Address set of issues with errno handling

System calls mixed up in error code paths are causing two issues which
several code paths have not correctly handled:
1) For write() calls, sometimes the system may return less bytes than
what has been written without errno being set.  Some paths were careful
enough to consider that case, and assumed that errno should be set to
ENOSPC, other calls missed that.
2) errno generated by a system call is overwritten by other system calls
which may succeed once an error code path is taken, causing what is
reported to the user to be incorrect.

This patch uses the brute-force approach of correcting all those code
paths.  Some refactoring could happen in the future, but this is let as
future work, which is not targeted for back-branches anyway.

Author: Michael Paquier
Reviewed-by: Ashutosh Sharma
Discussion: https://postgr.es/m/20180622061535.GD5215@paquier.xyz

src/backend/access/heap/rewriteheap.c

@@ -1163,9 +1163,14 @@ heap_xlog_logical_rewrite(XLogReaderState *r)
 
 	/* write out tail end of mapping file (again) */
 	if (write(fd, data, len) != len)
+	{
+		/* if write didn't set errno, assume problem is no disk space */
+		if (errno == 0)
+			errno = ENOSPC;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not write to file \"%s\": %m", path)));
+	}
 
 	/*
 	 * Now fsync all previously written data. We could improve things and only

src/backend/access/transam/twophase.c

@@ -1171,12 +1171,17 @@ ReadTwoPhaseFile(TransactionId xid, bool give_warnings)
 	 */
 	if (fstat(fd, &stat))
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
 		if (give_warnings)
+		{
+			errno = save_errno;
 			ereport(WARNING,
 					(errcode_for_file_access(),
 					 errmsg("could not stat two-phase state file \"%s\": %m",
 							path)));
+		}
 		return NULL;
 	}
 
@@ -1203,12 +1208,17 @@ ReadTwoPhaseFile(TransactionId xid, bool give_warnings)
 
 	if (read(fd, buf, stat.st_size) != stat.st_size)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
 		if (give_warnings)
+		{
+			errno = save_errno;
 			ereport(WARNING,
 					(errcode_for_file_access(),
 					 errmsg("could not read two-phase state file \"%s\": %m",
 							path)));
+		}
 		pfree(buf);
 		return NULL;
 	}
@@ -1550,14 +1560,24 @@ RecreateTwoPhaseFile(TransactionId xid, void *content, int len)
 	/* Write content and CRC */
 	if (write(fd, content, len) != len)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not write two-phase state file: %m")));
 	}
 	if (write(fd, &statefile_crc, sizeof(pg_crc32c)) != sizeof(pg_crc32c))
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not write two-phase state file: %m")));
@@ -1569,7 +1589,10 @@ RecreateTwoPhaseFile(TransactionId xid, void *content, int len)
 	 */
 	if (pg_fsync(fd) != 0)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+		errno = save_errno;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not fsync two-phase state file: %m")));

src/backend/access/transam/xlog.c

@@ -3087,7 +3087,10 @@ XLogFileInit(XLogSegNo logsegno, bool *use_existent, bool use_lock)
 
 	if (pg_fsync(fd) != 0)
 	{
+		int			save_errno = errno;
+
 		close(fd);
+		errno = save_errno;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not fsync file \"%s\": %m", tmppath)));
@@ -11255,8 +11258,10 @@ retry:
 	if (lseek(readFile, (off_t) readOff, SEEK_SET) < 0)
 	{
 		char		fname[MAXFNAMELEN];
+		int			save_errno = errno;
 
 		XLogFileName(fname, curFileTLI, readSegNo);
+		errno = save_errno;
 		ereport(emode_for_corrupt_record(emode, targetPagePtr + reqLen),
 				(errcode_for_file_access(),
 				 errmsg("could not seek in log segment %s to offset %u: %m",
@@ -11267,8 +11272,10 @@ retry:
 	if (read(readFile, readBuf, XLOG_BLCKSZ) != XLOG_BLCKSZ)
 	{
 		char		fname[MAXFNAMELEN];
+		int			save_errno = errno;
 
 		XLogFileName(fname, curFileTLI, readSegNo);
+		errno = save_errno;
 		ereport(emode_for_corrupt_record(emode, targetPagePtr + reqLen),
 				(errcode_for_file_access(),
 				 errmsg("could not read from log segment %s, offset %u: %m",

src/backend/access/transam/xlogutils.c

@@ -710,9 +710,11 @@ XLogRead(char *buf, TimeLineID tli, XLogRecPtr startptr, Size count)
 			if (lseek(sendFile, (off_t) startoff, SEEK_SET) < 0)
 			{
 				char		path[MAXPGPATH];
+				int			save_errno = errno;
 
 				XLogFilePath(path, tli, sendSegNo);
 
+				errno = save_errno;
 				ereport(ERROR,
 						(errcode_for_file_access(),
 				  errmsg("could not seek in log segment %s to offset %u: %m",
@@ -731,9 +733,11 @@ XLogRead(char *buf, TimeLineID tli, XLogRecPtr startptr, Size count)
 		if (readbytes <= 0)
 		{
 			char		path[MAXPGPATH];
+			int			save_errno = errno;
 
 			XLogFilePath(path, tli, sendSegNo);
 
+			errno = save_errno;
 			ereport(ERROR,
 					(errcode_for_file_access(),
 					 errmsg("could not read from log segment %s, offset %u, length %lu: %m",

src/backend/replication/basebackup.c

@@ -392,6 +392,8 @@ perform_base_backup(basebackup_options *opt, DIR *tblspcdir)
 			fp = AllocateFile(pathbuf, "rb");
 			if (fp == NULL)
 			{
+				int			save_errno = errno;
+
 				/*
 				 * Most likely reason for this is that the file was already
 				 * removed by a checkpoint, so check for that to get a better
@@ -399,6 +401,7 @@ perform_base_backup(basebackup_options *opt, DIR *tblspcdir)
 				 */
 				CheckXLogRemoved(segno, tli);
 
+				errno = save_errno;
 				ereport(ERROR,
 						(errcode_for_file_access(),
 						 errmsg("could not open file \"%s\": %m", pathbuf)));

src/backend/replication/logical/origin.c

@@ -549,7 +549,12 @@ CheckPointReplicationOrigin(void)
 	/* write magic */
 	if ((write(tmpfd, &magic, sizeof(magic))) != sizeof(magic))
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(tmpfd);
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 		ereport(PANIC,
 				(errcode_for_file_access(),
 				 errmsg("could not write to file \"%s\": %m",
@@ -588,7 +593,12 @@ CheckPointReplicationOrigin(void)
 		if ((write(tmpfd, &disk_state, sizeof(disk_state))) !=
 			sizeof(disk_state))
 		{
+			int			save_errno = errno;
+
 			CloseTransientFile(tmpfd);
+
+			/* if write didn't set errno, assume problem is no disk space */
+			errno = save_errno ? save_errno : ENOSPC;
 			ereport(PANIC,
 					(errcode_for_file_access(),
 					 errmsg("could not write to file \"%s\": %m",
@@ -604,7 +614,12 @@ CheckPointReplicationOrigin(void)
 	FIN_CRC32C(crc);
 	if ((write(tmpfd, &crc, sizeof(crc))) != sizeof(crc))
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(tmpfd);
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 		ereport(PANIC,
 				(errcode_for_file_access(),
 				 errmsg("could not write to file \"%s\": %m",

src/backend/replication/logical/reorderbuffer.c

@@ -2352,7 +2352,9 @@ ReorderBufferSerializeChange(ReorderBuffer *rb, ReorderBufferTXN *txn,
 		int			save_errno = errno;
 
 		CloseTransientFile(fd);
-		errno = save_errno;
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not write to data file for XID %u: %m",

src/backend/replication/logical/snapbuild.c

@@ -1583,7 +1583,12 @@ SnapBuildSerialize(SnapBuild *builder, XLogRecPtr lsn)
 
 	if ((write(fd, ondisk, needed_length)) != needed_length)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not write to file \"%s\": %m", tmppath)));
@@ -1599,7 +1604,10 @@ SnapBuildSerialize(SnapBuild *builder, XLogRecPtr lsn)
 	 */
 	if (pg_fsync(fd) != 0)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+		errno = save_errno;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not fsync file \"%s\": %m", tmppath)));
@@ -1681,7 +1689,10 @@ SnapBuildRestore(SnapBuild *builder, XLogRecPtr lsn)
 	readBytes = read(fd, &ondisk, SnapBuildOnDiskConstantSize);
 	if (readBytes != SnapBuildOnDiskConstantSize)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+		errno = save_errno;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not read file \"%s\", read %d of %d: %m",
@@ -1707,7 +1718,10 @@ SnapBuildRestore(SnapBuild *builder, XLogRecPtr lsn)
 	readBytes = read(fd, &ondisk.builder, sizeof(SnapBuild));
 	if (readBytes != sizeof(SnapBuild))
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+		errno = save_errno;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not read file \"%s\", read %d of %d: %m",
@@ -1722,7 +1736,10 @@ SnapBuildRestore(SnapBuild *builder, XLogRecPtr lsn)
 	readBytes = read(fd, ondisk.builder.was_running.was_xip, sz);
 	if (readBytes != sz)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+		errno = save_errno;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not read file \"%s\", read %d of %d: %m",
@@ -1736,7 +1753,10 @@ SnapBuildRestore(SnapBuild *builder, XLogRecPtr lsn)
 	readBytes = read(fd, ondisk.builder.committed.xip, sz);
 	if (readBytes != sz)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+		errno = save_errno;
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not read file \"%s\", read %d of %d: %m",

src/backend/replication/slot.c

@@ -1085,7 +1085,9 @@ SaveSlotToPath(ReplicationSlot *slot, const char *dir, int elevel)
 		int			save_errno = errno;
 
 		CloseTransientFile(fd);
-		errno = save_errno;
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 		ereport(elevel,
 				(errcode_for_file_access(),
 				 errmsg("could not write to file \"%s\": %m",
@@ -1184,7 +1186,10 @@ RestoreSlotFromDisk(const char *name)
 	 */
 	if (pg_fsync(fd) != 0)
 	{
+		int			save_errno = errno;
+
 		CloseTransientFile(fd);
+		errno = save_errno;
 		ereport(PANIC,
 				(errcode_for_file_access(),
 				 errmsg("could not fsync file \"%s\": %m",

src/bin/pg_basebackup/receivelog.c

@@ -149,6 +149,9 @@ open_walfile(StreamCtl *stream, XLogRecPtr startpoint)
 	{
 		if (write(f, zerobuf, XLOG_BLCKSZ) != XLOG_BLCKSZ)
 		{
+			/* if write didn't set errno, assume problem is no disk space */
+			if (errno == 0)
+				errno = ENOSPC;
 			fprintf(stderr,
 					_("%s: could not pad transaction log file \"%s\": %s\n"),
 					progname, fn, strerror(errno));
@@ -334,7 +337,9 @@ writeTimeLineHistoryFile(StreamCtl *stream, char *filename, char *content)
 		 */
 		close(fd);
 		unlink(tmppath);
-		errno = save_errno;
+
+		/* if write didn't set errno, assume problem is no disk space */
+		errno = save_errno ? save_errno : ENOSPC;
 
 		fprintf(stderr, _("%s: could not write timeline history file \"%s\": %s\n"),
 				progname, tmppath, strerror(errno));
@@ -343,7 +348,10 @@ writeTimeLineHistoryFile(StreamCtl *stream, char *filename, char *content)
 
 	if (fsync(fd) != 0)
 	{
+		int			save_errno = errno;
+
 		close(fd);
+		errno = save_errno;
 		fprintf(stderr, _("%s: could not fsync file \"%s\": %s\n"),
 				progname, tmppath, strerror(errno));
 		return false;
@@ -1185,6 +1193,9 @@ ProcessXLogDataMsg(PGconn *conn, StreamCtl *stream, char *copybuf, int len,
 				  copybuf + hdr_len + bytes_written,
 				  bytes_to_write) != bytes_to_write)
 		{
+			/* if write didn't set errno, assume problem is no disk space */
+			if (errno == 0)
+				errno = ENOSPC;
 			fprintf(stderr,
 				  _("%s: could not write %u bytes to WAL file \"%s\": %s\n"),
 					progname, bytes_to_write, current_walfile_name,