[Libguestfs] [PATCH] daemon: Fix read-file so it fails gracefully for large files (RHBZ#589039).

[Richard W.M. Jones]

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
virt-top is 'top' for virtual machines.  Tiny program with many
powerful monitoring features, net stats, disk stats, logging, etc.
http://et.redhat.com/~rjones/virt-top
-------------- next part --------------
>From 9a503ec7d7fa16e09abab53cc7f768b72020ce08 Mon Sep 17 00:00:00 2001
From: Richard Jones <rjones at redhat.com>
Date: Fri, 7 May 2010 13:30:42 +0100
Subject: [PATCH] daemon: Fix read-file so it fails gracefully for large files (RHBZ#589039).

Pengzhen Cao noticed that read-file would fail for files
larger than the protocol size; this is *not* the bug.  However
it would also lose protocol synchronization after this.

The reason was that functions which return RBufferOut in the
generator must not 'touch' the *size_r parameter along error
return paths.

I fixed read-file and initrd-cat, and I checked that pread was
doing the right thing.

This also adds regression tests for read-file with various categories
of large file.
---
 daemon/file.c    |   14 ++++++++++----
 daemon/initrd.c  |   12 ++++++++----
 src/generator.ml |   15 ++++++++++++++-
 3 files changed, 32 insertions(+), 9 deletions(-)

diff --git a/daemon/file.c b/daemon/file.c
index 7a0f8f9..2399828 100644
--- a/daemon/file.c
+++ b/daemon/file.c
@@ -336,25 +336,24 @@ do_read_file (const char *path, size_t *size_r)
     return NULL;
   }
 
-  *size_r = statbuf.st_size;
   /* The actual limit on messages is smaller than this.  This
    * check just limits the amount of memory we'll try and allocate
    * here.  If the message is larger than the real limit, that will
    * be caught later when we try to serialize the message.
    */
-  if (*size_r >= GUESTFS_MESSAGE_MAX) {
+  if (statbuf.st_size >= GUESTFS_MESSAGE_MAX) {
     reply_with_error ("%s: file is too large for the protocol, use guestfs_download instead", path);
     close (fd);
     return NULL;
   }
-  r = malloc (*size_r);
+  r = malloc (statbuf.st_size);
   if (r == NULL) {
     reply_with_perror ("malloc");
     close (fd);
     return NULL;
   }
 
-  if (xread (fd, r, *size_r) == -1) {
+  if (xread (fd, r, statbuf.st_size) == -1) {
     reply_with_perror ("read: %s", path);
     close (fd);
     free (r);
@@ -367,6 +366,10 @@ do_read_file (const char *path, size_t *size_r)
     return NULL;
   }
 
+  /* Mustn't touch *size_r until we are sure that we won't return any
+   * error (RHBZ#589039).
+   */
+  *size_r = statbuf.st_size;
   return r;
 }
 
@@ -418,6 +421,9 @@ do_pread (const char *path, int count, int64_t offset, size_t *size_r)
     return NULL;
   }
 
+  /* Mustn't touch *size_r until we are sure that we won't return any
+   * error (RHBZ#589039).
+   */
   *size_r = r;
   return buf;
 }
diff --git a/daemon/initrd.c b/daemon/initrd.c
index addeb14..932564b 100644
--- a/daemon/initrd.c
+++ b/daemon/initrd.c
@@ -142,25 +142,24 @@ do_initrd_cat (const char *path, const char *filename, size_t *size_r)
     goto cleanup;
   }
 
-  *size_r = statbuf.st_size;
   /* The actual limit on messages is smaller than this.  This
    * check just limits the amount of memory we'll try and allocate
    * here.  If the message is larger than the real limit, that will
    * be caught later when we try to serialize the message.
    */
-  if (*size_r >= GUESTFS_MESSAGE_MAX) {
+  if (statbuf.st_size >= GUESTFS_MESSAGE_MAX) {
     reply_with_error ("%s:%s: file is too large for the protocol",
                       path, filename);
     goto cleanup;
   }
 
-  ret = malloc (*size_r);
+  ret = malloc (statbuf.st_size);
   if (ret == NULL) {
     reply_with_perror ("malloc");
     goto cleanup;
   }
 
-  if (xread (fd, ret, *size_r) == -1) {
+  if (xread (fd, ret, statbuf.st_size) == -1) {
     reply_with_perror ("read: %s:%s", path, filename);
     free (ret);
     ret = NULL;
@@ -175,6 +174,11 @@ do_initrd_cat (const char *path, const char *filename, size_t *size_r)
   }
   fd = -1;
 
+  /* Mustn't touch *size_r until we are sure that we won't return any
+   * error (RHBZ#589039).
+   */
+  *size_r = statbuf.st_size;
+
  cleanup:
   if (fd >= 0)
     close (fd);
diff --git a/src/generator.ml b/src/generator.ml
index cbed94b..9dd8f8c 100755
--- a/src/generator.ml
+++ b/src/generator.ml
@@ -3272,7 +3272,20 @@ for full details.");
 
   ("read_file", (RBufferOut "content", [Pathname "path"]), 150, [ProtocolLimitWarning],
    [InitISOFS, Always, TestOutputBuffer (
-      [["read_file"; "/known-4"]], "abc\ndef\nghi")],
+      [["read_file"; "/known-4"]], "abc\ndef\nghi");
+    (* Test various near large, large and too large files (RHBZ#589039). *)
+    InitBasicFS, Always, TestLastFail (
+      [["touch"; "/a"];
+       ["truncate_size"; "/a"; "4194303"]; (* GUESTFS_MESSAGE_MAX - 1 *)
+       ["read_file"; "/a"]]);
+    InitBasicFS, Always, TestLastFail (
+      [["touch"; "/a"];
+       ["truncate_size"; "/a"; "4194304"]; (* GUESTFS_MESSAGE_MAX *)
+       ["read_file"; "/a"]]);
+    InitBasicFS, Always, TestLastFail (
+      [["touch"; "/a"];
+       ["truncate_size"; "/a"; "41943040"]; (* GUESTFS_MESSAGE_MAX * 10 *)
+       ["read_file"; "/a"]])],
    "read a file",
    "\
 This calls returns the contents of the file C<path> as a
-- 
1.6.6.1