Re: Segfault downloading two files with -O -J
On Sat, 20 Aug 2011 00:30:43 +0200 (CEST) Daniel Stenberg dan...@haxx.se wrote: On Thu, 18 Aug 2011, Paul Howarth wrote: https://github.com/bagder/curl/commit/5eb2396cd15cbbf73b02 It doesn't for me: Hm, ok then perhaps something else fixed it since with my git version I cannot reproduce. I also ran it successfully with valgrind to verify a little extra. Can you try the git version and check in your end? I've tried with the 20110821 daily snapshot and get slightly different output but still a core dump: $ curl -O -J 'http://koji.fedoraproject.org/koji/getfile?taskID=3281612name=ibus-1.3.99.20110419-15.fc15.x86_64.rpm' 'http://koji.fedoraproject.org/koji/getfile?taskID=3281612name=ibus-libs-1.3.99.20110419-15.fc15.x86_64.rpm' /dev/null % Total% Received % Xferd Average Speed TimeTime Time Current Dload Upload Total SpentLeft Speed 0 00 00 0 0 0 --:--:-- --:--:-- --:--:-- 0Warning: Refusing to overwrite ibus-1.3.99.20110419-15.fc15.x86_64.rpm: File Warning: exists 0 388k0 10400 0 2369 0 0:02:48 --:--:-- 0:02:48 2482 curl: (23) Failed writing body (0 != 1040) Segmentation fault (core dumped) Paul. --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
Re: Segfault downloading two files with -O -J
On Sun, 21 Aug 2011 10:38:25 +0100 Paul Howarth p...@city-fan.org wrote: On Sat, 20 Aug 2011 00:30:43 +0200 (CEST) Daniel Stenberg dan...@haxx.se wrote: On Thu, 18 Aug 2011, Paul Howarth wrote: https://github.com/bagder/curl/commit/5eb2396cd15cbbf73b02 It doesn't for me: Hm, ok then perhaps something else fixed it since with my git version I cannot reproduce. I also ran it successfully with valgrind to verify a little extra. Can you try the git version and check in your end? I've tried with the 20110821 daily snapshot and get slightly different output but still a core dump: $ curl -O -J 'http://koji.fedoraproject.org/koji/getfile?taskID=3281612name=ibus-1.3.99.20110419-15.fc15.x86_64.rpm' 'http://koji.fedoraproject.org/koji/getfile?taskID=3281612name=ibus-libs-1.3.99.20110419-15.fc15.x86_64.rpm' /dev/null % Total% Received % Xferd Average Speed Time Time Time Current Dload Upload Total SpentLeft Speed 0 00 00 0 0 0 --:--:-- --:--:-- --:--:-- 0Warning: Refusing to overwrite ibus-1.3.99.20110419-15.fc15.x86_64.rpm: File Warning: exists 0 388k0 10400 0 2369 0 0:02:48 --:--:-- 0:02:48 2482 curl: (23) Failed writing body (0 != 1040) Segmentation fault (core dumped) I got this backtrace from gdb: #0 0x004067ce in header_callback (ptr=0x62a150, size=optimized out, nmemb=optimized out, stream=0x0) at main.c:4610 #1 0x77d9c04b in Curl_client_write (conn=0x62be30, type=optimized out, ptr=0x62a150 Content-Disposition: attachment; filename=ibus-libs-1.3.99.20110419-15.fc15.x86_64.rpm\r\n, len=88) at sendf.c:483 #2 0x77d9ab9e in Curl_http_readwrite_headers (data=0x621560, conn=0x62be30, nread=0x7fffd5a0, stop_reading=0x7fffd5af) at http.c:3397 #3 0x77dae33a in readwrite_data (done=0x7fffd63f, didwhat=read_sleb128: Corrupted DWARF expression. ) at transfer.c:464 #4 Curl_readwrite (conn=0x62be30, done=0x7fffd63f) at transfer.c:1025 #5 0x77dafb84 in Transfer (conn=0x62be30) at transfer.c:1392 #6 Curl_do_perform (data=0x621560) at transfer.c:2133 #7 0x00404369 in operate (argv=optimized out, argc=0, config=0x7fffd720) at main.c:5643 #8 main (argc=0, argv=optimized out) at main.c:5968 Paul. Paul. --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
Re: Compiling libcurl staticly within DLL-module with mingw32
I have tried to redefine CURL_EXTERN myself, in curl/curl.h after where it was defined, both to __declspec(dllimport) and __declspec(dllexport) with no luck - libcurl replaces all dll symbols by it's own when linked staticly. --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
Re: Segfault downloading two files with -O -J
On Sun, 21 Aug 2011, Paul Howarth wrote: I got this backtrace from gdb: Thanks. I've pushed commit 61ae7e9ce77a now which I believe now makes the fix more complete. Sorry about that. -- / daniel.haxx.se --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
Re: Segfault downloading two files with -O -J
On Sun, 21 Aug 2011 13:02:29 +0200 (CEST) Daniel Stenberg dan...@haxx.se wrote: On Sun, 21 Aug 2011, Paul Howarth wrote: I got this backtrace from gdb: Thanks. I've pushed commit 61ae7e9ce77a now which I believe now makes the fix more complete. Sorry about that. Thanks, that seems to have nailed it here. Paul. --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
RE: [PATCH] Curl_gethostname() may or may not return the hostname as a FQDN
On Thu, 11 Aug 2011, Steve Holme wrote: I have had an attempt at trying to use Curl_resolv() and Curl_resolv_timeout() but unfortunately the CANONICAL name is always empty. It would appear that the underlying Curl_resolver_getaddrinfo() does not set the ai_flags to include: AI_CANONNAME :( (sorry for the major delay) Argh, that's correct. We don't have anything in libcurl before this that needs that feature. I wonder how large percentage of use cases this will work for anyway. Lots of machines today that speak SMTP will not be able to resolve their own names. Is it really worth persuing or should we work on better documenting how a client should do to provide the full host name itself? -- / daniel.haxx.se --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
RE: [PATCH] Added AUTH NTLM for SMTP
Hi Yang, There's no real need for you to generate an additional patch, I can manage from this point without it. But of course you can provide it if you wish so ;-) Many thanks - it is much appreciated. I spent just over 2 days (about 20 hours) writing the original modification to Libcurl and since then nearly 35 hours on producing the various patches. At what stage are we now? I see there hasn't been any feedback about the code moving to curl_ntlm.[ch] so can we start to introduce the smtp code changes to introduce the NTLM feature? Kind Regards Steve --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
RE: [PATCH] Curl_gethostname() may or may not return the hostname as a FQDN
Hiya Daniel, (sorry for the major delay) No problem - I assumed you were busy with more important Curl matters / have a personal life ;-) Argh, that's correct. We don't have anything in libcurl before this that needs that feature. I wonder how large percentage of use cases this will work for anyway. Lots of machines today that speak SMTP will not be able to resolve their own names. I must admit I honestly can't answer that. My own testing of Libcurl has been performed against Exchange Server 2010 and that doesn't appear to have any problems. However, I have found various articles around the web that talk about this problem with the Windows SMTP service. Is it really worth persuing or should we work on better documenting how a client should do to provide the full host name itself? Good question. How easy / difficult would it be to add the AI_CANONNAME flag to Curl_resolv()'s underlying functions and what impact would this have on their performance? Over the last week I have had some thoughts about the Curl_gethostname() function and I must admit I really don't like changing the function's argument list so dramatically (as I did in my example code). If we are to do this I would much rather we change the behaviour of Curl_gethostname() to return just the hostname, in a non-fully qualified manner, thus stripping off the domain if necessary and adding a separate Curl_getfullyqualifiedhostname() or Curl_getfqhostname() function that calls Curl_resolv() as required. The benefit of doing this is that the behaviour of Curl_gethostname() is then defined as returning just the hostname and is constant across all platforms rather than at present where it may return the hostname, it may return the fully qualified hostname. The programmer who is calling the function has no idea what the underlying stack is going to return unless they perform a test on the string (as is the case of the existing NTLM code). If we decide not to modify Curl_resolv()'s underlying functions then I would still recommend we modify the output of Curl_gethostname() so it returns a defined value, but we will also need to think of a better way of passing the fully qualified machine name in the EHLO and HELO commands when talking to an SMTP server instead of using Curl_gethostname(). In my case, this would just move the problem from Libcurl to where I am using it as I would still need to perform an getaddrinfo() to get the fully qualified machine name ;-) Best Regards Steve --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
Re: Compiling libcurl staticly within DLL-module with mingw32
After compiling libcurl with -DCURL_STATICLIB (CPPFLAGS=-DCURL_STATICLIB ./configure --prefix=/home/crosscompile/i686-mingw32/tree/ ... etc) all worked! Big thanks to B4gder from #c...@freenode.irc for this hint! --- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
