Merge pull request #380 from bshifter/logmsg-check-program

check program name for rfc3164
axoflow · Nov 18, 2024 · e152a6c · e152a6c
2 parents 0b99654 + 6fbb002
commit e152a6c
Show file tree

Hide file tree

Showing 13 changed files with 151 additions and 0 deletions.
diff --git a/lib/cfg-grammar.y b/lib/cfg-grammar.y
@@ -275,6 +275,7 @@ main_location_print (FILE *yyo, YYLTYPE const * const yylocp)
 %token KW_BAD_HOSTNAME                10094
 %token KW_LOG_LEVEL                   10095
 %token KW_IDLE_TIMEOUT                10096
+%token KW_CHECK_PROGRAM               10097
 
 %token KW_KEEP_TIMESTAMP              10100
 
@@ -1102,6 +1103,7 @@ options_item
 	| KW_CHAIN_HOSTNAMES '(' yesno ')'	{ configuration->chain_hostnames = $3; }
 	| KW_KEEP_HOSTNAME '(' yesno ')'	{ configuration->keep_hostname = $3; }
 	| KW_CHECK_HOSTNAME '(' yesno ')'	{ configuration->check_hostname = $3; }
+  | KW_CHECK_PROGRAM '(' yesno ')' { configuration->check_program = $3; }
 	| KW_BAD_HOSTNAME '(' string ')'	{ cfg_bad_hostname_set(configuration, $3); free($3); }
 	| KW_TIME_REOPEN '(' positive_integer ')'		{ configuration->time_reopen = $3; }
 	| KW_TIME_REAP '(' nonnegative_integer ')'		{ configuration->time_reap = $3; }
@@ -1485,6 +1487,7 @@ source_reader_option
         /* NOTE: plugins need to set "last_reader_options" in order to incorporate this rule in their grammar */
 
 	: KW_CHECK_HOSTNAME '(' yesno ')'	{ last_reader_options->check_hostname = $3; }
+  | KW_CHECK_PROGRAM '(' yesno ')' { last_reader_options->check_program = $3; }
 	| KW_FLAGS '(' source_reader_option_flags ')'
 	| KW_LOG_FETCH_LIMIT '(' positive_integer ')'	{ last_reader_options->fetch_limit = $3; }
         | KW_FORMAT '(' string ')'              { last_reader_options->parse_options.format = g_strdup($3); free($3); }
@@ -1496,6 +1499,7 @@ source_reader_option
 source_reader_option_flags
         : string source_reader_option_flags     { CHECK_ERROR(log_reader_options_process_flag(last_reader_options, $1), @1, "Unknown flag \"%s\"", $1); free($1); }
         | KW_CHECK_HOSTNAME source_reader_option_flags     { log_reader_options_process_flag(last_reader_options, "check-hostname"); }
+        | KW_CHECK_PROGRAM source_reader_option_flags     { log_reader_options_process_flag(last_reader_options, "check-program"); }
 	|
 	;
 

diff --git a/lib/cfg-parser.c b/lib/cfg-parser.c
@@ -116,6 +116,7 @@ static CfgLexerKeyword main_keywords[] =
   { "normalize_hostnames", KW_NORMALIZE_HOSTNAMES },
   { "keep_hostname",      KW_KEEP_HOSTNAME },
   { "check_hostname",     KW_CHECK_HOSTNAME },
+  { "check_program",      KW_CHECK_PROGRAM },
   { "bad_hostname",       KW_BAD_HOSTNAME },
   { "custom_domain",      KW_CUSTOM_DOMAIN },
   { "keep_timestamp",     KW_KEEP_TIMESTAMP },

diff --git a/lib/cfg.h b/lib/cfg.h
@@ -82,6 +82,7 @@ struct _GlobalConfig
   gboolean chain_hostnames;
   gboolean keep_hostname;
   gboolean check_hostname;
+  gboolean check_program;
   gboolean bad_hostname_compiled;
   regex_t bad_hostname;
   gchar *bad_hostname_re;

diff --git a/lib/logmsg/logmsg.c b/lib/logmsg/logmsg.c
@@ -2030,6 +2030,7 @@ log_msg_tags_init(void)
   log_tags_register_predefined_tag("syslog.invalid_hostname", LM_T_SYSLOG_INVALID_HOSTNAME);
   log_tags_register_predefined_tag("syslog.unexpected_framing", LM_T_SYSLOG_UNEXPECTED_FRAMING);
   log_tags_register_predefined_tag("syslog.rfc3164_missing_header", LM_T_SYSLOG_RFC3164_MISSING_HEADER);
+  log_tags_register_predefined_tag("syslog.rfc3164_invalid_program", LM_T_SYSLOG_RFC_3164_INVALID_PROGRAM);
 
   log_tags_register_predefined_tag("syslog.rfc5424_missing_hostname", LM_T_SYSLOG_RFC5424_MISSING_HOSTNAME);
   log_tags_register_predefined_tag("syslog.rfc5424_missing_app_name", LM_T_SYSLOG_RFC5424_MISSING_APP_NAME);

diff --git a/lib/logmsg/logmsg.h b/lib/logmsg/logmsg.h
@@ -134,6 +134,9 @@ enum
   LM_T_SYSLOG_RFC5424_MISSING_MESSAGE,
   /* message field missing */
   LM_T_SYSLOG_MISSING_MESSAGE,
+  /* invalid program name */
+  LM_T_SYSLOG_RFC_3164_INVALID_PROGRAM,
+
   LM_T_PREDEFINED_MAX,
 };
 

diff --git a/lib/logreader.c b/lib/logreader.c
@@ -852,6 +852,11 @@ log_reader_options_init(LogReaderOptions *options, GlobalConfig *cfg, const gcha
     options->parse_options.flags |= LP_ASSUME_UTF8;
   if (cfg->threaded)
     options->flags |= LR_THREADED;
+  if (options->check_program == -1)
+    options->check_program = cfg->check_program;
+  if (options->check_program)
+    options->parse_options.flags |= LP_CHECK_PROGRAM;
+
   options->initialized = TRUE;
 }
 

diff --git a/lib/logreader.h b/lib/logreader.h
@@ -53,6 +53,7 @@ typedef struct _LogReaderOptions
   gint fetch_limit;
   const gchar *group_name;
   gboolean check_hostname;
+  gboolean check_program;
 } LogReaderOptions;
 
 typedef struct _LogReader LogReader;

diff --git a/lib/msg-format.c b/lib/msg-format.c
@@ -333,6 +333,7 @@ CfgFlagHandler msg_format_flag_handlers[] =
   { "no-rfc3164-fallback",        CFH_SET, offsetof(MsgFormatOptions, flags), LP_NO_RFC3164_FALLBACK },
   { "piggyback-errors",           CFH_SET, offsetof(MsgFormatOptions, flags), LP_PIGGYBACK_ERRORS },
   { "no-piggyback-errors",      CFH_CLEAR, offsetof(MsgFormatOptions, flags), LP_PIGGYBACK_ERRORS },
+  { "check-program",              CFH_SET, offsetof(MsgFormatOptions, flags), LP_CHECK_PROGRAM },
   { NULL },
 };
 

diff --git a/lib/msg-format.h b/lib/msg-format.h
@@ -60,6 +60,7 @@ enum
   LP_NO_HEADER = 0x2000,
   LP_NO_RFC3164_FALLBACK = 0x4000,
   LP_PIGGYBACK_ERRORS = 0x8000,
+  LP_CHECK_PROGRAM = 0x10000,
 };
 
 typedef struct _MsgFormatHandler MsgFormatHandler;

diff --git a/modules/syslogformat/syslog-format.c b/modules/syslogformat/syslog-format.c
@@ -326,6 +326,24 @@ _syslog_format_parse_version(LogMessage *msg, const guchar **data, gint *length)
   return TRUE;
 }
 
+static const gchar program_name_allowed_specials[] = ".-_()/";
+static gsize program_name_allowed_spacial_chars_len = G_N_ELEMENTS(program_name_allowed_specials) - 1;
+
+static inline gboolean
+_validate_program_char(const guchar ch, gboolean *has_alpha)
+{
+  if (isalpha(ch))
+    {
+      *has_alpha = TRUE;
+      return TRUE;
+    }
+  if (isdigit(ch))
+    return TRUE;
+  if (memchr(program_name_allowed_specials, ch, program_name_allowed_spacial_chars_len))
+    return TRUE;
+  return FALSE;
+}
+
 static void
 _syslog_format_parse_legacy_program_name(LogMessage *msg, const guchar **data, gint *length, guint flags)
 {
@@ -336,10 +354,24 @@ _syslog_format_parse_legacy_program_name(LogMessage *msg, const guchar **data, g
   src = *data;
   left = *length;
   prog_start = src;
+  gboolean has_alpha_char = FALSE;
+
   while (left && *src != ' ' && *src != '[' && *src != ':')
     {
+      if (G_UNLIKELY(flags & LP_CHECK_PROGRAM) && !_validate_program_char(*src, &has_alpha_char))
+        {
+          log_msg_set_tag_by_id(msg, LM_T_SYSLOG_RFC_3164_INVALID_PROGRAM);
+          return;
+        }
       _skip_char(&src, &left);
     }
+
+  if (G_UNLIKELY(flags & LP_CHECK_PROGRAM) && !has_alpha_char)
+    {
+      log_msg_set_tag_by_id(msg, LM_T_SYSLOG_RFC_3164_INVALID_PROGRAM);
+      return;
+    }
+
   log_msg_set_value(msg, LM_V_PROGRAM, (gchar *) prog_start, src - prog_start);
   if (left > 0 && *src == '[')
     {

diff --git a/modules/syslogformat/syslog-parser-grammar.ym b/modules/syslogformat/syslog-parser-grammar.ym
@@ -96,6 +96,7 @@ parser_syslog_opt
 parser_syslog_opt_flags
         : string parser_syslog_opt_flags                { CHECK_ERROR(msg_format_options_process_flag(&((SyslogParser *) last_parser)->parse_options, $1), @1, "Unknown flag %s", $1); free($1); }
         | KW_CHECK_HOSTNAME parser_syslog_opt_flags     { msg_format_options_process_flag(&((SyslogParser *) last_parser)->parse_options, "check-hostname");  }
+        | KW_CHECK_PROGRAM parser_syslog_opt_flags       { msg_format_options_process_flag(&((SyslogParser *) last_parser)->parse_options, "check-program");  }
         |
 	;
 

diff --git a/modules/syslogformat/tests/test_syslog_format.c b/modules/syslogformat/tests/test_syslog_format.c
@@ -102,6 +102,92 @@ Test(syslog_format, rfc3164_error_invalid_pri)
   log_msg_unref(msg);
 }
 
+Test(syslog_format, rfc3164_check_program_valid_name)
+{
+  const gchar *data =
+    "<189> Feb  3 12:34:56 host program_0123456789-abcdefghijklmnopqrstuvwxyz(ABCDEFGHIJKLMNOPQRSTUVWXYZ).valid/chars[pid]: message";
+  gsize data_length = strlen(data);
+
+  LogMessage *msg = log_msg_new_empty();
+
+  parse_options.flags |= LP_CHECK_PROGRAM;
+
+  gsize problem_position;
+  cr_assert(syslog_format_handler(&parse_options, msg, (const guchar *) data, data_length, &problem_position));
+  assert_log_message_value_by_name(msg, "HOST", "host");
+  assert_log_message_value_by_name(msg, "PROGRAM",
+                                   "program_0123456789-abcdefghijklmnopqrstuvwxyz(ABCDEFGHIJKLMNOPQRSTUVWXYZ).valid/chars");
+  assert_log_message_value_by_name(msg, "PID", "pid");
+  assert_log_message_value_by_name(msg, "MSG", "message");
+  assert_log_message_value_by_name(msg, "MSGFORMAT", "syslog:rfc3164");
+  cr_assert(!log_msg_is_tag_by_name(msg, "syslog.rfc3164_invalid_program"));
+
+  log_msg_unref(msg);
+}
+
+Test(syslog_format, rfc3164_check_program_decimal_number)
+{
+  const gchar *data = "<189> Feb  3 12:34:56 host 323235243.2354[pid]: message";
+  gsize data_length = strlen(data);
+
+  LogMessage *msg = log_msg_new_empty();
+
+  parse_options.flags |= LP_CHECK_PROGRAM;
+
+  gsize problem_position;
+  cr_assert(syslog_format_handler(&parse_options, msg, (const guchar *) data, data_length, &problem_position));
+  assert_log_message_value_by_name(msg, "HOST", "host");
+  assert_log_message_value_by_name(msg, "PROGRAM", "");
+  assert_log_message_value_by_name(msg, "PID", "");
+  assert_log_message_value_by_name(msg, "MSG", "323235243.2354[pid]: message");
+  assert_log_message_value_by_name(msg, "MSGFORMAT", "syslog:rfc3164");
+  assert_log_message_has_tag(msg, "syslog.rfc3164_invalid_program");
+
+  log_msg_unref(msg);
+}
+
+Test(syslog_format, rfc3164_check_program_invalid_character)
+{
+  const gchar *data = "<189> Feb  3 12:34:56 host program![pid]: message";
+  gsize data_length = strlen(data);
+
+  LogMessage *msg = log_msg_new_empty();
+
+  parse_options.flags |= LP_CHECK_PROGRAM;
+
+  gsize problem_position;
+  cr_assert(syslog_format_handler(&parse_options, msg, (const guchar *) data, data_length, &problem_position));
+  assert_log_message_value_by_name(msg, "HOST", "host");
+  assert_log_message_value_by_name(msg, "PROGRAM", "");
+  assert_log_message_value_by_name(msg, "PID", "");
+  assert_log_message_value_by_name(msg, "MSG", "program![pid]: message");
+  assert_log_message_value_by_name(msg, "MSGFORMAT", "syslog:rfc3164");
+  assert_log_message_has_tag(msg, "syslog.rfc3164_invalid_program");
+
+  log_msg_unref(msg);
+}
+
+Test(syslog_format, rfc3164_check_program_ip_address)
+{
+  const gchar *data = "<189> Feb  3 12:34:56 host 127.0.0.1[pid]: message";
+  gsize data_length = strlen(data);
+
+  LogMessage *msg = log_msg_new_empty();
+
+  parse_options.flags |= LP_CHECK_PROGRAM;
+
+  gsize problem_position;
+  cr_assert(syslog_format_handler(&parse_options, msg, (const guchar *) data, data_length, &problem_position));
+  assert_log_message_value_by_name(msg, "HOST", "host");
+  assert_log_message_value_by_name(msg, "PROGRAM", "");
+  assert_log_message_value_by_name(msg, "PID", "");
+  assert_log_message_value_by_name(msg, "MSG", "127.0.0.1[pid]: message");
+  assert_log_message_value_by_name(msg, "MSGFORMAT", "syslog:rfc3164");
+  assert_log_message_has_tag(msg, "syslog.rfc3164_invalid_program");
+
+  log_msg_unref(msg);
+}
+
 Test(syslog_format, rfc3164_error_missing_timestamp)
 {
   /* incorrect pri value */

diff --git a/news/feature-380.md b/news/feature-380.md
@@ -0,0 +1,14 @@
+`check-program`: Introduced as a flag for global or source options.
+
+By default, this flag is set to false. Enabling the check-program flag triggers `program` name validation for `RFC3164` messages. Valid `program` names must adhere to the following criteria:
+
+Contain only these characters: `[a-zA-Z0-9-_/().]`
+Include at least one alphabetical character.
+If a `program` name fails validation, it will be considered part of the log message.
+
+
+Example:
+
+```
+source { network(flags(check-hostname, check-program)); };
+```