Bug 59117 - When source package doesn't builds, makefile returns 0 anyway
When source package doesn't builds, makefile returns 0 anyway
Product: Red Hat Linux
Classification: Retired
Component: tripwire (Show other bugs)
i386 Linux
medium Severity medium
: ---
: ---
Assigned To: Mike A. Harris
David Lawrence
Depends On:
  Show dependency treegraph
Reported: 2002-01-31 05:49 EST by Sysoltsev Slawa
Modified: 2007-04-18 12:39 EDT (History)
0 users

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2002-01-31 12:06:33 EST
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)
rpm -bc --rcfile ../rpmrc tripwire.spec &> tripwire.log (4.57 KB, text/plain)
2002-01-31 12:02 EST, Sysoltsev Slawa
no flags Details

  None (edit)
Description Sysoltsev Slawa 2002-01-31 05:49:18 EST
From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)

Description of problem:
If we do some syntax error in sources and then package build will be fail, 
makefile return 0 anyway and rpm respectively returns 0 too.

Version-Release number of selected component (if applicable):

How reproducible:

Steps to Reproduce:
1. cd BUILD/tripwire-2.3.1-2/src/core
2. edit charutil_t.cpp: insert "#error ERROR" into first line
3. cd ..
4. make release
5. echo $?


Actual Results:  compiler reports an error, make stops but returns 0

Expected Results:  compiler reports an error, make stops and returns nonzero

Additional info:

It occurs because build output going to 'tee' utility and exit status is 
returned from 'tee' - always 0. Status returned from compiler is lost 
and 'make' returns 0 too.
Comment 1 Mike A. Harris 2002-01-31 11:23:55 EST
It isn't clear to me exactly what it is you're reporting.  If you could
expand the problem description with a bit more detail, it would be

Comment 2 Sysoltsev Slawa 2002-01-31 11:37:46 EST
Normally, when we execute `rpm -bc` we expect that returned error code will 
show building status - package builds or no on my system. 'tripwire' returns 
zero always, independently on was there errors thru building process or not. 
It's clear that we may look for executable results, don't find them and 
recognize that build wasn't complete or just spot an error messages in log 
files but usually build status seen by rpm's exit status.
As I mentioned before rpm returns zero, because make returns zero, and make 
return zero because it is exit status from tee utility and/but compiler exit 
status just lost.
Comment 3 Mike A. Harris 2002-01-31 11:51:49 EST
Can you please attach a build log of this so I can see what you're seeing

rpm -ba tripwire.spec &> tripwire.log

Please attach tripwire.log using the file attachment link below.
Comment 4 Sysoltsev Slawa 2002-01-31 12:02:48 EST
Created attachment 44156 [details]
rpm -bc --rcfile ../rpmrc tripwire.spec &> tripwire.log
Comment 5 Sysoltsev Slawa 2002-01-31 12:06:27 EST
I don't have sufficient permissions and works with rpm locally, so you see that 
I use custom rc file.
I don't need to install package - I just need to build it by Intel C/C++ 
compiler and I see '#exit 0' while package may not be built.
Comment 6 Mike A. Harris 2002-01-31 15:20:31 EST
Red Hat does not provide fix-bugs level support people recompiling
the packages with other compilers.  In particular, this seems like
a technical support issue rather than a bug per se.

I've noticed a lot of Intel people are filing bug reports against
Red Hat Linux packages lately, when using the Intel compiler
to compile things.

In the interest of saving everyone at both Intel and Red Hat a lot
of wasted time and effort, it would be appreciated if a memo or
something could be circulated around there perhaps stating that
people please not file bugs in bugzilla when using tools that we
have not supplied in the Red Hat Linux distribution, since we do not
support such.

We can't possibly spend all of our engineering resources fixing problems
that crop up with various other compilers which we do not even ship
let alone support.  That said, if such support is indeed desired, I'm
sure that Red Hat consulting services can offer some sort of technical
support contract or possibly custom services.

Note You need to log in before you can comment on or make changes to this bug.