From 2bea59f0420b24a873c668f48f883aa0d9657b49 Mon Sep 17 00:00:00 2001
From: Andrea Mazzoleni <amadvance@users.sourceforge.net>
Date: Sat, 12 Mar 2011 23:21:32 +0100
Subject: [PATCH] Adds more extensive tests

---
 INSTALL       |   5 +-
 Makefile.am   |  27 +++++----
 README        |  22 ++++++-
 config.h      |   6 +-
 configure.ac  |   2 +-
 mktest.c      | 161 ++++++++++++++++++++++++++++++++++++++++++++++++++
 snapraid.d    |  45 +++++++++-----
 snapraid.html |  58 ++++++++++++------
 snapraid.txt  |  47 ++++++++++-----
 testgen.c     |  76 ------------------------
 10 files changed, 308 insertions(+), 141 deletions(-)
 create mode 100644 mktest.c
 delete mode 100644 testgen.c

diff --git a/INSTALL b/INSTALL
index c5e0fccf..5c9af39d 100644
--- a/INSTALL
+++ b/INSTALL
@@ -11,9 +11,10 @@ type:
 
     make install
 
-then change the example configuration file snapraid.conf to fit
-your needs and copy it in /etc/snapraid.conf.
+then change the example configuration file snapraid.conf.example
+to fit your needs and copy it in /etc/snapraid.conf
 
 The documentation is in the man page:
 
     man snapraid
+
diff --git a/Makefile.am b/Makefile.am
index 919e8c88..9f689613 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -1,5 +1,5 @@
 bin_PROGRAMS = snapraid
-check_PROGRAMS = testgen
+check_PROGRAMS = mktest
 
 snapraid_SOURCES = \
 	snapraid.c \
@@ -33,8 +33,8 @@ noinst_HEADERS = \
 	tommyhashdyn.c \
 	tommylist.c
 
-testgen_SOURCES = \
-	testgen.c
+mktest_SOURCES = \
+	mktest.c
 
 EXTRA_DIST = \
 	autogen.sh \
@@ -63,29 +63,34 @@ check-local:
 	mkdir test/disk1
 	mkdir test/disk2
 	mkdir test/disk3
-	echo Create some files, sync and check
-	./testgen 1 3 2000
+	echo --- Create some files, sync and check
+	./mktest generate 1 3 1000 4000
 	./snapraid -c $(srcdir)/test.conf sync
 	./snapraid -c $(srcdir)/test.conf check
-	echo Create some more files, sync and check
-	./testgen 2 3 2000
+	echo --- Create some more files, sync and check
+	./mktest generate 2 3 1000 4000
 	./snapraid -c $(srcdir)/test.conf sync
 	./snapraid -c $(srcdir)/test.conf check 
-	echo Delete some files, fix and check
+	echo --- Delete some files, fix and check
 	rm test/disk1/a*
 	./snapraid -c $(srcdir)/test.conf check 2>/dev/null
 	./snapraid -c $(srcdir)/test.conf fix 2>/dev/null
 	./snapraid -c $(srcdir)/test.conf check
-	echo Delete the parity, fix and check
+	echo --- Delete the parity, fix and check
 	rm test/parity
 	./snapraid -c $(srcdir)/test.conf check 2>/dev/null
 	./snapraid -c $(srcdir)/test.conf fix 2>/dev/null
 	./snapraid -c $(srcdir)/test.conf check
-	echo Corrupt the parity, fix and check
-	dd if=/dev/zero of=test/parity bs=1024 count=1000 seek=1000 conv=notrunc
+	echo --- Corrupt the parity, fix and check
+	./mktest damage 1 10 10000 test/parity
 	./snapraid -c $(srcdir)/test.conf check 2>/dev/null
 	./snapraid -c $(srcdir)/test.conf fix 2>/dev/null
 	./snapraid -c $(srcdir)/test.conf check
+	echo --- Corrupt some files, fix and check
+	./mktest damage 1 1 500 test/disk2/b*
+	./snapraid -c $(srcdir)/test.conf check 2>/dev/null
+	./snapraid -c $(srcdir)/test.conf fix 2>/dev/null
+	./snapraid -c $(srcdir)/test.conf check	
 	echo Success!
 
 valgrind:
diff --git a/README b/README
index e8392241..a529d906 100644
--- a/README
+++ b/README
@@ -2,9 +2,29 @@ SnapRAID
 ========
 
 SnapRAID is a backup program for a disk array using redundancy.
-Just like a RAID system but based on snapshots.
+
+SnapRAID uses a disk of the array to store redundancy information,
+and it allows to recover from a complete disk failure.
+
+SnapRAID is mainly targeted for a home media server, where you have
+a lot of big files that rarely change.
+
+The main features of SnapRAID are:
+
+* You can start using SnapRAID with already filled disks.
+  You do not have to reformat or copy your data.
+* The disks of the array can have different sizes.
+  You have only to use the biggest one for the redundancy data.
+* If you accidentally delete some files in a disk, you can
+  recover them.
+* If more than one disk fails, you lose the data only on the failed disks.
+  All the data in the other disks is safe.
+* It doesn't lock-in your data.
+  You can stop using SnapRAID at any time.
 
 The official site of SnapRAID is:
 
     http://snapraid.sourceforge.net
 
+
+
diff --git a/config.h b/config.h
index eb91b443..b3e6c060 100644
--- a/config.h
+++ b/config.h
@@ -119,7 +119,7 @@
 #define PACKAGE_NAME "snapraid"
 
 /* Define to the full name and version of this package. */
-#define PACKAGE_STRING "snapraid 0.1"
+#define PACKAGE_STRING "snapraid 0.2"
 
 /* Define to the one symbol short name of this package. */
 #define PACKAGE_TARNAME "snapraid"
@@ -128,7 +128,7 @@
 #define PACKAGE_URL "http://snapraid.sourceforge.net"
 
 /* Define to the version of this package. */
-#define PACKAGE_VERSION "0.1"
+#define PACKAGE_VERSION "0.2"
 
 /* Define to 1 if you have the ANSI C header files. */
 #define STDC_HEADERS 1
@@ -159,7 +159,7 @@
 
 
 /* Version number of package */
-#define VERSION "0.1"
+#define VERSION "0.2"
 
 /* Define WORDS_BIGENDIAN to 1 if your processor stores words with the most
    significant byte first (like Motorola and SPARC, unlike Intel). */
diff --git a/configure.ac b/configure.ac
index 74c469bb..61181f04 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1,6 +1,6 @@
 dnl Process this file with autoconf to produce a configure script.
 AC_PREREQ([2.65])
-AC_INIT([snapraid], [0.1], [], [], [http://snapraid.sourceforge.net])
+AC_INIT([snapraid], [0.2], [], [], [http://snapraid.sourceforge.net])
 AM_INIT_AUTOMAKE([foreign no-dependencies])
 AC_CONFIG_SRCDIR([snapraid.c])
 AC_CONFIG_HEADER([config.h])
diff --git a/mktest.c b/mktest.c
new file mode 100644
index 00000000..98251f61
--- /dev/null
+++ b/mktest.c
@@ -0,0 +1,161 @@
+/*
+ * Copyright (C) 2011 Andrea Mazzoleni
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include "portable.h"
+
+/**
+ * Pseudo random number generator.
+ */
+unsigned long long seed = 0;
+
+unsigned rnd(unsigned max) 
+{
+	seed = seed * 6364136223846793005LL + 1442695040888963407LL;
+
+	return (seed >> 32) % max;
+}
+
+void generate(int disk, int size)
+{
+	char path[PATH_MAX];
+	char name[PATH_MAX];
+	FILE* f;
+	int count;
+	int i;
+	int l;
+
+	l = 1 + rnd(32);
+
+	for(i=0;i<l;++i) {
+		name[i] = 'a' + rnd(22);
+	}
+	name[i] = 0;
+
+	snprintf(path, sizeof(path), "test/disk%d/%s", disk, name);
+
+	f = fopen(path, "w");
+	if (!f) {
+		fprintf(stderr, "Error writing %s\n", path);
+		exit(EXIT_FAILURE);
+	}
+
+	count = size;
+	while (count) {
+		fputc(rnd(256), f);
+		--count;
+	}
+
+
+	fclose(f);
+}
+
+void damage(const char* path, int size)
+{
+	FILE* f;
+	struct stat st;
+	off_t start;
+	int count;
+
+	f = fopen(path, "r+");
+	if (!f) {
+		fprintf(stderr, "Error writing %s\n", path);
+		exit(EXIT_FAILURE);
+	}
+
+	if (fstat(fileno(f), &st) != 0) {
+		fprintf(stderr, "Error accessing %s\n", path);
+		exit(EXIT_FAILURE);
+	}
+
+	/* start at random position inside the file */
+	start = rnd(st.st_size);
+	if (fseek(f, start, SEEK_SET) != 0) {
+		fprintf(stderr, "Error seeking %s\n", path);
+		exit(EXIT_FAILURE);
+	}
+
+	/* write garbage, in case also over the end */
+	count = rnd(size);
+	while (count) {
+		fputc(rnd(256), f);
+		--count;
+	}
+
+	fclose(f);
+}
+
+void help(void)
+{
+	printf("Test for " PACKAGE " v" VERSION " by Andrea Mazzoleni, " PACKAGE_URL "\n");
+	printf("Usage:\n");
+	printf("\tmktest generate SEED DISK_NUM FILE_NUM FILE_SIZE\n");
+	printf("\tmktest damage SEED FAIL_NUM FAIL_SIZE FILE\n");
+}
+
+int main(int argc, char* argv[])
+{
+	int i, j;
+
+	if (argc < 2) {
+		help();
+		exit(EXIT_FAILURE);
+	}
+
+	if (strcmp(argv[1], "generate") == 0) {
+		int disk, file, size;
+
+		if (argc != 6) {
+			help();
+			exit(EXIT_FAILURE);
+		}
+
+		seed = atoi(argv[2]);
+		disk = atoi(argv[3]);
+		file = atoi(argv[4]);
+		size = atoi(argv[5]);
+
+		for(i=0;i<disk;++i) {
+			for(j=0;j<file;++j) {
+				generate(i+1, rnd(size));
+			}
+		}
+	} else if (strcmp(argv[1], "damage") == 0) {
+		int fail, size;
+		const char* file;
+
+		if (argc < 6) {
+			help();
+			exit(EXIT_FAILURE);
+		}
+
+		seed = atoi(argv[2]);
+		fail = atoi(argv[3]);
+		size = atoi(argv[4]);
+
+		for(i=5;i<argc;++i) {
+			for(j=0;j<fail;++j) {
+				damage(argv[i], rnd(size));
+			}
+		}
+	} else {
+		help();
+		exit(EXIT_FAILURE);
+	}
+
+	return 0;
+}
+
diff --git a/snapraid.d b/snapraid.d
index 074f2aa4..2e9e48cc 100644
--- a/snapraid.d
+++ b/snapraid.d
@@ -12,12 +12,10 @@ Description
 	SnapRAID is a backup program for a disk array using redundancy.
 
 	SnapRAID uses a disk of the array to store redundancy information,
-	and it allows to recover from a complete disk failure.
+	and it allows to recover from a disk failure.
 
-	SnapRAID is based on the snapshot concept like a backup program.
-	You decide the time to make the "sync" of the redundancy information,
-	and you are able to restore all the changes done at a disk of the array,
-	until the "sync".
+	SnapRAID is mainly targeted for a home media server, where you have
+	a lot of big files that rarely change.
 
 	The main features of SnapRAID are:
 
@@ -25,18 +23,37 @@ Description
 		You do not have to reformat or copy your data.
 	* The disks of the array can have different sizes.
 		You have only to use the biggest one for the redundancy data.
-	* The data in the disks is not changed in any way.
-		It doesn't even need write access at the data.
-	* If more than one disk fails, you lose the data only on the failed disks.
-		All the data in the other disks is safe.
+	* If you accidentally delete some files in a disk, you can
+		recover them.
+	* If more than one disk fails, you lose the data only on the
+		failed disks. All the data in the other disks is safe.
 	* It doesn't lock-in your data.
-		You can stop using SnapRAID at any time.
+		You can stop using SnapRAID at any time without the
+		need to reformat or move data.
+
+	The official site of SnapRAID is:
+
+		:http://snapraid.sourceforge.net
 
 Limitations
 	SnapRAID is in between a RAID and a backup program trying to get the best
-	benefits of them. Altough it also has some downsides that you should
-	consider before using it:
+	benefits of them. Altought it also has some limitations that you should
+	consider before using it.
 
+	The main one, is that if a disk fail, and you haven't recently synched,
+	you may not able to do a complete recover.
+	More specifically, you may be unable to recover up to the size of the
+	amount of the changed or deleted files, from the last sync operation.
+	This happens even if the files changed or deleted are not in the
+	failed disk.
+	New added files don't prevent the recovering of the already existing
+	files. You may only lose the just added files, if they were on the failed
+	disk.
+
+	This is the reason because SnapRAID is better suited for data that
+	rarely change.
+
+	Other limitations are:
 	* You have different filesystems for each disk.
 		Using a RAID you have only a big filesystem.
 	* It doesn't stripe data.
@@ -49,7 +66,7 @@ Limitations
 
 Getting Started
 	To use SnapRAID you need first to mount all the disks of your disk
-	array and select which one to dedicate as redundancy information.
+	array and select which one to dedicate at the redundancy information.
 
 	This disk will be dedicated to this pourpuse only, and you should
 	not store any other data on it.
@@ -79,7 +96,7 @@ Getting Started
 		:snapraid sync
 
 	This process will take some hours at best the first time, depending on
-	the size of the data already present in the data disks.
+	the size of the data already present in the disks.
 	If the disks are empty the process is immediate.
 	You can stop it at any time pressing Ctrl+C, and at the next run it
 	will start where interrupted.
diff --git a/snapraid.html b/snapraid.html
index 2cf18a95..f603a6de 100644
--- a/snapraid.html
+++ b/snapraid.html
@@ -22,12 +22,10 @@ Description
 SnapRAID is a backup program for a disk array using redundancy.
 <p>
 SnapRAID uses a disk of the array to store redundancy information,
-and it allows to recover from a complete disk failure.
+and it allows to recover from a disk failure.
 <p>
-SnapRAID is based on the snapshot concept like a backup program.
-You decide the time to make the "sync" of the redundancy information,
-and you are able to restore all the changes done at a disk of the array,
-until the "sync".
+SnapRAID is mainly targeted for a home media server, where you have
+a lot of big files that rarely change.
 <p>
 The main features of SnapRAID are:
 <ul>
@@ -40,29 +38,53 @@ The disks of the array can have different sizes.
 You have only to use the biggest one for the redundancy data.
 </li>
 <li>
-The data in the disks is not changed in any way.
-It doesn't even need write access at the data.
+If you accidentally delete some files in a disk, you can
+recover them.
 </li>
 <li>
-If more than one disk fails, you lose the data only on the failed disks.
-All the data in the other disks is safe.
+If more than one disk fails, you lose the data only on the
+failed disks. All the data in the other disks is safe.
 </li>
 <li>
 It doesn't lock-in your data.
-You can stop using SnapRAID at any time.
+You can stop using SnapRAID at any time without the
+need to reformat or move data.
 </li>
 </ul>
+<p>
+The official site of SnapRAID is:
+<p>
+<table width="100%" border="0" cellspacing="0" cellpadding="2"><tr><td width="5%"></td><td width="95%">
+<pre>
+<a href="http://snapraid.sourceforge.net">http://snapraid.sourceforge.net</a>
+</pre>
+</td></tr></table>
 <h2><a name="3">3</a> 
 Limitations
 </h2>
 SnapRAID is in between a RAID and a backup program trying to get the best
-benefits of them. Altough it also has some downsides that you should
-consider before using it:
-<ul>
-<li>
-You have different filesystems for each disk.
+benefits of them. Altought it also has some limitations that you should
+consider before using it.
+<p>
+The main one, is that if a disk fail, and you haven't recently synched,
+you may not able to do a complete recover.
+More specifically, you may be unable to recover up to the size of the
+amount of the changed or deleted files, from the last sync operation.
+This happens even if the files changed or deleted are not in the
+failed disk.
+New added files don't prevent the recovering of the already existing
+files. You may only lose the just added files, if they were on the failed
+disk.
+<p>
+This is the reason because SnapRAID is better suited for data that
+rarely change.
+<p>
+Other limitations are:
+* You have different filesystems for each disk.
+<table width="100%" border="0" cellspacing="0" cellpadding="2"><tr><td width="5%"></td><td width="95%">
 Using a RAID you have only a big filesystem.
-</li>
+</td></tr></table>
+<ul>
 <li>
 It doesn't stripe data.
 With RAID you get a speed boost with striping.
@@ -81,7 +103,7 @@ failure of the whole disk array.
 Getting Started
 </h2>
 To use SnapRAID you need first to mount all the disks of your disk
-array and select which one to dedicate as redundancy information.
+array and select which one to dedicate at the redundancy information.
 <p>
 This disk will be dedicated to this pourpuse only, and you should
 not store any other data on it.
@@ -123,7 +145,7 @@ snapraid sync
 </td></tr></table>
 <p>
 This process will take some hours at best the first time, depending on
-the size of the data already present in the data disks.
+the size of the data already present in the disks.
 If the disks are empty the process is immediate.
 You can stop it at any time pressing Ctrl+C, and at the next run it
 will start where interrupted.
diff --git a/snapraid.txt b/snapraid.txt
index 4b8520c2..1451a9e7 100644
--- a/snapraid.txt
+++ b/snapraid.txt
@@ -19,12 +19,10 @@ snapraid [-V, --version] [-h, --help]
 SnapRAID is a backup program for a disk array using redundancy.
 
 SnapRAID uses a disk of the array to store redundancy information,
-and it allows to recover from a complete disk failure.
+and it allows to recover from a disk failure.
 
-SnapRAID is based on the snapshot concept like a backup program.
-You decide the time to make the "sync" of the redundancy information,
-and you are able to restore all the changes done at a disk of the array,
-until the "sync".
+SnapRAID is mainly targeted for a home media server, where you have
+a lot of big files that rarely change.
 
 The main features of SnapRAID are:
 
@@ -32,23 +30,42 @@ The main features of SnapRAID are:
   You do not have to reformat or copy your data.
 * The disks of the array can have different sizes.
   You have only to use the biggest one for the redundancy data.
-* The data in the disks is not changed in any way.
-  It doesn't even need write access at the data.
-* If more than one disk fails, you lose the data only on the failed disks.
-  All the data in the other disks is safe.
+* If you accidentally delete some files in a disk, you can
+  recover them.
+* If more than one disk fails, you lose the data only on the
+  failed disks. All the data in the other disks is safe.
 * It doesn't lock-in your data.
-  You can stop using SnapRAID at any time.
+  You can stop using SnapRAID at any time without the
+  need to reformat or move data.
+
+The official site of SnapRAID is:
+
+    http://snapraid.sourceforge.net
 
 
 3 LIMITATIONS
 =============
 
 SnapRAID is in between a RAID and a backup program trying to get the best
-benefits of them. Altough it also has some downsides that you should
-consider before using it:
+benefits of them. Altought it also has some limitations that you should
+consider before using it.
 
+The main one, is that if a disk fail, and you haven't recently synched,
+you may not able to do a complete recover.
+More specifically, you may be unable to recover up to the size of the
+amount of the changed or deleted files, from the last sync operation.
+This happens even if the files changed or deleted are not in the
+failed disk.
+New added files don't prevent the recovering of the already existing
+files. You may only lose the just added files, if they were on the failed
+disk.
+
+This is the reason because SnapRAID is better suited for data that
+rarely change.
+
+Other limitations are:
 * You have different filesystems for each disk.
-  Using a RAID you have only a big filesystem.
+    Using a RAID you have only a big filesystem.
 * It doesn't stripe data.
   With RAID you get a speed boost with striping.
 * It doesn't support realtime recovery.
@@ -62,7 +79,7 @@ consider before using it:
 =================
 
 To use SnapRAID you need first to mount all the disks of your disk
-array and select which one to dedicate as redundancy information.
+array and select which one to dedicate at the redundancy information.
 
 This disk will be dedicated to this pourpuse only, and you should
 not store any other data on it.
@@ -92,7 +109,7 @@ redundancy information.
     snapraid sync
 
 This process will take some hours at best the first time, depending on
-the size of the data already present in the data disks.
+the size of the data already present in the disks.
 If the disks are empty the process is immediate.
 You can stop it at any time pressing Ctrl+C, and at the next run it
 will start where interrupted.
diff --git a/testgen.c b/testgen.c
deleted file mode 100644
index e7b6b790..00000000
--- a/testgen.c
+++ /dev/null
@@ -1,76 +0,0 @@
-/*
- * Copyright (C) 2011 Andrea Mazzoleni
- *
- * This program is free software: you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation, either version 3 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#include "portable.h"
-
-void gen(int disk, int size)
-{
-	char path[PATH_MAX];
-	char name[PATH_MAX];
-	FILE* f;
-	int i;
-	int l;
-
-	l = 1 + rand() % 32;
-
-	for(i=0;i<l;++i) {
-		name[i] = 'a' + (rand() % 22);
-	}
-	name[i] = 0;
-
-	snprintf(path, sizeof(path), "test/disk%d/%s", disk, name);
-
-	f = fopen(path, "w");
-	if (!f) {
-		fprintf(stderr, "Error writing %s\n", path);
-		exit(EXIT_FAILURE);
-	}
-
-	while (size) {
-		fputc(size, f);
-		--size;
-	}
-
-	fclose(f);
-}
-
-int main(int argc, char* argv[])
-{
-	int i, j;
-	int s = 1;
-	int d = 3;
-	int n = 100;
-
-	if (argc != 4) {
-		fprintf(stderr, "Syntax: testgen SEED DISK FILE\n");
-		exit(EXIT_FAILURE);
-	}
-
-	s = atoi(argv[1]);
-	d = atoi(argv[2]);
-	n = atoi(argv[3]);
-
-	srand(s);
-
-	for(i=0;i<d;++i) {
-		for(j=0;j<n;++j) {
-			gen(i+1, j);
-		}
-	}
-
-	return 0;
-}