste them below).
> > > When you test jhalfs, do you use any alternative patch locations?
> > 
> > I never let jhalfs download the files. I have them locally.
> > 
> > > --2025-06-01 03:38:06--
> > > https://www.linuxfromscratch.org/patches/lfs/development/bash-5.3-rc1-cross_compile_fix-1.patch
> > > Resolving www.linuxfromscratch.org (www.linuxfromscratch.org)...
> > > 208.118.68.85
> > > Connecting to www.linuxfromscratch.org
> > > (www.linuxfromscratch.org)|208.118.68.85|:443... connected.
> > > HTTP request sent, awaiting response... 404 Not Found
> > > 2025-06-01 03:38:06 ERROR 404: Not Found.
> > 
> > That's because we are now on bash-5.3-rc2 and it does not need the
> > patch.
> > The patch repository is updated daily according to the most recent
> > commits.
> > 
> > I don't know why you are not getting the other patches:
> 
> the topic of patches is a side question.
> I don't experience any missing patches currently, the output was from
> 2025-06-01.
> At what time the patch repository is synchronized daily?
> My pipelines run at 3 a.m. UTC, and picks up missing patches often.

Looks like lfs book is generated slightly after 7am GMT. Needed patches
are copied to location at that time...

> 
> For the main topic, it looks like I reproduce the ncurses failure
> locally.
> Please try to reproduce on your side.
> 
> When printing more lines or logs/828-ncurses-6.5-20250531, I see
> errors like:
> grep "Error " logs/828-ncurses-6.5-20250531
> make[2]: [Makefile:1001:
> /sources/ncurses-6.5-20250531/dest/usr/lib/libncursesw.so.6.5] Error
> 1
> (ignored)
> make[2]: [Makefile:345:
> /sources/ncurses-6.5-20250531/dest/usr/lib/libpanelw.so.6.5] Error 1
> (ignored)
> make[2]: [Makefile:386:
> /sources/ncurses-6.5-20250531/dest/usr/lib/libmenuw.so.6.5] Error 1
> (ignored)
> make[2]: [Makefile:426:
> /sources/ncurses-6.5-20250531/dest/usr/lib/libformw.so.6.5] Error 1
> (ignored)
> make[2]: [Makefile:397:
> /sources/ncurses-6.5-20250531/dest/usr/lib/libncurses++w.so.6.5]
> Error
> 1 (ignored)
> 
> For example, like below. I don't know whether this is an actual
> error.
> Why there a Errors printed as ignored by the Makefile?
> 
> make[2]: Entering directory '/sources/ncurses-6.5-20250531/panel'
> linking /sources/ncurses-6.5-20250531/dest/usr/lib/libpanelw.so.6.5
> gcc  -O2 --param max-inline-insns-single=1200  -shared
> -Wl,-soname,`basename
> /sources/ncurses-6.5-20250531/dest/usr/lib/libpanelw.so.6.5
> .6.5`.6,-stats,-lc -o
> /sources/ncurses-6.5-20250531/dest/usr/lib/libpanelw.so.6.5
> ../obj_s/panel.o ../obj_s/p_above.o ../obj_s/p_below.o
> ../obj_s/p_bottom.o ../obj_s/p_delete.o ../obj_s/p_hide.o
> ../obj_s/p_hidden.o ../obj_s/p_move.o ../obj_s/p_new.o
> ../obj_s/p_replace.o ../obj_s/p_show.o ../obj_s/p_top.o
> ../obj_s/p_update.o ../obj_s/p_user.o ../obj_s/p_win.o -L../lib
> -lncursesw
> /usr/bin/ld: total time in link: 0.031484
> cd /sources/ncurses-6.5-20250531/dest/usr/lib && (ln -s -f
> libpanelw.so.6.5 libpanelw.so.6; ln -s -f libpanelw.so.6
> libpanelw.so;
> )
> test -z "/sources/ncurses-6.5-20250531/dest" && /usr/sbin/ldconfig
> make[2]: [Makefile:345:
> /sources/ncurses-6.5-20250531/dest/usr/lib/libpanelw.so.6.5] Error 1
> (ignored)
> 
> 

"ignored" errors are not errors... Sometimes, Makefile's need to
execute an action if some condition is met. This action may fail if the
condition is not met. Rather than using a complicated if ... else
construct, it is easier to perform the action and tell "make" to ignore
the error if there is any.

Does ncurses build? I mean, does the build stop before finishing? If
not, it means there is no error, I'd say.

Pierre

-- 
http://lists.linuxfromscratch.org/sympa/info/alfs-discuss
Unsubscribe: See the above information page