1 These are problems known to exist at the time of this release. Feel free to
2 join in and help us correct one or more of these! Also be sure to check the
3 changelog of the current development status, as one or more of these problems
4 may have been fixed since this was written!
6 64. The threaded resolver used in libcurl on Windows has some kind of race
7 problem when multiple simultanoes resolves are done, like with the multi
8 interface transferring many files in parallell:
9 http://curl.haxx.se/mail/lib-2009-04/0028.html
11 63. When CURLOPT_CONNECT_ONLY is used, the handle cannot reliably be re-used
12 for any further requests or transfers. The work-around is then to close that
13 handle with curl_easy_cleanup() and create a new. Some more details:
14 http://curl.haxx.se/mail/lib-2009-04/0300.html
16 62. CURLOPT_TIMEOUT does not work properly with the regular multi and
17 multi_socket interfaces. The work-around for apps is to simply remove the
18 easy handle once the time is up. See also:
19 http://curl.haxx.se/bug/view.cgi?id=2501457
21 61. If an upload using Expect: 100-continue receives an HTTP 417 response,
22 it ought to be automatically resent without the Expect:. A workaround is
23 for the client application to redo the transfer after disabling Expect:.
24 http://curl.haxx.se/mail/archive-2008-02/0043.html
26 60. libcurl closes the connection if an HTTP 401 reply is received while it
27 is waiting for the the 100-continue response.
28 http://curl.haxx.se/mail/lib-2008-08/0462.html
30 58. It seems sensible to be able to use CURLOPT_NOBODY and
31 CURLOPT_FAILONERROR with FTP to detect if a file exists or not, but it is
32 not working: http://curl.haxx.se/mail/lib-2008-07/0295.html
34 57. On VMS-Alpha: When using an http-file-upload the file is not sent to the
35 Server with the correct content-length. Sending a file with 511 or less
36 bytes, content-length 512 is used. Sending a file with 513 - 1023 bytes,
37 content-length 1024 is used. Files with a length of a multiple of 512 Bytes
38 show the correct content-length. Only these files work for upload.
39 http://curl.haxx.se/bug/view.cgi?id=2057858
41 56. When libcurl sends CURLOPT_POSTQUOTE commands when connected to a SFTP
42 server using the multi interface, the commands are not being sent correctly
43 and instead the connection is "cancelled" (the operation is considered done)
44 prematurely. There is a half-baked (busy-looping) patch provided in the bug
45 report but it cannot be accepted as-is. See
46 http://curl.haxx.se/bug/view.cgi?id=2006544
48 55. libcurl fails to build with MIT Kerberos for Windows (KfW) due to KfW's
49 library header files exporting symbols/macros that should be kept private
50 to the KfW library. See ticket #5601 at http://krbdev.mit.edu/rt/
52 52. Gautam Kachroo's issue that identifies a problem with the multi interface
53 where a connection can be re-used without actually being properly
55 http://curl.haxx.se/mail/lib-2008-01/0277.html
57 49. If using --retry and the transfer timeouts (possibly due to using -m or
58 -y/-Y) the next attempt doesn't resume the transfer properly from what was
59 downloaded in the previous attempt but will truncate and restart at the
60 original position where it was at before the previous failed attempt. See
61 http://curl.haxx.se/mail/lib-2008-01/0080.html and Mandriva bug report
62 https://qa.mandriva.com/show_bug.cgi?id=22565
64 48. If a CONNECT response-headers are larger than BUFSIZE (16KB) when the
65 connection is meant to be kept alive (like for NTLM proxy auth), the
66 function will return prematurely and will confuse the rest of the HTTP
67 protocol code. This should be very rare.
69 43. There seems to be a problem when connecting to the Microsoft telnet server.
70 http://curl.haxx.se/bug/view.cgi?id=1720605
72 41. When doing an operation over FTP that requires the ACCT command (but not
73 when logging in), the operation will fail since libcurl doesn't detect this
74 and thus fails to issue the correct command:
75 http://curl.haxx.se/bug/view.cgi?id=1693337
77 39. Steffen Rumler's Race Condition in Curl_proxyCONNECT:
78 http://curl.haxx.se/mail/lib-2007-01/0045.html
80 38. Kumar Swamy Bhatt's problem in ftp/ssl "LIST" operation:
81 http://curl.haxx.se/mail/lib-2007-01/0103.html
83 37. Having more than one connection to the same host when doing NTLM
84 authentication (with performs multiple "passes" and authenticates a
85 connection rather than a HTTP request), and particularly when using the
86 multi interface, there's a risk that libcurl will re-use a wrong connection
87 when doing the different passes in the NTLM negotiation and thus fail to
88 negotiate (in seemingly mysterious ways).
90 35. Both SOCKS5 and SOCKS4 proxy connections are done blocking, which is very
91 bad when used with the multi interface.
93 34. The SOCKS4 connection codes don't properly acknowledge (connect) timeouts.
94 Also see #12. According to bug #1556528, even the SOCKS5 connect code does
95 not do it right: http://curl.haxx.se/bug/view.cgi?id=1556528,
97 31. "curl-config --libs" will include details set in LDFLAGS when configure is
98 run that might be needed only for building libcurl. Further, curl-config
99 --cflags suffers from the same effects with CFLAGS/CPPFLAGS.
101 30. You need to use -g to the command line tool in order to use RFC2732-style
102 IPv6 numerical addresses in URLs.
104 29. IPv6 URLs with zone ID is not supported.
105 http://www.ietf.org/internet-drafts/draft-fenner-literal-zone-02.txt (expired)
106 specifies the use of a plus sign instead of a percent when specifying zone
107 IDs in URLs to get around the problem of percent signs being
108 special. According to the reporter, Firefox deals with the URL _with_ a
109 percent letter (which seems like a blatant URL spec violation).
111 See http://curl.haxx.se/bug/view.cgi?id=1371118
113 26. NTLM authentication using SSPI (on Windows) when (lib)curl is running in
114 "system context" will make it use wrong(?) user name - at least when compared
115 to what winhttp does. See http://curl.haxx.se/bug/view.cgi?id=1281867
117 23. SOCKS-related problems:
118 A) libcurl doesn't support SOCKS for IPv6.
119 B) libcurl doesn't support FTPS over a SOCKS proxy.
120 E) libcurl doesn't support active FTP over a SOCKS proxy
122 We probably have even more bugs and lack of features when a SOCKS proxy is
125 22. Sending files to a FTP server using curl on VMS, might lead to curl
126 complaining on "unaligned file size" on completion. The problem is related
127 to VMS file structures and the perceived file sizes stat() returns. A
128 possible fix would involve sending a "STRU VMS" command.
129 http://curl.haxx.se/bug/view.cgi?id=1156287
131 21. FTP ASCII transfers do not follow RFC959. They don't convert the data
132 accordingly (not for sending nor for receiving). RFC 959 section 3.1.1.1
133 clearly describes how this should be done:
135 The sender converts the data from an internal character representation to
136 the standard 8-bit NVT-ASCII representation (see the Telnet
137 specification). The receiver will convert the data from the standard
138 form to his own internal form.
140 Since 7.15.4 at least line endings are converted.
142 16. FTP URLs passed to curl may contain NUL (0x00) in the RFC 1738 <user>,
143 <password>, and <fpath> components, encoded as "%00". The problem is that
144 curl_unescape does not detect this, but instead returns a shortened C
145 string. From a strict FTP protocol standpoint, NUL is a valid character
146 within RFC 959 <string>, so the way to handle this correctly in curl would
147 be to use a data structure other than a plain C string, one that can handle
148 embedded NUL characters. From a practical standpoint, most FTP servers
149 would not meaningfully support NUL characters within RFC 959 <string>,
150 anyway (e.g., UNIX pathnames may not contain NUL).
152 14. Test case 165 might fail on system which has libidn present, but with an
153 old iconv version (2.1.3 is a known bad version), since it doesn't recognize
154 the charset when named ISO8859-1. Changing the name to ISO-8859-1 makes the
155 test pass, but instead makes it fail on Solaris hosts that use its native
158 13. curl version 7.12.2 fails on AIX if compiled with --enable-ares.
159 The workaround is to combine --enable-ares with --disable-shared
161 12. When connecting to a SOCKS proxy, the (connect) timeout is not properly
162 acknowledged after the actual TCP connect (during the SOCKS "negotiate"
165 10. To get HTTP Negotiate authentication to work fine, you need to provide a
166 (fake) user name (this concerns both curl and the lib) because the code
167 wrongly only considers authentication if there's a user name provided.
168 http://curl.haxx.se/bug/view.cgi?id=1004841. How?
169 http://curl.haxx.se/mail/lib-2004-08/0182.html
171 8. Doing resumed upload over HTTP does not work with '-C -', because curl
172 doesn't do a HEAD first to get the initial size. This needs to be done
173 manually for HTTP PUT resume to work, and then '-C [index]'.
175 6. libcurl ignores empty path parts in FTP URLs, whereas RFC1738 states that
176 such parts should be sent to the server as 'CWD ' (without an argument).
177 The only exception to this rule, is that we knowingly break this if the
178 empty part is first in the path, as then we use the double slashes to
179 indicate that the user wants to reach the root dir (this exception SHALL
180 remain even when this bug is fixed).
182 5. libcurl doesn't treat the content-length of compressed data properly, as
183 it seems HTTP servers send the *uncompressed* length in that header and
184 libcurl thinks of it as the *compressed* length. Some explanations are here:
185 http://curl.haxx.se/mail/lib-2003-06/0146.html
187 2. If a HTTP server responds to a HEAD request and includes a body (thus
188 violating the RFC2616), curl won't wait to read the response but just stop
189 reading and return back. If a second request (let's assume a GET) is then
190 immediately made to the same server again, the connection will be re-used
191 fine of course, and the second request will be sent off but when the
192 response is to get read, the previous response-body is what curl will read
193 and havoc is what happens.
194 More details on this is found in this libcurl mailing list thread:
195 http://curl.haxx.se/mail/lib-2002-08/0000.html