Fernando de Oliveira wrote:
Em 10-07-2015 20:20, Fernando de Oliveira escreveu:
Em 10-07-2015 19:36, Bruce Dubbs escreveu:
I'm trying to test openldap-2.4.41 for including in the book. The
builds for both client and sever are fine, but the tests fail for me in
the server at an early point.
I went back to the book's version 2.4.40 and tried that. The tests fail
in exactly the same way:
>>>>> Starting test003-search for mdb...
...
Starting slapd on TCP/IP port 9011...
Testing slapd searching...
Waiting 5 seconds for slapd to start...
Waiting 5 seconds for slapd to start...
Waiting 5 seconds for slapd to start...
Waiting 5 seconds for slapd to start...
Waiting 5 seconds for slapd to start...
Waiting 5 seconds for slapd to start...
ldapsearch failed (255)!
>>>>> test003-search failed for mdb
(exit 255)
Makefile:312: recipe for target 'mdb-yes' failed
Can I get someone to try openldap-2.4.40 as in the book and see if the
tests work for you.
Thanks.
I'm running for 2.41, couldn't open VM for 2.40 logs, and had already
installed openldap-2.41 at 07/03/15 19:12, without tests.
Results are:
and no change for a long time.
Will leave it there and see tomorrow morning if anything changed.
At this time tests results are:
$ grep -E 'completed|^>>>>> Test|false' \
openldap-2.4.41-make-k-test-2015.07.10-20h02m10s.log
OK, I tried the same script on another system and got a lot further, but
it failed as
...
Query 15 not answerable
Query 16 answerable
Query 17 answerable
Successfully verified answerability
Filtering ldapsearch results...
Filtering original ldif...
Comparing filter output...
Comparison failed
>>>>> test020-proxycache failed for mdb
(exit 1)
I used *exactly* the same script as on the other system. Both systems
are x86_64. Dependencies may be different versions.
Hmm. I put some echo statements in the test020-proxycache script to see
what was failing and it passed. That suggests to me that the test
procedure has some race conditions that can make it fail. That may be
the problem on my other (even faster) system.
I'm sure you also noticed that the tests have a lot of waiting messages.
That makes the test time independent of processor speed, but if
upstream missed some, I suspect it can cause things to hang.
I'm now to test060 and I'm watching top. I don't think it's hung. It
appears to just run very slowly. the load average is around 1.0 but cpu
average is negligible, so again there are timing issues.
And the test finally finished:
0 tests for mdb were skipped.
On this particular system it took about 50 minutes for 'make test', but
as I said above, a lot of that time seems to be processor independent.
I'll write that up in the test description.
-- Bruce
--
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
