| 1 |
.\" -*- mode: troff; coding: utf-8 -*- |
| 2 |
.\" Automatically generated by Pod::Man 5.0102 (Pod::Simple 3.45) |
| 3 |
.\" |
| 4 |
.\" Standard preamble: |
| 5 |
.\" ======================================================================== |
| 6 |
.de Sp \" Vertical space (when we can't use .PP) |
| 7 |
.if t .sp .5v |
| 8 |
.if n .sp |
| 9 |
.. |
| 10 |
.de Vb \" Begin verbatim text |
| 11 |
.ft CW |
| 12 |
.nf |
| 13 |
.ne \\$1 |
| 14 |
.. |
| 15 |
.de Ve \" End verbatim text |
| 16 |
.ft R |
| 17 |
.fi |
| 18 |
.. |
| 19 |
.\" \*(C` and \*(C' are quotes in nroff, nothing in troff, for use with C<>. |
| 20 |
.ie n \{\ |
| 21 |
. ds C` "" |
| 22 |
. ds C' "" |
| 23 |
'br\} |
| 24 |
.el\{\ |
| 25 |
. ds C` |
| 26 |
. ds C' |
| 27 |
'br\} |
| 28 |
.\" |
| 29 |
.\" Escape single quotes in literal strings from groff's Unicode transform. |
| 30 |
.ie \n(.g .ds Aq \(aq |
| 31 |
.el .ds Aq ' |
| 32 |
.\" |
| 33 |
.\" If the F register is >0, we'll generate index entries on stderr for |
| 34 |
.\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index |
| 35 |
.\" entries marked with X<> in POD. Of course, you'll have to process the |
| 36 |
.\" output yourself in some meaningful fashion. |
| 37 |
.\" |
| 38 |
.\" Avoid warning from groff about undefined register 'F'. |
| 39 |
.de IX |
| 40 |
.. |
| 41 |
.nr rF 0 |
| 42 |
.if \n(.g .if rF .nr rF 1 |
| 43 |
.if (\n(rF:(\n(.g==0)) \{\ |
| 44 |
. if \nF \{\ |
| 45 |
. de IX |
| 46 |
. tm Index:\\$1\t\\n%\t"\\$2" |
| 47 |
.. |
| 48 |
. if !\nF==2 \{\ |
| 49 |
. nr % 0 |
| 50 |
. nr F 2 |
| 51 |
. \} |
| 52 |
. \} |
| 53 |
.\} |
| 54 |
.rr rF |
| 55 |
.\" ======================================================================== |
| 56 |
.\" |
| 57 |
.IX Title "OSSL-GUIDE-TLS-CLIENT-BLOCK 7SSL" |
| 58 |
.TH OSSL-GUIDE-TLS-CLIENT-BLOCK 7SSL 2026-04-07 3.6.2 OpenSSL |
| 59 |
.\" For nroff, turn off justification. Always turn off hyphenation; it makes |
| 60 |
.\" way too many mistakes in technical documents. |
| 61 |
.if n .ad l |
| 62 |
.nh |
| 63 |
.SH NAME |
| 64 |
ossl\-guide\-tls\-client\-block |
| 65 |
\&\- OpenSSL Guide: Writing a simple blocking TLS client |
| 66 |
.SH "SIMPLE BLOCKING TLS CLIENT EXAMPLE" |
| 67 |
.IX Header "SIMPLE BLOCKING TLS CLIENT EXAMPLE" |
| 68 |
This page will present various source code samples demonstrating how to write |
| 69 |
a simple TLS client application which connects to a server, sends an HTTP/1.0 |
| 70 |
request to it, and reads back the response. |
| 71 |
.PP |
| 72 |
We use a blocking socket for the purposes of this example. This means that |
| 73 |
attempting to read data from a socket that has no data available on it to read |
| 74 |
will block (and the function will not return), until data becomes available. |
| 75 |
For example, this can happen if we have sent our request, but we are still |
| 76 |
waiting for the server's response. Similarly any attempts to write to a socket |
| 77 |
that is not able to write at the moment will block until writing is possible. |
| 78 |
.PP |
| 79 |
This blocking behaviour simplifies the implementation of a client because you do |
| 80 |
not have to worry about what happens if data is not yet available. The |
| 81 |
application will simply wait until it is available. |
| 82 |
.PP |
| 83 |
The complete source code for this example blocking TLS client is available in |
| 84 |
the \fBdemos/guide\fR directory of the OpenSSL source distribution in the file |
| 85 |
\&\fBtls\-client\-block.c\fR. It is also available online at |
| 86 |
<https://github.com/openssl/openssl/blob/master/demos/guide/tls\-client\-block.c>. |
| 87 |
.PP |
| 88 |
We assume that you already have OpenSSL installed on your system; that you |
| 89 |
already have some fundamental understanding of OpenSSL concepts and TLS (see |
| 90 |
\&\fBossl\-guide\-libraries\-introduction\fR\|(7) and \fBossl\-guide\-tls\-introduction\fR\|(7)); |
| 91 |
and that you know how to write and build C code and link it against the |
| 92 |
libcrypto and libssl libraries that are provided by OpenSSL. It also assumes |
| 93 |
that you have a basic understanding of TCP/IP and sockets. |
| 94 |
.SS "Creating the SSL_CTX and SSL objects" |
| 95 |
.IX Subsection "Creating the SSL_CTX and SSL objects" |
| 96 |
The first step is to create an \fBSSL_CTX\fR object for our client. We use the |
| 97 |
\&\fBSSL_CTX_new\fR\|(3) function for this purpose. We could alternatively use |
| 98 |
\&\fBSSL_CTX_new_ex\fR\|(3) if we want to associate the \fBSSL_CTX\fR with a particular |
| 99 |
\&\fBOSSL_LIB_CTX\fR (see \fBossl\-guide\-libraries\-introduction\fR\|(7) to learn about |
| 100 |
\&\fBOSSL_LIB_CTX\fR). We pass as an argument the return value of the function |
| 101 |
\&\fBTLS_client_method\fR\|(3). You should use this method whenever you are writing a |
| 102 |
TLS client. This method will automatically use TLS version negotiation to select |
| 103 |
the highest version of the protocol that is mutually supported by both the |
| 104 |
client and the server. |
| 105 |
.PP |
| 106 |
.Vb 10 |
| 107 |
\& /* |
| 108 |
\& * Create an SSL_CTX which we can use to create SSL objects from. We |
| 109 |
\& * want an SSL_CTX for creating clients so we use TLS_client_method() |
| 110 |
\& * here. |
| 111 |
\& */ |
| 112 |
\& ctx = SSL_CTX_new(TLS_client_method()); |
| 113 |
\& if (ctx == NULL) { |
| 114 |
\& printf("Failed to create the SSL_CTX\en"); |
| 115 |
\& goto end; |
| 116 |
\& } |
| 117 |
.Ve |
| 118 |
.PP |
| 119 |
Since we are writing a client we must ensure that we verify the server's |
| 120 |
certificate. We do this by calling the \fBSSL_CTX_set_verify\fR\|(3) function and |
| 121 |
pass the \fBSSL_VERIFY_PEER\fR value to it. The final argument to this function |
| 122 |
is a callback that you can optionally supply to override the default handling |
| 123 |
for certificate verification. Most applications do not need to do this so this |
| 124 |
can safely be set to NULL to get the default handling. |
| 125 |
.PP |
| 126 |
.Vb 6 |
| 127 |
\& /* |
| 128 |
\& * Configure the client to abort the handshake if certificate |
| 129 |
\& * verification fails. Virtually all clients should do this unless you |
| 130 |
\& * really know what you are doing. |
| 131 |
\& */ |
| 132 |
\& SSL_CTX_set_verify(ctx, SSL_VERIFY_PEER, NULL); |
| 133 |
.Ve |
| 134 |
.PP |
| 135 |
In order for certificate verification to be successful you must have configured |
| 136 |
where the trusted certificate store to be used is located (see |
| 137 |
\&\fBossl\-guide\-tls\-introduction\fR\|(7)). In most cases you just want to use the |
| 138 |
default store so we call \fBSSL_CTX_set_default_verify_paths\fR\|(3). |
| 139 |
.PP |
| 140 |
.Vb 5 |
| 141 |
\& /* Use the default trusted certificate store */ |
| 142 |
\& if (!SSL_CTX_set_default_verify_paths(ctx)) { |
| 143 |
\& printf("Failed to set the default trusted certificate store\en"); |
| 144 |
\& goto end; |
| 145 |
\& } |
| 146 |
.Ve |
| 147 |
.PP |
| 148 |
We would also like to restrict the TLS versions that we are willing to accept to |
| 149 |
TLSv1.2 or above. TLS protocol versions earlier than that are generally to be |
| 150 |
avoided where possible. We can do that using |
| 151 |
\&\fBSSL_CTX_set_min_proto_version\fR\|(3): |
| 152 |
.PP |
| 153 |
.Vb 8 |
| 154 |
\& /* |
| 155 |
\& * TLSv1.1 or earlier are deprecated by IETF and are generally to be |
| 156 |
\& * avoided if possible. We require a minimum TLS version of TLSv1.2. |
| 157 |
\& */ |
| 158 |
\& if (!SSL_CTX_set_min_proto_version(ctx, TLS1_2_VERSION)) { |
| 159 |
\& printf("Failed to set the minimum TLS protocol version\en"); |
| 160 |
\& goto end; |
| 161 |
\& } |
| 162 |
.Ve |
| 163 |
.PP |
| 164 |
That is all the setup that we need to do for the \fBSSL_CTX\fR, so next we need to |
| 165 |
create an \fBSSL\fR object to represent the TLS connection. In a real application |
| 166 |
we might expect to be creating more than one TLS connection over time. In that |
| 167 |
case we would expect to reuse the \fBSSL_CTX\fR that we already created each time. |
| 168 |
There is no need to repeat those steps. In fact it is best not to since certain |
| 169 |
internal resources are cached in the \fBSSL_CTX\fR. You will get better performance |
| 170 |
by reusing an existing \fBSSL_CTX\fR instead of creating a new one each time. |
| 171 |
.PP |
| 172 |
Creating the \fBSSL\fR object is a simple matter of calling the \fBSSL_new\|(3)\fR |
| 173 |
function and passing the \fBSSL_CTX\fR we created as an argument. |
| 174 |
.PP |
| 175 |
.Vb 6 |
| 176 |
\& /* Create an SSL object to represent the TLS connection */ |
| 177 |
\& ssl = SSL_new(ctx); |
| 178 |
\& if (ssl == NULL) { |
| 179 |
\& printf("Failed to create the SSL object\en"); |
| 180 |
\& goto end; |
| 181 |
\& } |
| 182 |
.Ve |
| 183 |
.SS "Creating the socket and BIO" |
| 184 |
.IX Subsection "Creating the socket and BIO" |
| 185 |
TLS data is transmitted over an underlying transport layer. Normally a TCP |
| 186 |
socket. It is the application's responsibility for ensuring that the socket is |
| 187 |
created and associated with an SSL object (via a BIO). |
| 188 |
.PP |
| 189 |
Socket creation for use by a client is typically a 2 step process, i.e. |
| 190 |
constructing the socket; and connecting the socket. |
| 191 |
.PP |
| 192 |
How to construct a socket is platform specific \- but most platforms (including |
| 193 |
Windows) provide a POSIX compatible interface via the \fIsocket\fR function, e.g. |
| 194 |
to create an IPv4 TCP socket: |
| 195 |
.PP |
| 196 |
.Vb 1 |
| 197 |
\& int sock; |
| 198 |
\& |
| 199 |
\& sock = socket(AF_INET, SOCK_STREAM, 0); |
| 200 |
\& if (sock == \-1) |
| 201 |
\& return NULL; |
| 202 |
.Ve |
| 203 |
.PP |
| 204 |
Once the socket is constructed it must be connected to the remote server. Again |
| 205 |
the details are platform specific but most platforms (including Windows) |
| 206 |
provide the POSIX compatible \fIconnect\fR function. For example: |
| 207 |
.PP |
| 208 |
.Vb 2 |
| 209 |
\& struct sockaddr_in serveraddr; |
| 210 |
\& struct hostent *server; |
| 211 |
\& |
| 212 |
\& server = gethostbyname("www.openssl.org"); |
| 213 |
\& if (server == NULL) { |
| 214 |
\& close(sock); |
| 215 |
\& return NULL; |
| 216 |
\& } |
| 217 |
\& |
| 218 |
\& memset(&serveraddr, 0, sizeof(serveraddr)); |
| 219 |
\& serveraddr.sin_family = server\->h_addrtype; |
| 220 |
\& serveraddr.sin_port = htons(443); |
| 221 |
\& memcpy(&serveraddr.sin_addr.s_addr, server\->h_addr, server\->h_length); |
| 222 |
\& |
| 223 |
\& if (connect(sock, (struct sockaddr *)&serveraddr, |
| 224 |
\& sizeof(serveraddr)) == \-1) { |
| 225 |
\& close(sock); |
| 226 |
\& return NULL; |
| 227 |
\& } |
| 228 |
.Ve |
| 229 |
.PP |
| 230 |
OpenSSL provides portable helper functions to do these tasks which also |
| 231 |
integrate into the OpenSSL error system to log error data, e.g. |
| 232 |
.PP |
| 233 |
.Vb 3 |
| 234 |
\& int sock = \-1; |
| 235 |
\& BIO_ADDRINFO *res; |
| 236 |
\& const BIO_ADDRINFO *ai = NULL; |
| 237 |
\& |
| 238 |
\& /* |
| 239 |
\& * Lookup IP address info for the server. |
| 240 |
\& */ |
| 241 |
\& if (!BIO_lookup_ex(hostname, port, BIO_LOOKUP_CLIENT, family, SOCK_STREAM, 0, |
| 242 |
\& &res)) |
| 243 |
\& return NULL; |
| 244 |
\& |
| 245 |
\& /* |
| 246 |
\& * Loop through all the possible addresses for the server and find one |
| 247 |
\& * we can connect to. |
| 248 |
\& */ |
| 249 |
\& for (ai = res; ai != NULL; ai = BIO_ADDRINFO_next(ai)) { |
| 250 |
\& /* |
| 251 |
\& * Create a TCP socket. We could equally use non\-OpenSSL calls such |
| 252 |
\& * as "socket" here for this and the subsequent connect and close |
| 253 |
\& * functions. But for portability reasons and also so that we get |
| 254 |
\& * errors on the OpenSSL stack in the event of a failure we use |
| 255 |
\& * OpenSSL\*(Aqs versions of these functions. |
| 256 |
\& */ |
| 257 |
\& sock = BIO_socket(BIO_ADDRINFO_family(ai), SOCK_STREAM, 0, 0); |
| 258 |
\& if (sock == \-1) |
| 259 |
\& continue; |
| 260 |
\& |
| 261 |
\& /* Connect the socket to the server\*(Aqs address */ |
| 262 |
\& if (!BIO_connect(sock, BIO_ADDRINFO_address(ai), BIO_SOCK_NODELAY)) { |
| 263 |
\& BIO_closesocket(sock); |
| 264 |
\& sock = \-1; |
| 265 |
\& continue; |
| 266 |
\& } |
| 267 |
\& |
| 268 |
\& /* We have a connected socket so break out of the loop */ |
| 269 |
\& break; |
| 270 |
\& } |
| 271 |
\& |
| 272 |
\& /* Free the address information resources we allocated earlier */ |
| 273 |
\& BIO_ADDRINFO_free(res); |
| 274 |
.Ve |
| 275 |
.PP |
| 276 |
See \fBBIO_lookup_ex\fR\|(3), \fBBIO_socket\fR\|(3), \fBBIO_connect\fR\|(3), |
| 277 |
\&\fBBIO_closesocket\fR\|(3), \fBBIO_ADDRINFO_next\fR\|(3), \fBBIO_ADDRINFO_address\fR\|(3) and |
| 278 |
\&\fBBIO_ADDRINFO_free\fR\|(3) for further information on the functions used here. In |
| 279 |
the above example code the \fBhostname\fR and \fBport\fR variables are strings, e.g. |
| 280 |
"www.example.com" and "443". Note also the use of the family variable, which |
| 281 |
can take the values of AF_INET or AF_INET6 based on the command line \-6 option, |
| 282 |
to allow specific connections to an ipv4 or ipv6 enabled host. |
| 283 |
.PP |
| 284 |
Sockets created using the methods described above will automatically be blocking |
| 285 |
sockets \- which is exactly what we want for this example. |
| 286 |
.PP |
| 287 |
Once the socket has been created and connected we need to associate it with a |
| 288 |
BIO object: |
| 289 |
.PP |
| 290 |
.Vb 1 |
| 291 |
\& BIO *bio; |
| 292 |
\& |
| 293 |
\& /* Create a BIO to wrap the socket */ |
| 294 |
\& bio = BIO_new(BIO_s_socket()); |
| 295 |
\& if (bio == NULL) { |
| 296 |
\& BIO_closesocket(sock); |
| 297 |
\& return NULL; |
| 298 |
\& } |
| 299 |
\& |
| 300 |
\& /* |
| 301 |
\& * Associate the newly created BIO with the underlying socket. By |
| 302 |
\& * passing BIO_CLOSE here the socket will be automatically closed when |
| 303 |
\& * the BIO is freed. Alternatively you can use BIO_NOCLOSE, in which |
| 304 |
\& * case you must close the socket explicitly when it is no longer |
| 305 |
\& * needed. |
| 306 |
\& */ |
| 307 |
\& BIO_set_fd(bio, sock, BIO_CLOSE); |
| 308 |
.Ve |
| 309 |
.PP |
| 310 |
See \fBBIO_new\fR\|(3), \fBBIO_s_socket\fR\|(3) and \fBBIO_set_fd\fR\|(3) for further |
| 311 |
information on these functions. |
| 312 |
.PP |
| 313 |
Finally we associate the \fBSSL\fR object we created earlier with the \fBBIO\fR using |
| 314 |
the \fBSSL_set_bio\fR\|(3) function. Note that this passes ownership of the \fBBIO\fR |
| 315 |
object to the \fBSSL\fR object. Once ownership is passed the SSL object is |
| 316 |
responsible for its management and will free it automatically when the \fBSSL\fR is |
| 317 |
freed. So, once \fBSSL_set_bio\fR\|(3) has been been called, you should not call |
| 318 |
\&\fBBIO_free\fR\|(3) on the \fBBIO\fR. |
| 319 |
.PP |
| 320 |
.Vb 1 |
| 321 |
\& SSL_set_bio(ssl, bio, bio); |
| 322 |
.Ve |
| 323 |
.SS "Setting the server's hostname" |
| 324 |
.IX Subsection "Setting the server's hostname" |
| 325 |
We have already connected our underlying socket to the server, but the client |
| 326 |
still needs to know the server's hostname. It uses this information for 2 key |
| 327 |
purposes and we need to set the hostname for each one. |
| 328 |
.PP |
| 329 |
Firstly, the server's hostname is included in the initial ClientHello message |
| 330 |
sent by the client. This is known as the Server Name Indication (SNI). This is |
| 331 |
important because it is common for multiple hostnames to be fronted by a single |
| 332 |
server that handles requests for all of them. In other words a single server may |
| 333 |
have multiple hostnames associated with it and it is important to indicate which |
| 334 |
one we want to connect to. Without this information we may get a handshake |
| 335 |
failure, or we may get connected to the "default" server which may not be the |
| 336 |
one we were expecting. |
| 337 |
.PP |
| 338 |
To set the SNI hostname data we call the \fBSSL_set_tlsext_host_name\fR\|(3) function |
| 339 |
like this: |
| 340 |
.PP |
| 341 |
.Vb 8 |
| 342 |
\& /* |
| 343 |
\& * Tell the server during the handshake which hostname we are attempting |
| 344 |
\& * to connect to in case the server supports multiple hosts. |
| 345 |
\& */ |
| 346 |
\& if (!SSL_set_tlsext_host_name(ssl, hostname)) { |
| 347 |
\& printf("Failed to set the SNI hostname\en"); |
| 348 |
\& goto end; |
| 349 |
\& } |
| 350 |
.Ve |
| 351 |
.PP |
| 352 |
Here the \f(CW\*(C`hostname\*(C'\fR argument is a string representing the hostname of the |
| 353 |
server, e.g. "www.example.com". |
| 354 |
.PP |
| 355 |
Secondly, we need to tell OpenSSL what hostname we expect to see in the |
| 356 |
certificate coming back from the server. This is almost always the same one that |
| 357 |
we asked for in the original request. This is important because, without this, |
| 358 |
we do not verify that the hostname in the certificate is what we expect it to be |
| 359 |
and any certificate is acceptable unless your application explicitly checks this |
| 360 |
itself. We do this via the \fBSSL_set1_host\fR\|(3) function: |
| 361 |
.PP |
| 362 |
.Vb 10 |
| 363 |
\& /* |
| 364 |
\& * Ensure we check during certificate verification that the server has |
| 365 |
\& * supplied a certificate for the hostname that we were expecting. |
| 366 |
\& * Virtually all clients should do this unless you really know what you |
| 367 |
\& * are doing. |
| 368 |
\& */ |
| 369 |
\& if (!SSL_set1_host(ssl, hostname)) { |
| 370 |
\& printf("Failed to set the certificate verification hostname"); |
| 371 |
\& goto end; |
| 372 |
\& } |
| 373 |
.Ve |
| 374 |
.PP |
| 375 |
All of the above steps must happen before we attempt to perform the handshake |
| 376 |
otherwise they will have no effect. |
| 377 |
.SS "Performing the handshake" |
| 378 |
.IX Subsection "Performing the handshake" |
| 379 |
Before we can start sending or receiving application data over a TLS connection |
| 380 |
the TLS handshake must be performed. We can do this explicitly via the |
| 381 |
\&\fBSSL_connect\fR\|(3) function. |
| 382 |
.PP |
| 383 |
.Vb 12 |
| 384 |
\& /* Do the handshake with the server */ |
| 385 |
\& if (SSL_connect(ssl) < 1) { |
| 386 |
\& printf("Failed to connect to the server\en"); |
| 387 |
\& /* |
| 388 |
\& * If the failure is due to a verification error we can get more |
| 389 |
\& * information about it from SSL_get_verify_result(). |
| 390 |
\& */ |
| 391 |
\& if (SSL_get_verify_result(ssl) != X509_V_OK) |
| 392 |
\& printf("Verify error: %s\en", |
| 393 |
\& X509_verify_cert_error_string(SSL_get_verify_result(ssl))); |
| 394 |
\& goto end; |
| 395 |
\& } |
| 396 |
.Ve |
| 397 |
.PP |
| 398 |
The \fBSSL_connect\fR\|(3) function can return 1, 0 or less than 0. Only a return |
| 399 |
value of 1 is considered a success. For a simple blocking client we only need |
| 400 |
to concern ourselves with whether the call was successful or not. Anything else |
| 401 |
indicates that we have failed to connect to the server. |
| 402 |
.PP |
| 403 |
A common cause of failures at this stage is due to a problem verifying the |
| 404 |
server's certificate. For example if the certificate has expired, or it is not |
| 405 |
signed by a CA in our trusted certificate store. We can use the |
| 406 |
\&\fBSSL_get_verify_result\fR\|(3) function to find out more information about the |
| 407 |
verification failure. A return value of \fBX509_V_OK\fR indicates that the |
| 408 |
verification was successful (so the connection error must be due to some other |
| 409 |
cause). Otherwise we use the \fBX509_verify_cert_error_string\fR\|(3) function to get |
| 410 |
a human readable error message. |
| 411 |
.SS "Sending and receiving data" |
| 412 |
.IX Subsection "Sending and receiving data" |
| 413 |
Once the handshake is complete we are able to send and receive application data. |
| 414 |
Exactly what data is sent and in what order is usually controlled by some |
| 415 |
application level protocol. In this example we are using HTTP 1.0 which is a |
| 416 |
very simple request and response protocol. The client sends a request to the |
| 417 |
server. The server sends the response data and then immediately closes down the |
| 418 |
connection. |
| 419 |
.PP |
| 420 |
To send data to the server we use the \fBSSL_write_ex\fR\|(3) function and to receive |
| 421 |
data from the server we use the \fBSSL_read_ex\fR\|(3) function. In HTTP 1.0 the |
| 422 |
client always writes data first. Our HTTP request will include the hostname that |
| 423 |
we are connecting to. For simplicity, we write the HTTP request in three |
| 424 |
chunks. First we write the start of the request. Secondly we write the hostname |
| 425 |
we are sending the request to. Finally we send the end of the request. |
| 426 |
.PP |
| 427 |
.Vb 3 |
| 428 |
\& size_t written; |
| 429 |
\& const char *request_start = "GET / HTTP/1.0\er\enConnection: close\er\enHost: "; |
| 430 |
\& const char *request_end = "\er\en\er\en"; |
| 431 |
\& |
| 432 |
\& /* Write an HTTP GET request to the peer */ |
| 433 |
\& if (!SSL_write_ex(ssl, request_start, strlen(request_start), &written)) { |
| 434 |
\& printf("Failed to write start of HTTP request\en"); |
| 435 |
\& goto end; |
| 436 |
\& } |
| 437 |
\& if (!SSL_write_ex(ssl, hostname, strlen(hostname), &written)) { |
| 438 |
\& printf("Failed to write hostname in HTTP request\en"); |
| 439 |
\& goto end; |
| 440 |
\& } |
| 441 |
\& if (!SSL_write_ex(ssl, request_end, strlen(request_end), &written)) { |
| 442 |
\& printf("Failed to write end of HTTP request\en"); |
| 443 |
\& goto end; |
| 444 |
\& } |
| 445 |
.Ve |
| 446 |
.PP |
| 447 |
The \fBSSL_write_ex\fR\|(3) function returns 0 if it fails and 1 if it is successful. |
| 448 |
If it is successful then we can proceed to waiting for a response from the |
| 449 |
server. |
| 450 |
.PP |
| 451 |
.Vb 2 |
| 452 |
\& size_t readbytes; |
| 453 |
\& char buf[160]; |
| 454 |
\& |
| 455 |
\& /* |
| 456 |
\& * Get up to sizeof(buf) bytes of the response. We keep reading until the |
| 457 |
\& * server closes the connection. |
| 458 |
\& */ |
| 459 |
\& while (SSL_read_ex(ssl, buf, sizeof(buf), &readbytes)) { |
| 460 |
\& /* |
| 461 |
\& * OpenSSL does not guarantee that the returned data is a string or |
| 462 |
\& * that it is NUL terminated so we use fwrite() to write the exact |
| 463 |
\& * number of bytes that we read. The data could be non\-printable or |
| 464 |
\& * have NUL characters in the middle of it. For this simple example |
| 465 |
\& * we\*(Aqre going to print it to stdout anyway. |
| 466 |
\& */ |
| 467 |
\& fwrite(buf, 1, readbytes, stdout); |
| 468 |
\& } |
| 469 |
\& /* In case the response didn\*(Aqt finish with a newline we add one now */ |
| 470 |
\& printf("\en"); |
| 471 |
.Ve |
| 472 |
.PP |
| 473 |
We use the \fBSSL_read_ex\fR\|(3) function to read the response. We don't know |
| 474 |
exactly how much data we are going to receive back so we enter a loop reading |
| 475 |
blocks of data from the server and printing each block that we receive to the |
| 476 |
screen. The loop ends as soon as \fBSSL_read_ex\fR\|(3) returns 0 \- meaning that it |
| 477 |
failed to read any data. |
| 478 |
.PP |
| 479 |
A failure to read data could mean that there has been some error, or it could |
| 480 |
simply mean that server has sent all the data that it wants to send and has |
| 481 |
indicated that it has finished by sending a "close_notify" alert. This alert is |
| 482 |
a TLS protocol level message indicating that the endpoint has finished sending |
| 483 |
all of its data and it will not send any more. Both of these conditions result |
| 484 |
in a 0 return value from \fBSSL_read_ex\fR\|(3) and we need to use the function |
| 485 |
\&\fBSSL_get_error\fR\|(3) to determine the cause of the 0 return value. |
| 486 |
.PP |
| 487 |
.Vb 10 |
| 488 |
\& /* |
| 489 |
\& * Check whether we finished the while loop above normally or as the |
| 490 |
\& * result of an error. The 0 argument to SSL_get_error() is the return |
| 491 |
\& * code we received from the SSL_read_ex() call. It must be 0 in order |
| 492 |
\& * to get here. Normal completion is indicated by SSL_ERROR_ZERO_RETURN. |
| 493 |
\& */ |
| 494 |
\& if (SSL_get_error(ssl, 0) != SSL_ERROR_ZERO_RETURN) { |
| 495 |
\& /* |
| 496 |
\& * Some error occurred other than a graceful close down by the |
| 497 |
\& * peer |
| 498 |
\& */ |
| 499 |
\& printf ("Failed reading remaining data\en"); |
| 500 |
\& goto end; |
| 501 |
\& } |
| 502 |
.Ve |
| 503 |
.PP |
| 504 |
If \fBSSL_get_error\fR\|(3) returns \fBSSL_ERROR_ZERO_RETURN\fR then we know that the |
| 505 |
server has finished sending its data. Otherwise an error has occurred. |
| 506 |
.SS "Shutting down the connection" |
| 507 |
.IX Subsection "Shutting down the connection" |
| 508 |
Once we have finished reading data from the server then we are ready to close |
| 509 |
the connection down. We do this via the \fBSSL_shutdown\fR\|(3) function which has |
| 510 |
the effect of sending a TLS protocol level message (a "close_notify" alert) to |
| 511 |
the server saying that we have finished writing data: |
| 512 |
.PP |
| 513 |
.Vb 10 |
| 514 |
\& /* |
| 515 |
\& * The peer already shutdown gracefully (we know this because of the |
| 516 |
\& * SSL_ERROR_ZERO_RETURN above). We should do the same back. |
| 517 |
\& */ |
| 518 |
\& ret = SSL_shutdown(ssl); |
| 519 |
\& if (ret < 1) { |
| 520 |
\& /* |
| 521 |
\& * ret < 0 indicates an error. ret == 0 would be unexpected here |
| 522 |
\& * because that means "we\*(Aqve sent a close_notify and we\*(Aqre waiting |
| 523 |
\& * for one back". But we already know we got one from the peer |
| 524 |
\& * because of the SSL_ERROR_ZERO_RETURN above. |
| 525 |
\& */ |
| 526 |
\& printf("Error shutting down\en"); |
| 527 |
\& goto end; |
| 528 |
\& } |
| 529 |
.Ve |
| 530 |
.PP |
| 531 |
The \fBSSL_shutdown\fR\|(3) function will either return 1, 0, or less than 0. A |
| 532 |
return value of 1 is a success, and a return value less than 0 is an error. More |
| 533 |
precisely a return value of 1 means that we have sent a "close_notify" alert to |
| 534 |
the server, and that we have also received one back. A return value of 0 means |
| 535 |
that we have sent a "close_notify" alert to the server, but we have not yet |
| 536 |
received one back. Usually in this scenario you would call \fBSSL_shutdown\fR\|(3) |
| 537 |
again which (with a blocking socket) would block until the "close_notify" is |
| 538 |
received. However in this case we already know that the server has sent us a |
| 539 |
"close_notify" because of the SSL_ERROR_ZERO_RETURN that we received from the |
| 540 |
call to \fBSSL_read_ex\fR\|(3). So this scenario should never happen in practice. We |
| 541 |
just treat it as an error in this example. |
| 542 |
.SS "Final clean up" |
| 543 |
.IX Subsection "Final clean up" |
| 544 |
Before the application exits we have to clean up some memory that we allocated. |
| 545 |
If we are exiting due to an error we might also want to display further |
| 546 |
information about that error if it is available to the user: |
| 547 |
.PP |
| 548 |
.Vb 10 |
| 549 |
\& /* Success! */ |
| 550 |
\& res = EXIT_SUCCESS; |
| 551 |
\& end: |
| 552 |
\& /* |
| 553 |
\& * If something bad happened then we will dump the contents of the |
| 554 |
\& * OpenSSL error stack to stderr. There might be some useful diagnostic |
| 555 |
\& * information there. |
| 556 |
\& */ |
| 557 |
\& if (res == EXIT_FAILURE) |
| 558 |
\& ERR_print_errors_fp(stderr); |
| 559 |
\& |
| 560 |
\& /* |
| 561 |
\& * Free the resources we allocated. We do not free the BIO object here |
| 562 |
\& * because ownership of it was immediately transferred to the SSL object |
| 563 |
\& * via SSL_set_bio(). The BIO will be freed when we free the SSL object. |
| 564 |
\& */ |
| 565 |
\& SSL_free(ssl); |
| 566 |
\& SSL_CTX_free(ctx); |
| 567 |
\& return res; |
| 568 |
.Ve |
| 569 |
.PP |
| 570 |
To display errors we make use of the \fBERR_print_errors_fp\fR\|(3) function which |
| 571 |
simply dumps out the contents of any errors on the OpenSSL error stack to the |
| 572 |
specified location (in this case \fIstderr\fR). |
| 573 |
.PP |
| 574 |
We need to free up the \fBSSL\fR object that we created for the connection via the |
| 575 |
\&\fBSSL_free\fR\|(3) function. Also, since we are not going to be creating any more |
| 576 |
TLS connections we must also free up the \fBSSL_CTX\fR via a call to |
| 577 |
\&\fBSSL_CTX_free\fR\|(3). |
| 578 |
.SH TROUBLESHOOTING |
| 579 |
.IX Header "TROUBLESHOOTING" |
| 580 |
There are a number of things that might go wrong when running the demo |
| 581 |
application. This section describes some common things you might encounter. |
| 582 |
.SS "Failure to connect the underlying socket" |
| 583 |
.IX Subsection "Failure to connect the underlying socket" |
| 584 |
This could occur for numerous reasons. For example if there is a problem in the |
| 585 |
network route between the client and the server; or a firewall is blocking the |
| 586 |
communication; or the server is not in DNS. Check the network configuration. |
| 587 |
.SS "Verification failure of the server certificate" |
| 588 |
.IX Subsection "Verification failure of the server certificate" |
| 589 |
A verification failure of the server certificate would result in a failure when |
| 590 |
running the \fBSSL_connect\fR\|(3) function. \fBERR_print_errors_fp\fR\|(3) would display |
| 591 |
an error which would look something like this: |
| 592 |
.PP |
| 593 |
.Vb 2 |
| 594 |
\& Verify error: unable to get local issuer certificate |
| 595 |
\& 40E74AF1F47F0000:error:0A000086:SSL routines:tls_post_process_server_certificate:certificate verify failed:ssl/statem/statem_clnt.c:2069: |
| 596 |
.Ve |
| 597 |
.PP |
| 598 |
A server certificate verification failure could be caused for a number of |
| 599 |
reasons. For example |
| 600 |
.IP "Failure to correctly setup the trusted certificate store" 4 |
| 601 |
.IX Item "Failure to correctly setup the trusted certificate store" |
| 602 |
See the page \fBossl\-guide\-tls\-introduction\fR\|(7) and check that your trusted |
| 603 |
certificate store is correctly configured |
| 604 |
.IP "Unrecognised CA" 4 |
| 605 |
.IX Item "Unrecognised CA" |
| 606 |
If the CA used by the server's certificate is not in the trusted certificate |
| 607 |
store for the client then this will cause a verification failure during |
| 608 |
connection. Often this can occur if the server is using a self-signed |
| 609 |
certificate (i.e. a test certificate that has not been signed by a CA at all). |
| 610 |
.IP "Missing intermediate CAs" 4 |
| 611 |
.IX Item "Missing intermediate CAs" |
| 612 |
This is a server misconfiguration where the client has the relevant root CA in |
| 613 |
its trust store, but the server has not supplied all of the intermediate CA |
| 614 |
certificates between that root CA and the server's own certificate. Therefore |
| 615 |
a trust chain cannot be established. |
| 616 |
.IP "Mismatched hostname" 4 |
| 617 |
.IX Item "Mismatched hostname" |
| 618 |
If for some reason the hostname of the server that the client is expecting does |
| 619 |
not match the hostname in the certificate then this will cause verification to |
| 620 |
fail. |
| 621 |
.IP "Expired certificate" 4 |
| 622 |
.IX Item "Expired certificate" |
| 623 |
The date that the server's certificate is valid to has passed. |
| 624 |
.PP |
| 625 |
The "unable to get local issuer certificate" we saw in the example above means |
| 626 |
that we have been unable to find the issuer of the server's certificate (or one |
| 627 |
of its intermediate CA certificates) in our trusted certificate store (e.g. |
| 628 |
because the trusted certificate store is misconfigured, or there are missing |
| 629 |
intermediate CAs, or the issuer is simply unrecognised). |
| 630 |
.SH "FURTHER READING" |
| 631 |
.IX Header "FURTHER READING" |
| 632 |
See \fBossl\-guide\-tls\-client\-non\-block\fR\|(7) to read a tutorial on how to modify |
| 633 |
the client developed on this page to support a nonblocking socket. |
| 634 |
.PP |
| 635 |
See \fBossl\-guide\-tls\-server\-block\fR\|(7) for a tutorial on how to implement a |
| 636 |
simple TLS server handling one client at a time over a blocking socket. |
| 637 |
.PP |
| 638 |
See \fBossl\-guide\-quic\-client\-block\fR\|(7) to read a tutorial on how to modify the |
| 639 |
client developed on this page to support QUIC instead of TLS. |
| 640 |
.SH "SEE ALSO" |
| 641 |
.IX Header "SEE ALSO" |
| 642 |
\&\fBossl\-guide\-introduction\fR\|(7), \fBossl\-guide\-libraries\-introduction\fR\|(7), |
| 643 |
\&\fBossl\-guide\-libssl\-introduction\fR\|(7), \fBossl\-guide\-tls\-introduction\fR\|(7), |
| 644 |
\&\fBossl\-guide\-tls\-client\-non\-block\fR\|(7), \fBossl\-guide\-quic\-client\-block\fR\|(7) |
| 645 |
.SH COPYRIGHT |
| 646 |
.IX Header "COPYRIGHT" |
| 647 |
Copyright 2023\-2024 The OpenSSL Project Authors. All Rights Reserved. |
| 648 |
.PP |
| 649 |
Licensed under the Apache License 2.0 (the "License"). You may not use |
| 650 |
this file except in compliance with the License. You can obtain a copy |
| 651 |
in the file LICENSE in the source distribution or at |
| 652 |
<https://www.openssl.org/source/license.html>. |