diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-27 06:33:50 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-27 06:33:50 +0000 |
commit | fe39ffb8b90ae4e002ed73fe98617cd590abb467 (patch) | |
tree | b80e5956907d8aeaaffe4e4f0c068c0e6157ce8b /modules/proxy/mod_proxy_fcgi.c | |
parent | Initial commit. (diff) | |
download | apache2-upstream.tar.xz apache2-upstream.zip |
Adding upstream version 2.4.56.upstream/2.4.56upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to '')
-rw-r--r-- | modules/proxy/mod_proxy_fcgi.c | 1330 |
1 files changed, 1330 insertions, 0 deletions
diff --git a/modules/proxy/mod_proxy_fcgi.c b/modules/proxy/mod_proxy_fcgi.c new file mode 100644 index 0000000..3382b9b --- /dev/null +++ b/modules/proxy/mod_proxy_fcgi.c @@ -0,0 +1,1330 @@ +/* Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +#include "mod_proxy.h" +#include "util_fcgi.h" +#include "util_script.h" +#include "ap_expr.h" + +module AP_MODULE_DECLARE_DATA proxy_fcgi_module; + +typedef struct { + ap_expr_info_t *cond; + ap_expr_info_t *subst; + const char *envname; +} sei_entry; + +typedef struct { + int need_dirwalk; +} fcgi_req_config_t; + +/* We will assume FPM, but still differentiate */ +typedef enum { + BACKEND_DEFAULT_UNKNOWN = 0, + BACKEND_FPM, + BACKEND_GENERIC, +} fcgi_backend_t; + + +#define FCGI_MAY_BE_FPM(dconf) \ + (dconf && \ + ((dconf->backend_type == BACKEND_DEFAULT_UNKNOWN) || \ + (dconf->backend_type == BACKEND_FPM))) + +typedef struct { + fcgi_backend_t backend_type; + apr_array_header_t *env_fixups; +} fcgi_dirconf_t; + +/* + * Canonicalise http-like URLs. + * scheme is the scheme for the URL + * url is the URL starting with the first '/' + * def_port is the default port for this scheme. + */ +static int proxy_fcgi_canon(request_rec *r, char *url) +{ + char *host, sport[7]; + const char *err; + char *path; + apr_port_t port, def_port; + fcgi_req_config_t *rconf = NULL; + const char *pathinfo_type = NULL; + + if (ap_cstr_casecmpn(url, "fcgi:", 5) == 0) { + url += 5; + } + else { + return DECLINED; + } + + port = def_port = ap_proxy_port_of_scheme("fcgi"); + + ap_log_rerror(APLOG_MARK, APLOG_TRACE1, 0, r, + "canonicalising URL %s", url); + err = ap_proxy_canon_netloc(r->pool, &url, NULL, NULL, &host, &port); + if (err) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01059) + "error parsing URL %s: %s", url, err); + return HTTP_BAD_REQUEST; + } + + if (port != def_port) + apr_snprintf(sport, sizeof(sport), ":%d", port); + else + sport[0] = '\0'; + + if (ap_strchr_c(host, ':')) { + /* if literal IPv6 address */ + host = apr_pstrcat(r->pool, "[", host, "]", NULL); + } + + if (apr_table_get(r->notes, "proxy-nocanon")) { + path = url; /* this is the raw path */ + } + else { + path = ap_proxy_canonenc(r->pool, url, strlen(url), enc_path, 0, + r->proxyreq); + } + if (path == NULL) + return HTTP_BAD_REQUEST; + + r->filename = apr_pstrcat(r->pool, "proxy:fcgi://", host, sport, "/", + path, NULL); + + ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01060) + "set r->filename to %s", r->filename); + + rconf = ap_get_module_config(r->request_config, &proxy_fcgi_module); + if (rconf == NULL) { + rconf = apr_pcalloc(r->pool, sizeof(fcgi_req_config_t)); + ap_set_module_config(r->request_config, &proxy_fcgi_module, rconf); + } + + if (NULL != (pathinfo_type = apr_table_get(r->subprocess_env, "proxy-fcgi-pathinfo"))) { + /* It has to be on disk for this to work */ + if (!strcasecmp(pathinfo_type, "full")) { + rconf->need_dirwalk = 1; + ap_unescape_url_keep2f(path, 0); + } + else if (!strcasecmp(pathinfo_type, "first-dot")) { + char *split = ap_strchr(path, '.'); + if (split) { + char *slash = ap_strchr(split, '/'); + if (slash) { + r->path_info = apr_pstrdup(r->pool, slash); + ap_unescape_url_keep2f(r->path_info, 0); + *slash = '\0'; /* truncate path */ + } + } + } + else if (!strcasecmp(pathinfo_type, "last-dot")) { + char *split = ap_strrchr(path, '.'); + if (split) { + char *slash = ap_strchr(split, '/'); + if (slash) { + r->path_info = apr_pstrdup(r->pool, slash); + ap_unescape_url_keep2f(r->path_info, 0); + *slash = '\0'; /* truncate path */ + } + } + } + else { + /* before proxy-fcgi-pathinfo had multi-values. This requires the + * the FCGI server to fixup PATH_INFO because it's the entire path + */ + r->path_info = apr_pstrcat(r->pool, "/", path, NULL); + if (!strcasecmp(pathinfo_type, "unescape")) { + ap_unescape_url_keep2f(r->path_info, 0); + } + ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01061) + "set r->path_info to %s", r->path_info); + } + } + + return OK; +} + + +/* + ProxyFCGISetEnvIf "reqenv('PATH_INFO') =~ m#/foo(\d+)\.php$#" COVENV1 "$1" + ProxyFCGISetEnvIf "reqenv('PATH_INFO') =~ m#/foo(\d+)\.php$#" PATH_INFO "/foo.php" + ProxyFCGISetEnvIf "reqenv('PATH_TRANSLATED') =~ m#(/.*foo)(\d+)(.*)#" PATH_TRANSLATED "$1$3" +*/ +static apr_status_t fix_cgivars(request_rec *r, fcgi_dirconf_t *dconf) +{ + sei_entry *entries; + const char *err, *src; + int i = 0, rc = 0; + ap_regmatch_t regm[AP_MAX_REG_MATCH]; + + entries = (sei_entry *) dconf->env_fixups->elts; + for (i = 0; i < dconf->env_fixups->nelts; i++) { + sei_entry *entry = &entries[i]; + + rc = ap_expr_exec_re(r, entry->cond, AP_MAX_REG_MATCH, regm, &src, &err); + if (rc < 0) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(10241) + "fix_cgivars: Condition eval returned %d: %s", + rc, err); + return APR_EGENERAL; + } + else if (rc == 0) { + continue; /* evaluated false */ + } + + if (entry->envname[0] == '!') { + apr_table_unset(r->subprocess_env, entry->envname+1); + } + else { + const char *val = ap_expr_str_exec_re(r, entry->subst, AP_MAX_REG_MATCH, regm, &src, &err); + if (err) { + ap_log_rerror(APLOG_MARK, APLOG_INFO, 0, r, APLOGNO(03514) + "Error evaluating expression for replacement of %s: '%s'", + entry->envname, err); + continue; + } + if (APLOGrtrace4(r)) { + const char *oldval = apr_table_get(r->subprocess_env, entry->envname); + ap_log_rerror(APLOG_MARK, APLOG_TRACE4, 0, r, + "fix_cgivars: override %s from '%s' to '%s'", + entry->envname, oldval, val); + + } + apr_table_setn(r->subprocess_env, entry->envname, val); + } + } + return APR_SUCCESS; +} + +/* Wrapper for apr_socket_sendv that handles updating the worker stats. */ +static apr_status_t send_data(proxy_conn_rec *conn, + struct iovec *vec, + int nvec, + apr_size_t *len) +{ + apr_status_t rv = APR_SUCCESS; + apr_size_t written = 0, to_write = 0; + int i, offset; + apr_socket_t *s = conn->sock; + + for (i = 0; i < nvec; i++) { + to_write += vec[i].iov_len; + } + + offset = 0; + while (to_write) { + apr_size_t n = 0; + rv = apr_socket_sendv(s, vec + offset, nvec - offset, &n); + if (rv != APR_SUCCESS) { + break; + } + if (n > 0) { + written += n; + if (written >= to_write) + break; /* short circuit out */ + for (i = offset; i < nvec; ) { + if (n >= vec[i].iov_len) { + offset++; + n -= vec[i++].iov_len; + } else { + vec[i].iov_len -= n; + vec[i].iov_base = (char *) vec[i].iov_base + n; + break; + } + } + } + } + + conn->worker->s->transferred += written; + *len = written; + + return rv; +} + +/* Wrapper for apr_socket_recv that handles updating the worker stats. */ +static apr_status_t get_data(proxy_conn_rec *conn, + char *buffer, + apr_size_t *buflen) +{ + apr_status_t rv = apr_socket_recv(conn->sock, buffer, buflen); + + if (rv == APR_SUCCESS) { + conn->worker->s->read += *buflen; + } + + return rv; +} + +static apr_status_t get_data_full(proxy_conn_rec *conn, + char *buffer, + apr_size_t buflen) +{ + apr_size_t readlen; + apr_size_t cumulative_len = 0; + apr_status_t rv; + + do { + readlen = buflen - cumulative_len; + rv = get_data(conn, buffer + cumulative_len, &readlen); + if (rv != APR_SUCCESS) { + return rv; + } + cumulative_len += readlen; + } while (cumulative_len < buflen); + + return APR_SUCCESS; +} + +static apr_status_t send_begin_request(proxy_conn_rec *conn, + apr_uint16_t request_id) +{ + struct iovec vec[2]; + ap_fcgi_header header; + unsigned char farray[AP_FCGI_HEADER_LEN]; + ap_fcgi_begin_request_body brb; + unsigned char abrb[AP_FCGI_HEADER_LEN]; + apr_size_t len; + + ap_fcgi_fill_in_header(&header, AP_FCGI_BEGIN_REQUEST, request_id, + sizeof(abrb), 0); + + ap_fcgi_fill_in_request_body(&brb, AP_FCGI_RESPONDER, + ap_proxy_connection_reusable(conn) + ? AP_FCGI_KEEP_CONN : 0); + + ap_fcgi_header_to_array(&header, farray); + ap_fcgi_begin_request_body_to_array(&brb, abrb); + + vec[0].iov_base = (void *)farray; + vec[0].iov_len = sizeof(farray); + vec[1].iov_base = (void *)abrb; + vec[1].iov_len = sizeof(abrb); + + return send_data(conn, vec, 2, &len); +} + +static apr_status_t send_environment(proxy_conn_rec *conn, request_rec *r, + apr_pool_t *temp_pool, + apr_uint16_t request_id) +{ + const apr_array_header_t *envarr; + const apr_table_entry_t *elts; + struct iovec vec[2]; + ap_fcgi_header header; + unsigned char farray[AP_FCGI_HEADER_LEN]; + char *body; + apr_status_t rv; + apr_size_t avail_len, len, required_len; + int next_elem, starting_elem; + fcgi_req_config_t *rconf = ap_get_module_config(r->request_config, &proxy_fcgi_module); + fcgi_dirconf_t *dconf = ap_get_module_config(r->per_dir_config, &proxy_fcgi_module); + + if (rconf) { + if (rconf->need_dirwalk) { + ap_directory_walk(r); + } + } + + /* Strip proxy: prefixes */ + if (r->filename) { + char *newfname = NULL; + + if (!strncmp(r->filename, "proxy:balancer://", 17)) { + newfname = apr_pstrdup(r->pool, r->filename+17); + } + + if (!FCGI_MAY_BE_FPM(dconf)) { + if (!strncmp(r->filename, "proxy:fcgi://", 13)) { + /* If we strip this under FPM, and any internal redirect occurs + * on PATH_INFO, FPM may use PATH_TRANSLATED instead of + * SCRIPT_FILENAME (a la mod_fastcgi + Action). + */ + newfname = apr_pstrdup(r->pool, r->filename+13); + } + /* Query string in environment only */ + if (newfname && r->args && *r->args) { + char *qs = strrchr(newfname, '?'); + if (qs && !strcmp(qs+1, r->args)) { + *qs = '\0'; + } + } + } + + if (newfname) { + newfname = ap_strchr(newfname, '/'); + r->filename = newfname; + } + } + + ap_add_common_vars(r); + ap_add_cgi_vars(r); + + /* XXX are there any FastCGI specific env vars we need to send? */ + + /* Give admins final option to fine-tune env vars */ + if (APR_SUCCESS != (rv = fix_cgivars(r, dconf))) { + return rv; + } + + /* XXX mod_cgi/mod_cgid use ap_create_environment here, which fills in + * the TZ value specially. We could use that, but it would mean + * parsing the key/value pairs back OUT of the allocated env array, + * not to mention allocating a totally useless array in the first + * place, which would suck. */ + + envarr = apr_table_elts(r->subprocess_env); + elts = (const apr_table_entry_t *) envarr->elts; + + if (APLOGrtrace8(r)) { + int i; + + for (i = 0; i < envarr->nelts; ++i) { + ap_log_rerror(APLOG_MARK, APLOG_TRACE8, 0, r, APLOGNO(01062) + "sending env var '%s' value '%s'", + elts[i].key, elts[i].val); + } + } + + /* Send envvars over in as many FastCGI records as it takes, */ + next_elem = 0; /* starting with the first one */ + + avail_len = 16 * 1024; /* our limit per record, which could have been up + * to AP_FCGI_MAX_CONTENT_LEN + */ + + while (next_elem < envarr->nelts) { + starting_elem = next_elem; + required_len = ap_fcgi_encoded_env_len(r->subprocess_env, + avail_len, + &next_elem); + + if (!required_len) { + if (next_elem < envarr->nelts) { + ap_log_rerror(APLOG_MARK, APLOG_WARNING, 0, r, + APLOGNO(02536) "couldn't encode envvar '%s' in %" + APR_SIZE_T_FMT " bytes", + elts[next_elem].key, avail_len); + /* skip this envvar and continue */ + ++next_elem; + continue; + } + /* only an unused element at the end of the array */ + break; + } + + body = apr_palloc(temp_pool, required_len); + rv = ap_fcgi_encode_env(r, r->subprocess_env, body, required_len, + &starting_elem); + /* we pre-compute, so we can't run out of space */ + ap_assert(rv == APR_SUCCESS); + /* compute and encode must be in sync */ + ap_assert(starting_elem == next_elem); + + ap_fcgi_fill_in_header(&header, AP_FCGI_PARAMS, request_id, + (apr_uint16_t)required_len, 0); + ap_fcgi_header_to_array(&header, farray); + + vec[0].iov_base = (void *)farray; + vec[0].iov_len = sizeof(farray); + vec[1].iov_base = body; + vec[1].iov_len = required_len; + + rv = send_data(conn, vec, 2, &len); + apr_pool_clear(temp_pool); + + if (rv) { + return rv; + } + } + + /* Envvars sent, so say we're done */ + ap_fcgi_fill_in_header(&header, AP_FCGI_PARAMS, request_id, 0, 0); + ap_fcgi_header_to_array(&header, farray); + + vec[0].iov_base = (void *)farray; + vec[0].iov_len = sizeof(farray); + + return send_data(conn, vec, 1, &len); +} + +enum { + HDR_STATE_READING_HEADERS, + HDR_STATE_GOT_CR, + HDR_STATE_GOT_CRLF, + HDR_STATE_GOT_CRLFCR, + HDR_STATE_GOT_LF, + HDR_STATE_DONE_WITH_HEADERS +}; + +/* Try to find the end of the script headers in the response from the back + * end fastcgi server. STATE holds the current header parsing state for this + * request. + * + * Returns 0 if it can't find the end of the headers, and 1 if it found the + * end of the headers. */ +static int handle_headers(request_rec *r, int *state, + const char *readbuf, apr_size_t readlen) +{ + const char *itr = readbuf; + + while (readlen--) { + if (*itr == '\r') { + switch (*state) { + case HDR_STATE_GOT_CRLF: + *state = HDR_STATE_GOT_CRLFCR; + break; + + default: + *state = HDR_STATE_GOT_CR; + break; + } + } + else if (*itr == '\n') { + switch (*state) { + case HDR_STATE_GOT_LF: + *state = HDR_STATE_DONE_WITH_HEADERS; + break; + + case HDR_STATE_GOT_CR: + *state = HDR_STATE_GOT_CRLF; + break; + + case HDR_STATE_GOT_CRLFCR: + *state = HDR_STATE_DONE_WITH_HEADERS; + break; + + default: + *state = HDR_STATE_GOT_LF; + break; + } + } + else { + *state = HDR_STATE_READING_HEADERS; + } + + if (*state == HDR_STATE_DONE_WITH_HEADERS) + break; + + ++itr; + } + + if (*state == HDR_STATE_DONE_WITH_HEADERS) { + return 1; + } + + return 0; +} + +static apr_status_t dispatch(proxy_conn_rec *conn, proxy_dir_conf *conf, + request_rec *r, apr_pool_t *setaside_pool, + apr_uint16_t request_id, const char **err, + int *bad_request, int *has_responded, + apr_bucket_brigade *input_brigade) +{ + apr_bucket_brigade *ib, *ob; + int seen_end_of_headers = 0, done = 0, ignore_body = 0; + apr_status_t rv = APR_SUCCESS; + int script_error_status = HTTP_OK; + conn_rec *c = r->connection; + struct iovec vec[2]; + ap_fcgi_header header; + unsigned char farray[AP_FCGI_HEADER_LEN]; + apr_pollfd_t pfd; + apr_pollfd_t *flushpoll = NULL; + apr_int32_t flushpoll_fd; + int header_state = HDR_STATE_READING_HEADERS; + char stack_iobuf[AP_IOBUFSIZE]; + apr_size_t iobuf_size = AP_IOBUFSIZE; + char *iobuf = stack_iobuf; + + *err = NULL; + if (conn->worker->s->io_buffer_size_set) { + iobuf_size = conn->worker->s->io_buffer_size; + iobuf = apr_palloc(r->pool, iobuf_size); + } + + pfd.desc_type = APR_POLL_SOCKET; + pfd.desc.s = conn->sock; + pfd.p = r->pool; + pfd.reqevents = APR_POLLIN | APR_POLLOUT; + + if (conn->worker->s->flush_packets == flush_auto) { + flushpoll = apr_pcalloc(r->pool, sizeof(apr_pollfd_t)); + flushpoll->reqevents = APR_POLLIN; + flushpoll->desc_type = APR_POLL_SOCKET; + flushpoll->desc.s = conn->sock; + } + + ib = apr_brigade_create(r->pool, c->bucket_alloc); + ob = apr_brigade_create(r->pool, c->bucket_alloc); + + while (! done) { + apr_interval_time_t timeout; + apr_size_t len; + int n; + + /* We need SOME kind of timeout here, or virtually anything will + * cause timeout errors. */ + apr_socket_timeout_get(conn->sock, &timeout); + + rv = apr_poll(&pfd, 1, &n, timeout); + if (rv != APR_SUCCESS) { + if (APR_STATUS_IS_EINTR(rv)) { + continue; + } + *err = "polling"; + break; + } + + if (pfd.rtnevents & APR_POLLOUT) { + apr_size_t to_send, writebuflen; + int last_stdin = 0; + char *iobuf_cursor; + + if (APR_BRIGADE_EMPTY(input_brigade)) { + rv = ap_get_brigade(r->input_filters, ib, + AP_MODE_READBYTES, APR_BLOCK_READ, + iobuf_size); + } + else { + apr_bucket *e; + APR_BRIGADE_CONCAT(ib, input_brigade); + rv = apr_brigade_partition(ib, iobuf_size, &e); + if (rv == APR_SUCCESS) { + while (e != APR_BRIGADE_SENTINEL(ib) + && APR_BUCKET_IS_METADATA(e)) { + e = APR_BUCKET_NEXT(e); + } + apr_brigade_split_ex(ib, e, input_brigade); + } + else if (rv == APR_INCOMPLETE) { + rv = APR_SUCCESS; + } + } + if (rv != APR_SUCCESS) { + *err = "reading input brigade"; + *bad_request = 1; + break; + } + + if (APR_BUCKET_IS_EOS(APR_BRIGADE_LAST(ib))) { + last_stdin = 1; + } + + writebuflen = iobuf_size; + + rv = apr_brigade_flatten(ib, iobuf, &writebuflen); + + apr_brigade_cleanup(ib); + + if (rv != APR_SUCCESS) { + *err = "flattening brigade"; + break; + } + + to_send = writebuflen; + iobuf_cursor = iobuf; + while (to_send > 0) { + int nvec = 0; + apr_size_t write_this_time; + + write_this_time = + to_send < AP_FCGI_MAX_CONTENT_LEN ? to_send : AP_FCGI_MAX_CONTENT_LEN; + + ap_fcgi_fill_in_header(&header, AP_FCGI_STDIN, request_id, + (apr_uint16_t)write_this_time, 0); + ap_fcgi_header_to_array(&header, farray); + + vec[nvec].iov_base = (void *)farray; + vec[nvec].iov_len = sizeof(farray); + ++nvec; + if (writebuflen) { + vec[nvec].iov_base = iobuf_cursor; + vec[nvec].iov_len = write_this_time; + ++nvec; + } + + rv = send_data(conn, vec, nvec, &len); + if (rv != APR_SUCCESS) { + *err = "sending stdin"; + break; + } + + to_send -= write_this_time; + iobuf_cursor += write_this_time; + } + if (rv != APR_SUCCESS) { + break; + } + + if (last_stdin) { + pfd.reqevents = APR_POLLIN; /* Done with input data */ + + /* signal EOF (empty FCGI_STDIN) */ + ap_fcgi_fill_in_header(&header, AP_FCGI_STDIN, request_id, + 0, 0); + ap_fcgi_header_to_array(&header, farray); + + vec[0].iov_base = (void *)farray; + vec[0].iov_len = sizeof(farray); + + rv = send_data(conn, vec, 1, &len); + if (rv != APR_SUCCESS) { + *err = "sending empty stdin"; + break; + } + } + } + + if (pfd.rtnevents & APR_POLLIN) { + apr_size_t readbuflen; + apr_uint16_t clen, rid; + apr_bucket *b; + unsigned char plen; + unsigned char type, version; + int mayflush = 0; + + /* First, we grab the header... */ + rv = get_data_full(conn, (char *) farray, AP_FCGI_HEADER_LEN); + if (rv != APR_SUCCESS) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01067) + "Failed to read FastCGI header"); + break; + } + + ap_log_rdata(APLOG_MARK, APLOG_TRACE8, r, "FastCGI header", + farray, AP_FCGI_HEADER_LEN, 0); + + ap_fcgi_header_fields_from_array(&version, &type, &rid, + &clen, &plen, farray); + + if (version != AP_FCGI_VERSION_1) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01068) + "Got bogus version %d", (int)version); + rv = APR_EINVAL; + break; + } + + if (rid != request_id) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01069) + "Got bogus rid %d, expected %d", + rid, request_id); + rv = APR_EINVAL; + break; + } + +recv_again: + if (clen > iobuf_size) { + readbuflen = iobuf_size; + } else { + readbuflen = clen; + } + + /* Now get the actual data. Yes it sucks to do this in a second + * recv call, this will eventually change when we move to real + * nonblocking recv calls. */ + if (readbuflen != 0) { + rv = get_data(conn, iobuf, &readbuflen); + if (rv != APR_SUCCESS) { + *err = "reading response body"; + break; + } + } + + switch (type) { + case AP_FCGI_STDOUT: + if (clen != 0) { + b = apr_bucket_transient_create(iobuf, + readbuflen, + c->bucket_alloc); + + APR_BRIGADE_INSERT_TAIL(ob, b); + + if (! seen_end_of_headers) { + int st = handle_headers(r, &header_state, + iobuf, readbuflen); + + if (st == 1) { + int status; + seen_end_of_headers = 1; + + status = ap_scan_script_header_err_brigade_ex(r, ob, + NULL, APLOG_MODULE_INDEX); + /* suck in all the rest */ + if (status != OK) { + apr_bucket *tmp_b; + apr_brigade_cleanup(ob); + tmp_b = apr_bucket_eos_create(c->bucket_alloc); + APR_BRIGADE_INSERT_TAIL(ob, tmp_b); + + *has_responded = 1; + r->status = status; + rv = ap_pass_brigade(r->output_filters, ob); + if (rv != APR_SUCCESS) { + *err = "passing headers brigade to output filters"; + break; + } + else if (status == HTTP_NOT_MODIFIED + || status == HTTP_PRECONDITION_FAILED) { + /* Special 'status' cases handled: + * 1) HTTP 304 response MUST NOT contain + * a message-body, ignore it. + * 2) HTTP 412 response. + * The break is not added since there might + * be more bytes to read from the FCGI + * connection. Even if the message-body is + * ignored (and the EOS bucket has already + * been sent) we want to avoid subsequent + * bogus reads. */ + ignore_body = 1; + } + else { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01070) + "Error parsing script headers"); + rv = APR_EINVAL; + break; + } + } + + if (ap_proxy_should_override(conf, r->status) && ap_is_initial_req(r)) { + /* + * set script_error_status to discard + * everything after the headers + */ + script_error_status = r->status; + /* + * prevent ap_die() from treating this as a + * recursive error, initially: + */ + r->status = HTTP_OK; + } + + if (script_error_status == HTTP_OK + && !APR_BRIGADE_EMPTY(ob) && !ignore_body) { + /* Send the part of the body that we read while + * reading the headers. + */ + *has_responded = 1; + rv = ap_pass_brigade(r->output_filters, ob); + if (rv != APR_SUCCESS) { + *err = "passing brigade to output filters"; + break; + } + mayflush = 1; + } + apr_brigade_cleanup(ob); + + apr_pool_clear(setaside_pool); + } + else { + /* We're still looking for the end of the + * headers, so this part of the data will need + * to persist. */ + apr_bucket_setaside(b, setaside_pool); + } + } else { + /* we've already passed along the headers, so now pass + * through the content. we could simply continue to + * setaside the content and not pass until we see the + * 0 content-length (below, where we append the EOS), + * but that could be a huge amount of data; so we pass + * along smaller chunks + */ + if (script_error_status == HTTP_OK && !ignore_body) { + *has_responded = 1; + rv = ap_pass_brigade(r->output_filters, ob); + if (rv != APR_SUCCESS) { + *err = "passing brigade to output filters"; + break; + } + mayflush = 1; + } + apr_brigade_cleanup(ob); + } + + /* If we didn't read all the data, go back and get the + * rest of it. */ + if (clen > readbuflen) { + clen -= readbuflen; + goto recv_again; + } + } else { + /* XXX what if we haven't seen end of the headers yet? */ + + if (script_error_status == HTTP_OK) { + b = apr_bucket_eos_create(c->bucket_alloc); + APR_BRIGADE_INSERT_TAIL(ob, b); + + *has_responded = 1; + rv = ap_pass_brigade(r->output_filters, ob); + if (rv != APR_SUCCESS) { + *err = "passing brigade to output filters"; + break; + } + } + + /* XXX Why don't we cleanup here? (logic from AJP) */ + } + break; + + case AP_FCGI_STDERR: + /* TODO: Should probably clean up this logging a bit... */ + if (clen) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01071) + "Got error '%.*s'", (int)readbuflen, iobuf); + } + + if (clen > readbuflen) { + clen -= readbuflen; + goto recv_again; + } + break; + + case AP_FCGI_END_REQUEST: + done = 1; + break; + + default: + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01072) + "Got bogus record %d", type); + break; + } + /* Leave on above switch's inner error. */ + if (rv != APR_SUCCESS) { + break; + } + + if (plen) { + rv = get_data_full(conn, iobuf, plen); + if (rv != APR_SUCCESS) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, rv, r, APLOGNO(02537) + "Error occurred reading padding"); + break; + } + } + + if (mayflush && ((conn->worker->s->flush_packets == flush_on) || + ((conn->worker->s->flush_packets == flush_auto) && + (apr_poll(flushpoll, 1, &flushpoll_fd, + conn->worker->s->flush_wait) == APR_TIMEUP)))) { + apr_bucket* flush_b = apr_bucket_flush_create(r->connection->bucket_alloc); + APR_BRIGADE_INSERT_TAIL(ob, flush_b); + rv = ap_pass_brigade(r->output_filters, ob); + if (rv != APR_SUCCESS) { + *err = "passing headers brigade to output filters"; + break; + } + mayflush = 0; + } + } + } + + apr_brigade_destroy(ib); + apr_brigade_destroy(ob); + + if (script_error_status != HTTP_OK) { + ap_die(script_error_status, r); /* send ErrorDocument */ + *has_responded = 1; + } + + return rv; +} + +/* + * process the request and write the response. + */ +static int fcgi_do_request(apr_pool_t *p, request_rec *r, + proxy_conn_rec *conn, + conn_rec *origin, + proxy_dir_conf *conf, + apr_uri_t *uri, + char *url, char *server_portstr, + apr_bucket_brigade *input_brigade) +{ + /* Request IDs are arbitrary numbers that we assign to a + * single request. This would allow multiplex/pipelining of + * multiple requests to the same FastCGI connection, but + * we don't support that, and always use a value of '1' to + * keep things simple. */ + apr_uint16_t request_id = 1; + apr_status_t rv; + apr_pool_t *temp_pool; + const char *err; + int bad_request = 0, + has_responded = 0; + + /* Step 1: Send AP_FCGI_BEGIN_REQUEST */ + rv = send_begin_request(conn, request_id); + if (rv != APR_SUCCESS) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, rv, r, APLOGNO(01073) + "Failed Writing Request to %s:", server_portstr); + conn->close = 1; + return HTTP_SERVICE_UNAVAILABLE; + } + + apr_pool_create(&temp_pool, r->pool); + apr_pool_tag(temp_pool, "proxy_fcgi_do_request"); + + /* Step 2: Send Environment via FCGI_PARAMS */ + rv = send_environment(conn, r, temp_pool, request_id); + if (rv != APR_SUCCESS) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, rv, r, APLOGNO(01074) + "Failed writing Environment to %s:", server_portstr); + conn->close = 1; + return HTTP_SERVICE_UNAVAILABLE; + } + + /* Step 3: Read records from the back end server and handle them. */ + rv = dispatch(conn, conf, r, temp_pool, request_id, + &err, &bad_request, &has_responded, + input_brigade); + if (rv != APR_SUCCESS) { + /* If the client aborted the connection during retrieval or (partially) + * sending the response, don't return a HTTP_SERVICE_UNAVAILABLE, since + * this is not a backend problem. */ + if (r->connection->aborted) { + ap_log_rerror(APLOG_MARK, APLOG_TRACE1, rv, r, + "The client aborted the connection."); + conn->close = 1; + return OK; + } + + ap_log_rerror(APLOG_MARK, APLOG_ERR, rv, r, APLOGNO(01075) + "Error dispatching request to %s: %s%s%s", + server_portstr, + err ? "(" : "", + err ? err : "", + err ? ")" : ""); + conn->close = 1; + if (has_responded) { + return AP_FILTER_ERROR; + } + if (bad_request) { + return ap_map_http_request_error(rv, HTTP_BAD_REQUEST); + } + if (APR_STATUS_IS_TIMEUP(rv)) { + return HTTP_GATEWAY_TIME_OUT; + } + return HTTP_SERVICE_UNAVAILABLE; + } + + return OK; +} + +#define FCGI_SCHEME "FCGI" + +#define MAX_MEM_SPOOL 16384 + +/* + * This handles fcgi:(dest) URLs + */ +static int proxy_fcgi_handler(request_rec *r, proxy_worker *worker, + proxy_server_conf *conf, + char *url, const char *proxyname, + apr_port_t proxyport) +{ + int status; + char server_portstr[32]; + conn_rec *origin = NULL; + proxy_conn_rec *backend = NULL; + apr_bucket_brigade *input_brigade; + apr_off_t input_bytes = 0; + apr_uri_t *uri; + + proxy_dir_conf *dconf = ap_get_module_config(r->per_dir_config, + &proxy_module); + + apr_pool_t *p = r->pool; + + + ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01076) + "url: %s proxyname: %s proxyport: %d", + url, proxyname, proxyport); + + if (ap_cstr_casecmpn(url, "fcgi:", 5) != 0) { + ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01077) "declining URL %s", url); + return DECLINED; + } + + ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01078) "serving URL %s", url); + + /* Create space for state information */ + status = ap_proxy_acquire_connection(FCGI_SCHEME, &backend, worker, + r->server); + if (status != OK) { + if (backend) { + backend->close = 1; + ap_proxy_release_connection(FCGI_SCHEME, backend, r->server); + } + return status; + } + + backend->is_ssl = 0; + + /* Step One: Determine Who To Connect To */ + uri = apr_palloc(p, sizeof(*uri)); + status = ap_proxy_determine_connection(p, r, conf, worker, backend, + uri, &url, proxyname, proxyport, + server_portstr, + sizeof(server_portstr)); + if (status != OK) { + goto cleanup; + } + + /* We possibly reuse input data prefetched in previous call(s), e.g. for a + * balancer fallback scenario. + */ + apr_pool_userdata_get((void **)&input_brigade, "proxy-fcgi-input", p); + if (input_brigade == NULL) { + const char *old_te = apr_table_get(r->headers_in, "Transfer-Encoding"); + const char *old_cl = NULL; + if (old_te) { + apr_table_unset(r->headers_in, "Content-Length"); + } + else { + old_cl = apr_table_get(r->headers_in, "Content-Length"); + } + + input_brigade = apr_brigade_create(p, r->connection->bucket_alloc); + apr_pool_userdata_setn(input_brigade, "proxy-fcgi-input", NULL, p); + + /* Prefetch (nonlocking) the request body so to increase the chance + * to get the whole (or enough) body and determine Content-Length vs + * chunked or spooled. By doing this before connecting or reusing the + * backend, we want to minimize the delay between this connection is + * considered alive and the first bytes sent (should the client's link + * be slow or some input filter retain the data). This is a best effort + * to prevent the backend from closing (from under us) what it thinks is + * an idle connection, hence to reduce to the minimum the unavoidable + * local is_socket_connected() vs remote keepalive race condition. + */ + status = ap_proxy_prefetch_input(r, backend, input_brigade, + APR_NONBLOCK_READ, &input_bytes, + MAX_MEM_SPOOL); + if (status != OK) { + goto cleanup; + } + + /* + * The request body is streamed by default, using either C-L or + * chunked T-E, like this: + * + * The whole body (including no body) was received on prefetch, i.e. + * the input brigade ends with EOS => C-L = input_bytes. + * + * C-L is known and reliable, i.e. only protocol filters in the input + * chain thus none should change the body => use C-L from client. + * + * The administrator has not "proxy-sendcl" which prevents T-E => use + * T-E and chunks. + * + * Otherwise we need to determine and set a content-length, so spool + * the entire request body to memory/temporary file (MAX_MEM_SPOOL), + * such that we finally know its length => C-L = input_bytes. + */ + if (!APR_BRIGADE_EMPTY(input_brigade) + && APR_BUCKET_IS_EOS(APR_BRIGADE_LAST(input_brigade))) { + /* The whole thing fit, so our decision is trivial, use the input + * bytes for the Content-Length. If we expected no body, and read + * no body, do not set the Content-Length. + */ + if (old_cl || old_te || input_bytes) { + apr_table_setn(r->headers_in, "Content-Length", + apr_off_t_toa(p, input_bytes)); + if (old_te) { + apr_table_unset(r->headers_in, "Transfer-Encoding"); + } + } + } + else if (old_cl && r->input_filters == r->proto_input_filters) { + /* Streaming is possible by preserving the existing C-L */ + } + else if (!apr_table_get(r->subprocess_env, "proxy-sendcl")) { + /* Streaming is possible using T-E: chunked */ + } + else { + /* No streaming, C-L is the only option so spool to memory/file */ + apr_bucket_brigade *tmp_bb; + apr_off_t remaining_bytes = 0; + + AP_DEBUG_ASSERT(MAX_MEM_SPOOL >= input_bytes); + tmp_bb = apr_brigade_create(p, r->connection->bucket_alloc); + status = ap_proxy_spool_input(r, backend, tmp_bb, &remaining_bytes, + MAX_MEM_SPOOL - input_bytes); + if (status != OK) { + goto cleanup; + } + + APR_BRIGADE_CONCAT(input_brigade, tmp_bb); + input_bytes += remaining_bytes; + + apr_table_setn(r->headers_in, "Content-Length", + apr_off_t_toa(p, input_bytes)); + if (old_te) { + apr_table_unset(r->headers_in, "Transfer-Encoding"); + } + } + } + + /* This scheme handler does not reuse connections by default, to + * avoid tying up a fastcgi that isn't expecting to work on + * parallel requests. But if the user went out of their way to + * type the default value of disablereuse=off, we'll allow it. + */ + backend->close = 1; + if (worker->s->disablereuse_set && !worker->s->disablereuse) { + backend->close = 0; + } + + /* Step Two: Make the Connection */ + if (ap_proxy_check_connection(FCGI_SCHEME, backend, r->server, 0, + PROXY_CHECK_CONN_EMPTY) + && ap_proxy_connect_backend(FCGI_SCHEME, backend, worker, + r->server)) { + ap_log_rerror(APLOG_MARK, APLOG_ERR, 0, r, APLOGNO(01079) + "failed to make connection to backend: %s", + backend->hostname); + status = HTTP_SERVICE_UNAVAILABLE; + goto cleanup; + } + + /* Step Three: Process the Request */ + status = fcgi_do_request(p, r, backend, origin, dconf, uri, url, + server_portstr, input_brigade); + +cleanup: + ap_proxy_release_connection(FCGI_SCHEME, backend, r->server); + return status; +} + +static void *fcgi_create_dconf(apr_pool_t *p, char *path) +{ + fcgi_dirconf_t *a; + + a = (fcgi_dirconf_t *)apr_pcalloc(p, sizeof(fcgi_dirconf_t)); + a->backend_type = BACKEND_DEFAULT_UNKNOWN; + a->env_fixups = apr_array_make(p, 20, sizeof(sei_entry)); + + return a; +} + +static void *fcgi_merge_dconf(apr_pool_t *p, void *basev, void *overridesv) +{ + fcgi_dirconf_t *a, *base, *over; + + a = (fcgi_dirconf_t *)apr_pcalloc(p, sizeof(fcgi_dirconf_t)); + base = (fcgi_dirconf_t *)basev; + over = (fcgi_dirconf_t *)overridesv; + + a->backend_type = (over->backend_type != BACKEND_DEFAULT_UNKNOWN) + ? over->backend_type + : base->backend_type; + a->env_fixups = apr_array_append(p, base->env_fixups, over->env_fixups); + return a; +} + +static const char *cmd_servertype(cmd_parms *cmd, void *in_dconf, + const char *val) +{ + fcgi_dirconf_t *dconf = in_dconf; + + if (!strcasecmp(val, "GENERIC")) { + dconf->backend_type = BACKEND_GENERIC; + } + else if (!strcasecmp(val, "FPM")) { + dconf->backend_type = BACKEND_FPM; + } + else { + return "ProxyFCGIBackendType requires one of the following arguments: " + "'GENERIC', 'FPM'"; + } + + return NULL; +} + + +static const char *cmd_setenv(cmd_parms *cmd, void *in_dconf, + const char *arg1, const char *arg2, + const char *arg3) +{ + fcgi_dirconf_t *dconf = in_dconf; + const char *err; + sei_entry *new; + const char *envvar = arg2; + + new = apr_array_push(dconf->env_fixups); + new->cond = ap_expr_parse_cmd(cmd, arg1, 0, &err, NULL); + if (err) { + return apr_psprintf(cmd->pool, "Could not parse expression \"%s\": %s", + arg1, err); + } + + if (envvar[0] == '!') { + /* Unset mode. */ + if (arg3) { + return apr_psprintf(cmd->pool, "Third argument (\"%s\") is not " + "allowed when using ProxyFCGISetEnvIf's unset " + "mode (%s)", arg3, envvar); + } + else if (!envvar[1]) { + /* i.e. someone tried to give us a name of just "!" */ + return "ProxyFCGISetEnvIf: \"!\" is not a valid variable name"; + } + + new->subst = NULL; + } + else { + /* Set mode. */ + if (!arg3) { + /* A missing expr-value should be treated as empty. */ + arg3 = ""; + } + + new->subst = ap_expr_parse_cmd(cmd, arg3, AP_EXPR_FLAG_STRING_RESULT, &err, NULL); + if (err) { + return apr_psprintf(cmd->pool, "Could not parse expression \"%s\": %s", + arg3, err); + } + } + + new->envname = envvar; + + return NULL; +} +static void register_hooks(apr_pool_t *p) +{ + proxy_hook_scheme_handler(proxy_fcgi_handler, NULL, NULL, APR_HOOK_FIRST); + proxy_hook_canon_handler(proxy_fcgi_canon, NULL, NULL, APR_HOOK_FIRST); +} + +static const command_rec command_table[] = { + AP_INIT_TAKE1("ProxyFCGIBackendType", cmd_servertype, NULL, OR_FILEINFO, + "Specify the type of FastCGI server: 'Generic', 'FPM'"), + AP_INIT_TAKE23("ProxyFCGISetEnvIf", cmd_setenv, NULL, OR_FILEINFO, + "expr-condition env-name expr-value"), + { NULL } +}; + +AP_DECLARE_MODULE(proxy_fcgi) = { + STANDARD20_MODULE_STUFF, + fcgi_create_dconf, /* create per-directory config structure */ + fcgi_merge_dconf, /* merge per-directory config structures */ + NULL, /* create per-server config structure */ + NULL, /* merge per-server config structures */ + command_table, /* command apr_table_t */ + register_hooks /* register hooks */ +}; |