Skip to contentSkip to conversation

mountd: fix checking for errors when exporting filesystems.

PATCH In review NeilBrown 3 participants 3 messages Dec 13, 2012 – Dec 17, 2012

Subsystems (2)
NFS client NFS server
lore
Share & tools
lore

NeilBrown

author
#0 lore raw

PATCH 1/1 mountd: fix checking for errors when exporting filesystems.

commit 5604b35a61e22930873ffc4e9971002f578e7978
nfs-utils: Increase the stdio file buffer size for procfs files

changed writes to some sysfs files to be line buffered (_IOLBF) where they weren't before. While this probably makes sense, it introduced a bug.

With fully buffered streams, you don't expect to get an error until you call fflush().
With line buffered streams you can get the error from fprintf() et al.

qword_eol() only tests the return from fflush(), not from fprintf(). Consequently errors were not noticed.

One result of this is that if you export, with crossmnt, a filesystem underneath which are mounted non-exportable filesystems (e.g. /proc) then an 'ls -l' on the client will block indefinitely waiting for a meaningful 'yes' or 'no' from the server, but will never get one.

This patch changes qword_eol to test both fprintf and fflush.

Signed-off-by NeilBrown <[email protected]>
support/nfs/cacheio.c +9 -4
index e641c45..61e07a8 100644
@@ -162,11 +162,16 @@ int qword_eol(FILE *f)
 {
 	int err;
 
-	fprintf(f,"\n");
-	err = fflush(f);
-	if (err) {
-		xlog_warn("qword_eol: fflush failed: errno %d (%s)",
+	err = fprintf(f,"\n");
+	if (err < 0) {
+		xlog_warn("qword_eol: fprintf failed: errno %d (%s)",
 			    errno, strerror(errno));
+	} else {
+		err = fflush(f);
+		if (err) {
+			xlog_warn("qword_eol: fflush failed: errno %d (%s)",
+				  errno, strerror(errno));
+		}
 	}
 	/*
 	 * We must send one line (and one line only) in a single write
#1 lore raw
Quoted 15 lines · NeilBrown

commit 5604b35a61e22930873ffc4e9971002f578e7978
nfs-utils: Increase the stdio file buffer size for procfs files

changed writes to some sysfs files to be line buffered (_IOLBF) where they weren't before. While this probably makes sense, it introduced a bug.

With fully buffered streams, you don't expect to get an error until you call fflush().
With line buffered streams you can get the error from fprintf() et al.

qword_eol() only tests the return from fflush(), not from fprintf(). Consequently errors were not noticed.

Quoted 6 lines

One result of this is that if you export, with crossmnt, a filesystem underneath which are mounted non-exportable filesystems (e.g. /proc) then an 'ls -l' on the client will block indefinitely waiting for a meaningful 'yes' or 'no' from the server, but will never get one.

This patch changes qword_eol to test both fprintf and fflush.

Makes sense to me.--b.

Quoted 28 lines
Signed-off-by NeilBrown <[email protected]>
support/nfs/cacheio.c +9 -4
index e641c45..61e07a8 100644
@@ -162,11 +162,16 @@ int qword_eol(FILE *f)
 {
 	int err;
 
-	fprintf(f,"\n");
-	err = fflush(f);
-	if (err) {
-		xlog_warn("qword_eol: fflush failed: errno %d (%s)",
+	err = fprintf(f,"\n");
+	if (err < 0) {
+		xlog_warn("qword_eol: fprintf failed: errno %d (%s)",
 			    errno, strerror(errno));
+	} else {
+		err = fflush(f);
+		if (err) {
+			xlog_warn("qword_eol: fflush failed: errno %d (%s)",
+				  errno, strerror(errno));
+		}
 	}
 	/*
 	 * We must send one line (and one line only) in a single write
#2 lore raw
In reply to NeilBrown
Quoted 15 lines · NeilBrown

commit 5604b35a61e22930873ffc4e9971002f578e7978
nfs-utils: Increase the stdio file buffer size for procfs files

changed writes to some sysfs files to be line buffered (_IOLBF) where they weren't before. While this probably makes sense, it introduced a bug.

With fully buffered streams, you don't expect to get an error until you call fflush().
With line buffered streams you can get the error from fprintf() et al.

qword_eol() only tests the return from fflush(), not from fprintf(). Consequently errors were not noticed.

Quoted 8 lines

One result of this is that if you export, with crossmnt, a filesystem underneath which are mounted non-exportable filesystems (e.g. /proc) then an 'ls -l' on the client will block indefinitely waiting for a meaningful 'yes' or 'no' from the server, but will never get one.

This patch changes qword_eol to test both fprintf and fflush.

Signed-off-by NeilBrown <[email protected]>

Committed...

steved.

Quoted 27 lines
support/nfs/cacheio.c +9 -4
index e641c45..61e07a8 100644
@@ -162,11 +162,16 @@ int qword_eol(FILE *f)
 {
 	int err;
 
-	fprintf(f,"\n");
-	err = fflush(f);
-	if (err) {
-		xlog_warn("qword_eol: fflush failed: errno %d (%s)",
+	err = fprintf(f,"\n");
+	if (err < 0) {
+		xlog_warn("qword_eol: fprintf failed: errno %d (%s)",
 			    errno, strerror(errno));
+	} else {
+		err = fflush(f);
+		if (err) {
+			xlog_warn("qword_eol: fflush failed: errno %d (%s)",
+				  errno, strerror(errno));
+		}
 	}
 	/*
 	 * We must send one line (and one line only) in a single write