cURL / Mailing Lists / curl-users / Single Mail

curl-users

Re: curl-7.10.4-pre4

From: Ralph Mitchell <rmitchell_at_eds.com>
Date: Wed, 19 Mar 2003 00:00:56 -0600

Daniel Stenberg wrote:

> On Tue, 18 Mar 2003, Ralph Mitchell wrote:
>
> > Test 100 runs a whole lot happier with a sleep(5) right after starting the
> > ftpserver...
>
> Ok, then it might work nicer with my recent fixes that first verifies that
> the FTP server is responding nicely before it runs any tests on it.

OK, good.

>
> > Test 401 fails with this in log/stderr401 (even with sleep(10) after starting
> > ftpserver and ftpsserver):
> >
> > * About to connect() to 127.0.0.1:8821
> > * Connected to 127.0.0.1 port 8821
> > * SSL connection timeout
> > * Closing connection #0
> > curl: (28) SSL connection timeout
>
> Does it fail with this? The test case is actually supposed to do exactly
> this: "FTPS download with strict timeout and slow CWD". It is a test that the
> timeout function works when working with a FTPS URL!

Well, it *says* it failed - I figured it ought to know...

    $ ./runtests.pl 401
    ********* System characteristics ********
    * curl 7.10.4-pre4 (sparc-unknown-netbsdelf1.6)
    * libcurl/7.10.4-pre4 OpenSSL/0.9.6g ipv6 zlib/1.1.4
    * Host: baradur
    * System: NetBSD baradur 1.6 NetBSD 1.6 (Baradur) #0: Thu Mar 13 08:00:16 CST
2003 root_at_baradur:/usr/src/sys/arch/sparc/compile/Baradur sparc
    * Memory debugging: OFF
    * Netrc debugging: OFF
    * HTTPS server: ON
    * FTPS server: ON
    * libcurl SSL: ON
    *****************************************
    test 401...[FTPS download with strict timeout and slow CWD]
     error OK0a1,4
> USER anonymous
> PASS curl_by_daniel_at_haxx.se
> PWD
> CWD path/to/file
     protocol FAILED

     - abort tests
    TESTDONE: 0 tests out of 1 reported OK: 0%
    TESTFAIL: These test cases failed: 401

> > and test 503 fails with this in log/stderr503:
> >
> > URL: http://127.0.0.1:8433/503
>
> This looks like either another problem with the test suite or a possible
> libcurl problem.
>
> If you got some more time to help out, then try filling the
> tests/libtest/lib503.c source file with fprintf(stderr, "reach point 432\n");
> style outputs so that we can figure out where/why it returns/exits from that
> function.

I'll try beating on this one a bit tonight.

Ralph Mitchell

-------------------------------------------------------
This SF.net email is sponsored by: Does your code think in ink?
You could win a Tablet PC. Get a free Tablet PC hat just for playing.
What are you waiting for?
http://ads.sourceforge.net/cgi-bin/redirect.pl?micr5043en
Received on 2003-03-19