/** Clear the GeoIP database and reload it from the file
* <b>filename</b>. Return 0 on success, -1 on failure.
*
* Recognized line formats are:
* INTIPLOW,INTIPHIGH,CC
* and
* "INTIPLOW","INTIPHIGH","CC","CC3","COUNTRY NAME"
* where INTIPLOW and INTIPHIGH are IPv4 addresses encoded as 4-byte unsigned
* integers, and CC is a country code.
*
* It also recognizes, and skips over, blank lines and lines that start
* with '#' (comments).
*/
int
geoip_load_file(const char *filename, const or_options_t *options)
{
FILE *f;
const char *msg = "";
int severity = options_need_geoip_info(options, &msg) ? LOG_WARN : LOG_INFO;
crypto_digest_t *geoip_digest_env = NULL;
clear_geoip_db();
if (!(f = tor_fopen_cloexec(filename, "r"))) {
log_fn(severity, LD_GENERAL, "Failed to open GEOIP file %s. %s",
filename, msg);
return -1;
}
if (!geoip_countries)
init_geoip_countries();
if (geoip_entries) {
SMARTLIST_FOREACH(geoip_entries, geoip_entry_t *, e, tor_free(e));
smartlist_free(geoip_entries);
}
geoip_entries = smartlist_new();
geoip_digest_env = crypto_digest_new();
log_notice(LD_GENERAL, "Parsing GEOIP file %s.", filename);
while (!feof(f)) {
char buf[512];
if (fgets(buf, (int)sizeof(buf), f) == NULL)
break;
crypto_digest_add_bytes(geoip_digest_env, buf, strlen(buf));
/* FFFF track full country name. */
geoip_parse_entry(buf);
}
/*XXXX abort and return -1 if no entries/illformed?*/
fclose(f);
smartlist_sort(geoip_entries, geoip_compare_entries_);
/* Okay, now we need to maybe change our mind about what is in which
* country. */
refresh_all_country_info();
/* Remember file digest so that we can include it in our extra-info
* descriptors. */
crypto_digest_get_digest(geoip_digest_env, geoip_digest, DIGEST_LEN);
crypto_digest_free(geoip_digest_env);
return 0;
}
/** Clear appropriate GeoIP database, based on <b>family</b>, and
* reload it from the file <b>filename</b>. Return 0 on success, -1 on
* failure.
*
* Recognized line formats for IPv4 are:
* INTIPLOW,INTIPHIGH,CC
* and
* "INTIPLOW","INTIPHIGH","CC","CC3","COUNTRY NAME"
* where INTIPLOW and INTIPHIGH are IPv4 addresses encoded as 4-byte unsigned
* integers, and CC is a country code.
*
* Recognized line format for IPv6 is:
* IPV6LOW,IPV6HIGH,CC
* where IPV6LOW and IPV6HIGH are IPv6 addresses and CC is a country code.
*
* It also recognizes, and skips over, blank lines and lines that start
* with '#' (comments).
*/
int
geoip_load_file(sa_family_t family, const char *filename)
{
FILE *f;
const char *msg = "";
const or_options_t *options = get_options();
int severity = options_need_geoip_info(options, &msg) ? LOG_WARN : LOG_INFO;
crypto_digest_t *geoip_digest_env = NULL;
tor_assert(family == AF_INET || family == AF_INET6);
if (!(f = tor_fopen_cloexec(filename, "r"))) {
log_fn(severity, LD_GENERAL, "Failed to open GEOIP file %s. %s",
filename, msg);
return -1;
}
if (!geoip_countries)
init_geoip_countries();
if (family == AF_INET) {
if (geoip_ipv4_entries) {
SMARTLIST_FOREACH(geoip_ipv4_entries, geoip_ipv4_entry_t *, e,
tor_free(e));
smartlist_free(geoip_ipv4_entries);
}
geoip_ipv4_entries = smartlist_new();
} else { /* AF_INET6 */
if (geoip_ipv6_entries) {
SMARTLIST_FOREACH(geoip_ipv6_entries, geoip_ipv6_entry_t *, e,
tor_free(e));
smartlist_free(geoip_ipv6_entries);
}
geoip_ipv6_entries = smartlist_new();
}
geoip_digest_env = crypto_digest_new();
log_notice(LD_GENERAL, "Parsing GEOIP %s file %s.",
(family == AF_INET) ? "IPv4" : "IPv6", filename);
while (!feof(f)) {
char buf[512];
if (fgets(buf, (int)sizeof(buf), f) == NULL)
break;
crypto_digest_add_bytes(geoip_digest_env, buf, strlen(buf));
/* FFFF track full country name. */
geoip_parse_entry(buf, family);
}
/*XXXX abort and return -1 if no entries/illformed?*/
fclose(f);
/* Sort list and remember file digests so that we can include it in
* our extra-info descriptors. */
if (family == AF_INET) {
smartlist_sort(geoip_ipv4_entries, geoip_ipv4_compare_entries_);
/* Okay, now we need to maybe change our mind about what is in
* which country. We do this for IPv4 only since that's what we
* store in node->country. */
refresh_all_country_info();
crypto_digest_get_digest(geoip_digest_env, geoip_digest, DIGEST_LEN);
} else {
/* AF_INET6 */
smartlist_sort(geoip_ipv6_entries, geoip_ipv6_compare_entries_);
crypto_digest_get_digest(geoip_digest_env, geoip6_digest, DIGEST_LEN);
}
crypto_digest_free(geoip_digest_env);
return 0;
}
/**
* Read the measured bandwidth list <b>from_file</b>:
* - store all the headers in <b>bw_file_headers</b>,
* - apply bandwidth lines to the list of vote_routerstatus_t in
* <b>routerstatuses</b>,
* - cache bandwidth lines for dirserv_get_bandwidth_for_router(),
* - expire old entries in the measured bandwidth cache, and
* - store the DIGEST_SHA256 of the contents of the file in <b>digest_out</b>.
*
* Returns -1 on error, 0 otherwise.
*
* If the file can't be read, or is empty:
* - <b>bw_file_headers</b> is empty,
* - <b>routerstatuses</b> is not modified,
* - the measured bandwidth cache is not modified, and
* - <b>digest_out</b> is the zero-byte digest.
*
* Otherwise, if there is an error later in the file:
* - <b>bw_file_headers</b> contains all the headers up to the error,
* - <b>routerstatuses</b> is updated with all the relay lines up to the error,
* - the measured bandwidth cache is updated with all the relay lines up to
* the error,
* - if the timestamp is valid and recent, old entries in the measured
* bandwidth cache are expired, and
* - <b>digest_out</b> is the digest up to the first read error (if any).
* The digest is taken over all the readable file contents, even if the
* file is outdated or unparseable.
*/
int
dirserv_read_measured_bandwidths(const char *from_file,
smartlist_t *routerstatuses,
smartlist_t *bw_file_headers,
uint8_t *digest_out)
{
FILE *fp = tor_fopen_cloexec(from_file, "r");
int applied_lines = 0;
time_t file_time, now;
int ok;
/* This flag will be 1 only when the first successful bw measurement line
* has been encountered, so that measured_bw_line_parse don't give warnings
* if there are additional header lines, as introduced in Bandwidth List spec
* version 1.1.0 */
int line_is_after_headers = 0;
int rv = -1;
char *line = NULL;
size_t n = 0;
crypto_digest_t *digest = crypto_digest256_new(DIGEST_SHA256);
if (fp == NULL) {
log_warn(LD_CONFIG, "Can't open bandwidth file at configured location: %s",
from_file);
goto err;
}
if (tor_getline(&line,&n,fp) <= 0) {
log_warn(LD_DIRSERV, "Empty bandwidth file");
goto err;
}
/* If the line could be gotten, add it to the digest */
crypto_digest_add_bytes(digest, (const char *) line, strlen(line));
if (!strlen(line) || line[strlen(line)-1] != '\n') {
log_warn(LD_DIRSERV, "Long or truncated time in bandwidth file: %s",
escaped(line));
/* Continue adding lines to the digest. */
goto continue_digest;
}
line[strlen(line)-1] = '\0';
file_time = (time_t)tor_parse_ulong(line, 10, 0, ULONG_MAX, &ok, NULL);
if (!ok) {
log_warn(LD_DIRSERV, "Non-integer time in bandwidth file: %s",
escaped(line));
goto continue_digest;
}
now = approx_time();
if ((now - file_time) > MAX_MEASUREMENT_AGE) {
log_warn(LD_DIRSERV, "Bandwidth measurement file stale. Age: %u",
(unsigned)(time(NULL) - file_time));
goto continue_digest;
}
/* If timestamp was correct and bw_file_headers is not NULL,
* add timestamp to bw_file_headers */
if (bw_file_headers)
smartlist_add_asprintf(bw_file_headers, "timestamp=%lu",
(unsigned long)file_time);
if (routerstatuses)
smartlist_sort(routerstatuses, compare_vote_routerstatus_entries);
while (!feof(fp)) {
measured_bw_line_t parsed_line;
if (tor_getline(&line, &n, fp) >= 0) {
crypto_digest_add_bytes(digest, (const char *) line, strlen(line));
if (measured_bw_line_parse(&parsed_line, line,
line_is_after_headers) != -1) {
/* This condition will be true when the first complete valid bw line
* has been encountered, which means the end of the header lines. */
line_is_after_headers = 1;
/* Also cache the line for dirserv_get_bandwidth_for_router() */
dirserv_cache_measured_bw(&parsed_line, file_time);
if (measured_bw_line_apply(&parsed_line, routerstatuses) > 0)
applied_lines++;
/* if the terminator is found, it is the end of header lines, set the
* flag but do not store anything */
} else if (strcmp(line, BW_FILE_HEADERS_TERMINATOR) == 0) {
line_is_after_headers = 1;
/* if the line was not a correct relay line nor the terminator and
* the end of the header lines has not been detected yet
* and it is key_value and bw_file_headers did not reach the maximum
* number of headers,
* then assume this line is a header and add it to bw_file_headers */
} else if (bw_file_headers &&
(line_is_after_headers == 0) &&
string_is_key_value(LOG_DEBUG, line) &&
!strchr(line, ' ') &&
(smartlist_len(bw_file_headers)
< MAX_BW_FILE_HEADER_COUNT_IN_VOTE)) {
line[strlen(line)-1] = '\0';
smartlist_add_strdup(bw_file_headers, line);
};
}
}
/* Now would be a nice time to clean the cache, too */
dirserv_expire_measured_bw_cache(now);
log_info(LD_DIRSERV,
"Bandwidth measurement file successfully read. "
"Applied %d measurements.", applied_lines);
rv = 0;
continue_digest:
/* Continue parsing lines to return the digest of the Bandwidth File. */
while (!feof(fp)) {
if (tor_getline(&line, &n, fp) >= 0) {
crypto_digest_add_bytes(digest, (const char *) line, strlen(line));
}
}
err:
if (line) {
// we need to raw_free this buffer because we got it from tor_getdelim()
raw_free(line);
}
if (fp)
fclose(fp);
if (digest_out)
crypto_digest_get_digest(digest, (char *) digest_out, DIGEST256_LEN);
crypto_digest_free(digest);
return rv;
}
