several FAIL calling: make check

[GoogleCodeExporter]

Answering the following questions is a big help:

1. What version of distcc are you using svn trunk (rev. 709)

2. What platform are you running on 

2.6.28-15-generic i686 GNU/Linux
cc (Ubuntu 4.3.3-5ubuntu4) 4.3.3
Python 2.6.2

3. What were you trying to do (e.g. "install distcc", "build Mozilla")?
make check

4. What went wrong? 
a lot of failure during tests, some coming from python script other coming
from the test result evaluation (see attached log)

Original issue reported on code.google.com by keller.eric on 4 Sep 2009 at 9:25

Attachments:

• make.check.res

[GoogleCodeExporter]

These errors all look like they have the same root cause.  distcc exits with 
exit
code 107, which is documented in the distcc man page as "I/O error".
I don't know what's causing it, and it's a little surprising that you didn't 
get any
output at all from distcc.  Your disk isn't full, is it?

There's another error in the logs which I think is a red herring.
The first invocation of distccd exits with exit code 102, which is documented 
in the
distcc man page as

       102    Bind failed.

There should be more detail in the distccd.log files mentioned in the output 
from
"make check".  Knowing the contents of one of those files would be helpful.
But this error usually happens because some other process is already using the 
port.
"make check" tries to use port 42000.  However, if that fails, it then goes on 
and
tries to use port 42001, 42002, etc.  The second invocation return exit status 
0, so
it looks like that invocation using port 42001 was successful.  Maybe you have
another instance of the distccd process hanging around; if so, please kill it.

BTW, I think there may be some tests in the distcc test suite which don't do 
this retry.
If you happen to be running another application which is already using that 
port on
your machine, you'll need to
either close or kill that application, or edit the line

    DISTCC_TEST_PORT             = 42000

in test/testdistcc.py.

Of course those are work-arounds.  To really fix this problem, 
test/testdistcc.py
should be modified to keep trying higher ports until it finds one which is free.
It already does this in most cases, but I think not in all cases.

If you ignore this failure of "make check", is distcc working correctly for you 
(with
DISTCC_HOSTS=127.0.0.1 or DISTCC_HOSTS=127.0.01.,cpp,lzo for pump mode)?

Original comment by fergus.h...@gmail.com on 4 Sep 2009 at 9:57

[GoogleCodeExporter]

So, my best guess at this point is that your disk was full.
If I don't hear any more from you, I'll close this issue.

Original comment by fergus.h...@gmail.com on 1 Dec 2009 at 4:39

[GoogleCodeExporter]

Original comment by fergus.h...@gmail.com on 2 Mar 2010 at 8:04

• Changed state: WontFix