pg_ctl: Detect current standby state from pg_control

petere · petere · commit c1dc51d4844e · 2016-09-21T12:00:00.000-04:00
pg_ctl used to determine whether a server was in standby mode by looking
for a recovery.conf file.  With this change, it instead looks into
pg_control, which is potentially more accurate.  There are also
occasional discussions about removing recovery.conf, so this removes one
dependency.

Reviewed-by: Michael Paquier &lt;michael.paquier@gmail.com&gt;
diff --git a/src/backend/utils/misc/pg_controldata.c b/src/backend/utils/misc/pg_controldata.c
@@ -52,6 +52,9 @@ pg_control_system(PG_FUNCTION_ARGS)
 
 	/* read the control file */
 	ControlFile = get_controlfile(DataDir, NULL);
+	if (!ControlFile)
+		ereport(ERROR,
+				(errmsg("calculated CRC checksum does not match value stored in file")));
 
 	values[0] = Int32GetDatum(ControlFile->pg_control_version);
 	nulls[0] = false;
@@ -128,6 +131,9 @@ pg_control_checkpoint(PG_FUNCTION_ARGS)
 
 	/* Read the control file. */
 	ControlFile = get_controlfile(DataDir, NULL);
+	if (!ControlFile)
+		ereport(ERROR,
+				(errmsg("calculated CRC checksum does not match value stored in file")));
 
 	/*
 	 * Calculate name of the WAL file containing the latest checkpoint's REDO
@@ -230,6 +236,9 @@ pg_control_recovery(PG_FUNCTION_ARGS)
 
 	/* read the control file */
 	ControlFile = get_controlfile(DataDir, NULL);
+	if (!ControlFile)
+		ereport(ERROR,
+				(errmsg("calculated CRC checksum does not match value stored in file")));
 
 	values[0] = LSNGetDatum(ControlFile->minRecoveryPoint);
 	nulls[0] = false;
@@ -295,6 +304,9 @@ pg_control_init(PG_FUNCTION_ARGS)
 
 	/* read the control file */
 	ControlFile = get_controlfile(DataDir, NULL);
+	if (!ControlFile)
+		ereport(ERROR,
+				(errmsg("calculated CRC checksum does not match value stored in file")));
 
 	values[0] = Int32GetDatum(ControlFile->maxAlign);
 	nulls[0] = false;
diff --git a/src/bin/pg_controldata/pg_controldata.c b/src/bin/pg_controldata/pg_controldata.c
@@ -156,6 +156,10 @@ main(int argc, char *argv[])
 
 	/* get a copy of the control file */
 	ControlFile = get_controlfile(DataDir, progname);
+	if (!ControlFile)
+		printf(_("WARNING: Calculated CRC checksum does not match value stored in file.\n"
+				 "Either the file is corrupt, or it has a different layout than this program\n"
+				 "is expecting.  The results below are untrustworthy.\n\n"));
 
 	/*
 	 * This slightly-chintzy coding will work as long as the control file
diff --git a/src/bin/pg_ctl/pg_ctl.c b/src/bin/pg_ctl/pg_ctl.c
@@ -19,6 +19,8 @@
 
 #include "postgres_fe.h"
 
+#include "catalog/pg_control.h"
+#include "common/controldata_utils.h"
 #include "libpq-fe.h"
 #include "pqexpbuffer.h"
 
@@ -96,7 +98,6 @@ static char postopts_file[MAXPGPATH];
 static char version_file[MAXPGPATH];
 static char pid_file[MAXPGPATH];
 static char backup_file[MAXPGPATH];
-static char recovery_file[MAXPGPATH];
 static char promote_file[MAXPGPATH];
 
 #ifdef WIN32
@@ -158,6 +159,8 @@ static bool postmaster_is_alive(pid_t pid);
 static void unlimit_core_size(void);
 #endif
 
+static DBState get_control_dbstate(void);
+
 
 #ifdef WIN32
 static void
@@ -988,12 +991,12 @@ do_stop(void)
 		/*
 		 * If backup_label exists, an online backup is running. Warn the user
 		 * that smart shutdown will wait for it to finish. However, if
-		 * recovery.conf is also present, we're recovering from an online
+		 * the server is in archive recovery, we're recovering from an online
 		 * backup instead of performing one.
 		 */
 		if (shutdown_mode == SMART_MODE &&
 			stat(backup_file, &statbuf) == 0 &&
-			stat(recovery_file, &statbuf) != 0)
+			get_control_dbstate() != DB_IN_ARCHIVE_RECOVERY)
 		{
 			print_msg(_("WARNING: online backup mode is active\n"
 						"Shutdown will not complete until pg_stop_backup() is called.\n\n"));
@@ -1076,12 +1079,12 @@ do_restart(void)
 		/*
 		 * If backup_label exists, an online backup is running. Warn the user
 		 * that smart shutdown will wait for it to finish. However, if
-		 * recovery.conf is also present, we're recovering from an online
+		 * the server is in archive recovery, we're recovering from an online
 		 * backup instead of performing one.
 		 */
 		if (shutdown_mode == SMART_MODE &&
 			stat(backup_file, &statbuf) == 0 &&
-			stat(recovery_file, &statbuf) != 0)
+			get_control_dbstate() != DB_IN_ARCHIVE_RECOVERY)
 		{
 			print_msg(_("WARNING: online backup mode is active\n"
 						"Shutdown will not complete until pg_stop_backup() is called.\n\n"));
@@ -1168,7 +1171,6 @@ do_promote(void)
 {
 	FILE	   *prmfile;
 	pgpid_t		pid;
-	struct stat statbuf;
 
 	pid = get_pgpid(false);
 
@@ -1187,8 +1189,7 @@ do_promote(void)
 		exit(1);
 	}
 
-	/* If recovery.conf doesn't exist, the server is not in standby mode */
-	if (stat(recovery_file, &statbuf) != 0)
+	if (get_control_dbstate() != DB_IN_ARCHIVE_RECOVERY)
 	{
 		write_stderr(_("%s: cannot promote server; "
 					   "server is not in standby mode\n"),
@@ -2115,6 +2116,35 @@ adjust_data_dir(void)
 }
 
 
+static DBState
+get_control_dbstate(void)
+{
+	DBState ret;
+
+	for (;;)
+	{
+		ControlFileData *control_file_data = get_controlfile(pg_data, progname);
+
+		if (control_file_data)
+		{
+			ret = control_file_data->state;
+			pfree(control_file_data);
+			return ret;
+		}
+
+		if (wait_seconds > 0)
+		{
+			pg_usleep(1000000);		/* 1 sec */
+			wait_seconds--;
+			continue;
+		}
+
+		write_stderr(_("%s: control file appears to be corrupt\n"), progname);
+		exit(1);
+	}
+}
+
+
 int
 main(int argc, char **argv)
 {
@@ -2401,7 +2431,6 @@ main(int argc, char **argv)
 		snprintf(version_file, MAXPGPATH, "%s/PG_VERSION", pg_data);
 		snprintf(pid_file, MAXPGPATH, "%s/postmaster.pid", pg_data);
 		snprintf(backup_file, MAXPGPATH, "%s/backup_label", pg_data);
-		snprintf(recovery_file, MAXPGPATH, "%s/recovery.conf", pg_data);
 	}
 
 	switch (ctl_command)
diff --git a/src/common/controldata_utils.c b/src/common/controldata_utils.c
@@ -33,9 +33,11 @@
  *
  * Get controlfile values. The caller is responsible
  * for pfreeing the result.
+ *
+ * Returns NULL if the CRC did not match.
  */
 ControlFileData *
-get_controlfile(char *DataDir, const char *progname)
+get_controlfile(const char *DataDir, const char *progname)
 {
 	ControlFileData *ControlFile;
 	int			fd;
@@ -82,13 +84,10 @@ get_controlfile(char *DataDir, const char *progname)
 	FIN_CRC32C(crc);
 
 	if (!EQ_CRC32C(crc, ControlFile->crc))
-#ifndef FRONTEND
-		elog(ERROR, _("calculated CRC checksum does not match value stored in file"));
-#else
-		printf(_("WARNING: Calculated CRC checksum does not match value stored in file.\n"
-				 "Either the file is corrupt, or it has a different layout than this program\n"
-				 "is expecting.  The results below are untrustworthy.\n\n"));
-#endif
+	{
+		pfree(ControlFile);
+		return NULL;
+	}
 
 	/* Make sure the control file is valid byte order. */
 	if (ControlFile->pg_control_version % 65536 == 0 &&
diff --git a/src/include/common/controldata_utils.h b/src/include/common/controldata_utils.h
@@ -12,6 +12,6 @@
 
 #include "catalog/pg_control.h"
 
-extern ControlFileData *get_controlfile(char *DataDir, const char *progname);
+extern ControlFileData *get_controlfile(const char *DataDir, const char *progname);
 
 #endif   /* COMMON_CONTROLDATA_UTILS_H */
