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 76. The SOCKET type in Win64 is 64 bits large (and thus so is curl_socket_t on
7 that platform), and long is only 32 bits. It makes it impossible for
8 curl_easy_getinfo() to return a socket properly with the CURLINFO_LASTSOCKET
9 option as for all other operating systems.
11 75. NTLM authentication involving unicode user name or password.
12 http://curl.haxx.se/mail/lib-2009-10/0024.html
13 http://curl.haxx.se/bug/view.cgi?id=2944325
15 74. The HTTP spec allows headers to be merged and become comma-separated
16 instead of being repeated several times. This also include Authenticate: and
17 Proxy-Authenticate: headers and while this hardly every happens in real life
18 it will confuse libcurl which does not properly support it for all headers -
19 like those Authenticate headers.
21 73. if a connection is made to a FTP server but the server then just never
22 sends the 220 response or otherwise is dead slow, libcurl will not
23 acknowledge the connection timeout during that phase but only the "real"
24 timeout - which may surprise users as it is probably considered to be the
25 connect phase to most people. Brought up (and is being misunderstood) in:
26 http://curl.haxx.se/bug/view.cgi?id=2844077
28 72. "Pausing pipeline problems."
29 http://curl.haxx.se/mail/lib-2009-07/0214.html
31 70. Problem re-using easy handle after call to curl_multi_remove_handle
32 http://curl.haxx.se/mail/lib-2009-07/0249.html
34 68. "More questions about ares behavior".
35 http://curl.haxx.se/mail/lib-2009-08/0012.html
37 67. When creating multipart formposts. The file name part can be encoded with
38 something beyond ascii but currently libcurl will only pass in the verbatim
39 string the app provides. There are several browsers that already do this
40 encoding. The key seems to be the updated draft to RFC2231:
41 http://tools.ietf.org/html/draft-reschke-rfc2231-in-http-02
43 66. When using telnet, the time limitation options don't work.
44 http://curl.haxx.se/bug/view.cgi?id=2818950
46 65. When doing FTP over a socks proxy or CONNECT through HTTP proxy and the
47 multi interface is used, libcurl will fail if the (passive) TCP connection
48 for the data transfer isn't more or less instant as the code does not
49 properly wait for the connect to be confirmed. See test case 564 for a first
52 63. When CURLOPT_CONNECT_ONLY is used, the handle cannot reliably be re-used
53 for any further requests or transfers. The work-around is then to close that
54 handle with curl_easy_cleanup() and create a new. Some more details:
55 http://curl.haxx.se/mail/lib-2009-04/0300.html
57 61. If an upload using Expect: 100-continue receives an HTTP 417 response,
58 it ought to be automatically resent without the Expect:. A workaround is
59 for the client application to redo the transfer after disabling Expect:.
60 http://curl.haxx.se/mail/archive-2008-02/0043.html
62 60. libcurl closes the connection if an HTTP 401 reply is received while it
63 is waiting for the the 100-continue response.
64 http://curl.haxx.se/mail/lib-2008-08/0462.html
66 58. It seems sensible to be able to use CURLOPT_NOBODY and
67 CURLOPT_FAILONERROR with FTP to detect if a file exists or not, but it is
68 not working: http://curl.haxx.se/mail/lib-2008-07/0295.html
70 57. On VMS-Alpha: When using an http-file-upload the file is not sent to the
71 Server with the correct content-length. Sending a file with 511 or less
72 bytes, content-length 512 is used. Sending a file with 513 - 1023 bytes,
73 content-length 1024 is used. Files with a length of a multiple of 512 Bytes
74 show the correct content-length. Only these files work for upload.
75 http://curl.haxx.se/bug/view.cgi?id=2057858
77 56. When libcurl sends CURLOPT_POSTQUOTE commands when connected to a SFTP
78 server using the multi interface, the commands are not being sent correctly
79 and instead the connection is "cancelled" (the operation is considered done)
80 prematurely. There is a half-baked (busy-looping) patch provided in the bug
81 report but it cannot be accepted as-is. See
82 http://curl.haxx.se/bug/view.cgi?id=2006544
84 55. libcurl fails to build with MIT Kerberos for Windows (KfW) due to KfW's
85 library header files exporting symbols/macros that should be kept private
86 to the KfW library. See ticket #5601 at http://krbdev.mit.edu/rt/
88 52. Gautam Kachroo's issue that identifies a problem with the multi interface
89 where a connection can be re-used without actually being properly
91 http://curl.haxx.se/mail/lib-2008-01/0277.html
93 49. If using --retry and the transfer timeouts (possibly due to using -m or
94 -y/-Y) the next attempt doesn't resume the transfer properly from what was
95 downloaded in the previous attempt but will truncate and restart at the
96 original position where it was at before the previous failed attempt. See
97 http://curl.haxx.se/mail/lib-2008-01/0080.html and Mandriva bug report
98 https://qa.mandriva.com/show_bug.cgi?id=22565
100 48. If a CONNECT response-headers are larger than BUFSIZE (16KB) when the
101 connection is meant to be kept alive (like for NTLM proxy auth), the
102 function will return prematurely and will confuse the rest of the HTTP
103 protocol code. This should be very rare.
105 43. There seems to be a problem when connecting to the Microsoft telnet server.
106 http://curl.haxx.se/bug/view.cgi?id=1720605
108 41. When doing an operation over FTP that requires the ACCT command (but not
109 when logging in), the operation will fail since libcurl doesn't detect this
110 and thus fails to issue the correct command:
111 http://curl.haxx.se/bug/view.cgi?id=1693337
113 39. Steffen Rumler's Race Condition in Curl_proxyCONNECT:
114 http://curl.haxx.se/mail/lib-2007-01/0045.html
116 38. Kumar Swamy Bhatt's problem in ftp/ssl "LIST" operation:
117 http://curl.haxx.se/mail/lib-2007-01/0103.html
119 37. Having more than one connection to the same host when doing NTLM
120 authentication (with performs multiple "passes" and authenticates a
121 connection rather than a HTTP request), and particularly when using the
122 multi interface, there's a risk that libcurl will re-use a wrong connection
123 when doing the different passes in the NTLM negotiation and thus fail to
124 negotiate (in seemingly mysterious ways).
126 35. Both SOCKS5 and SOCKS4 proxy connections are done blocking, which is very
127 bad when used with the multi interface.
129 34. The SOCKS4 connection codes don't properly acknowledge (connect) timeouts.
130 Also see #12. According to bug #1556528, even the SOCKS5 connect code does
131 not do it right: http://curl.haxx.se/bug/view.cgi?id=1556528,
133 31. "curl-config --libs" will include details set in LDFLAGS when configure is
134 run that might be needed only for building libcurl. Further, curl-config
135 --cflags suffers from the same effects with CFLAGS/CPPFLAGS.
137 30. You need to use -g to the command line tool in order to use RFC2732-style
138 IPv6 numerical addresses in URLs.
140 29. IPv6 URLs with zone ID is not nicely supported.
141 http://www.ietf.org/internet-drafts/draft-fenner-literal-zone-02.txt (expired)
142 specifies the use of a plus sign instead of a percent when specifying zone
143 IDs in URLs to get around the problem of percent signs being
144 special. According to the reporter, Firefox deals with the URL _with_ a
145 percent letter (which seems like a blatant URL spec violation).
146 libcurl supports zone IDs where the percent sign is URL-escaped (i.e. %25).
148 See http://curl.haxx.se/bug/view.cgi?id=1371118
150 26. NTLM authentication using SSPI (on Windows) when (lib)curl is running in
151 "system context" will make it use wrong(?) user name - at least when compared
152 to what winhttp does. See http://curl.haxx.se/bug/view.cgi?id=1281867
154 23. SOCKS-related problems:
155 A) libcurl doesn't support SOCKS for IPv6.
156 B) libcurl doesn't support FTPS over a SOCKS proxy.
157 E) libcurl doesn't support active FTP over a SOCKS proxy
159 We probably have even more bugs and lack of features when a SOCKS proxy is
162 22. Sending files to a FTP server using curl on VMS, might lead to curl
163 complaining on "unaligned file size" on completion. The problem is related
164 to VMS file structures and the perceived file sizes stat() returns. A
165 possible fix would involve sending a "STRU VMS" command.
166 http://curl.haxx.se/bug/view.cgi?id=1156287
168 21. FTP ASCII transfers do not follow RFC959. They don't convert the data
169 accordingly (not for sending nor for receiving). RFC 959 section 3.1.1.1
170 clearly describes how this should be done:
172 The sender converts the data from an internal character representation to
173 the standard 8-bit NVT-ASCII representation (see the Telnet
174 specification). The receiver will convert the data from the standard
175 form to his own internal form.
177 Since 7.15.4 at least line endings are converted.
179 16. FTP URLs passed to curl may contain NUL (0x00) in the RFC 1738 <user>,
180 <password>, and <fpath> components, encoded as "%00". The problem is that
181 curl_unescape does not detect this, but instead returns a shortened C
182 string. From a strict FTP protocol standpoint, NUL is a valid character
183 within RFC 959 <string>, so the way to handle this correctly in curl would
184 be to use a data structure other than a plain C string, one that can handle
185 embedded NUL characters. From a practical standpoint, most FTP servers
186 would not meaningfully support NUL characters within RFC 959 <string>,
187 anyway (e.g., UNIX pathnames may not contain NUL).
189 14. Test case 165 might fail on a system which has libidn present, but with an
190 old iconv version (2.1.3 is a known bad version), since it doesn't recognize
191 the charset when named ISO8859-1. Changing the name to ISO-8859-1 makes the
192 test pass, but instead makes it fail on Solaris hosts that use its native
195 13. curl version 7.12.2 fails on AIX if compiled with --enable-ares.
196 The workaround is to combine --enable-ares with --disable-shared
198 12. When connecting to a SOCKS proxy, the (connect) timeout is not properly
199 acknowledged after the actual TCP connect (during the SOCKS "negotiate"
202 10. To get HTTP Negotiate authentication to work fine, you need to provide a
203 (fake) user name (this concerns both curl and the lib) because the code
204 wrongly only considers authentication if there's a user name provided.
205 http://curl.haxx.se/bug/view.cgi?id=1004841. How?
206 http://curl.haxx.se/mail/lib-2004-08/0182.html
208 8. Doing resumed upload over HTTP does not work with '-C -', because curl
209 doesn't do a HEAD first to get the initial size. This needs to be done
210 manually for HTTP PUT resume to work, and then '-C [index]'.
212 6. libcurl ignores empty path parts in FTP URLs, whereas RFC1738 states that
213 such parts should be sent to the server as 'CWD ' (without an argument).
214 The only exception to this rule, is that we knowingly break this if the
215 empty part is first in the path, as then we use the double slashes to
216 indicate that the user wants to reach the root dir (this exception SHALL
217 remain even when this bug is fixed).
219 5. libcurl doesn't treat the content-length of compressed data properly, as
220 it seems HTTP servers send the *uncompressed* length in that header and
221 libcurl thinks of it as the *compressed* length. Some explanations are here:
222 http://curl.haxx.se/mail/lib-2003-06/0146.html
224 2. If a HTTP server responds to a HEAD request and includes a body (thus
225 violating the RFC2616), curl won't wait to read the response but just stop
226 reading and return back. If a second request (let's assume a GET) is then
227 immediately made to the same server again, the connection will be re-used
228 fine of course, and the second request will be sent off but when the
229 response is to get read, the previous response-body is what curl will read
230 and havoc is what happens.
231 More details on this is found in this libcurl mailing list thread:
232 http://curl.haxx.se/mail/lib-2002-08/0000.html