I'm trying to encrypt my data with AES algorithm with CBC mode. For this reason I use .Net Library 'Bouncy Castle'. I have not a background in crypto, so I'm trying to use it in a straightforward way. Here is my encrypt code
public byte[] encrypt(byte[] key, byte[] iv,byte[] data)
{
IBlockCipher engine=new AesFastEngine();
KeyParameter keyParam = new KeyParameter(key);
CbcBlockCipher cipher = new CbcBlockCipher(engine);
ICipherParameters parameters = new ParametersWithIV(keyParam, iv);
byte[] output=new byte[16+data.Length];
cipher.Init(true, parameters);
cipher.ProcessBlock(data, 0, output, data.Length);
//process output
byte[] cipherArray = new byte[data.Length];
/*
int k=0;
for (int i = 0; i < output.Length; i++)
{
if (output[i]!= 0)
{
cipherArray[k++] = output[i];
}
}
*/
return cipherArray;
}
When I try an input that is not a multiply of 16, I get an exception. When I pad the array to the right with a number of (16-length%16) with zeros on the left, I can get a result. But the result is a problem for me as well. It gives me a result like this:
[0][0][0][0[111][22][33][44][66][77][33][12][32][23][0][0][0][0][0]
zeros on the both left and right.
I thought it may be about my use of ProcessBlock(data, 0, output, data.Length) function. I use it with the assumption that output will be my ciphered text, but it seems that output should be longer than the input length. since I don't have a documentation about this function, I may be using it in a wrong way. Any help would be appreciated
Bouncy Castle will do the padding for you, to start you need to set up your cihper as:
PaddedBufferedBlockCipher cipher = new PaddedBufferedBlockCipher(new CbcBlockCipher(engine), new Pkcs7Padding());
For the rest of your code to work would need to use cipher.GetOutputSize(data.Length) ProcessBytes,and DoFinal so the padding is added correctly.
byte[] output = new byte[cipher.GetOutputSize(data.Length)];
int len = cipher.ProcessBytes(data, 0, data.Length, output, 0);
cipher.DoFinal(output, len);
I have a simple example on of using AES-GCM in Bouncy Castle on CodeReview
AES-GCM adds authenticated encryption, but the basic principle of using the api is the same.
I also have a C# port of high level encryption framework, Kecyzar, that I used Bouncy Castle as the backend, although it's a harder example, the abstracted encryption code SymmetricStream is setup for using AES-CBC in BouncyAesKey
Typically one would use a standard padding algorithm to ensure that plaintext data is aligned with the block-size for a cipher.
You are currently hand-coding zero padding. This is not a great choice as it forbids the original data ending in a zero byte - how can you distinguish that from the padding?
I would recommend you use a standard padding, such as PKCS #7 padding. Note that this is often referred to as "PKCS #5 padding", as they are very similar.
You may wish to refer to this other SO question - Encrypt/Decrypt using Bouncy Castle in C# - for an example of using standard padding.
Related
I need to implement custom chaining mode for encryption. I need to use symmetric block cipher (AES, 3DES, DES, IDEA). The problem I encounter is, the libraries/wrappers I found do not allow it.
BouncyCastle has those as Enum: Mode = CipherMode.CBC, so I do not see - how I can use my own. System.Security.Cryptography seem to do same way.
Is there any .NET library or wrapper, that allows custom made chaining modes?
Right now only idea I have got is to encrypt every block using CBC with IV full of zero bits and implement my chaining mode on top of it, but it does not seems as good idea.
I don't know of any library that supports a chaining callback, it sort of goes against the black-box-ness of most cryptographic API.
The way to accomplish your goal is to encrypt using ECB, since that is "just apply the encryption algorithm to this data". For example, to do CBC:
private byte[] _iv;
private ICryptoTransform _encryptor;
private void EncryptBlock(byte[] input, byte[] output)
{
byte[] buf = (byte[])input.Clone();
for (int i = 0; i < buf.Length; i++)
{
buf[i] ^= _iv[i];
}
_encryptor.TransformBlock(buf, 0, buf.Length, output, 0);
Buffer.BlockCopy(out, 0, _iv, 0, output.Length);
}
(all sorts of error checking omitted)
Given somewhere you initialized things as
using (Aes aes = Aes.Create())
{
aes.Mode = CipherMode.ECB;
aes.Key = key;
_encryptor = aes.CreateEncryptor();
_decryptor = aes.CreateDecryptor();
}
(etc).
I have the following code:
public static string PerformEncryption(string text, string uniqueKey, bool encrypt = false)
{
byte[] textBytes = encrypt ? Encoding.UTF8.GetBytes(text) : Convert.FromBase64String(text);
byte[] resultArray;
var staticKey = Convert.FromBase64String(ConfigReader.SecretKey);
using (TripleDESCryptoServiceProvider tDes = new TripleDESCryptoServiceProvider())
{
tDes.Mode = CipherMode.ECB;
tDes.Padding = PaddingMode.PKCS7;
tDes.Key = GenerateTripleDesKey(uniqueKey, staticKey);
CTransform = encrypt ? tDes.CreateEncryptor() : tDes.CreateDecryptor();
resultArray = CTransform.TransformFinalBlock(textBytes, 0, textBytes.Length);
tDes.Clear();
}
if (encrypt)
return Convert.ToBase64String(resultArray, 0, resultArray.Length);
return Encoding.UTF8.GetString(resultArray);
}
private static byte[] GenerateTripleDesKey(string uniqueKey, byte[] staticKey)
{
byte[] keyArray;
using (SHA512CryptoServiceProvider hash = new SHA512CryptoServiceProvider())
keyArray = hash.ComputeHash(Encoding.UTF8.GetBytes(string.Format("{0}{1}", uniqueKey, staticKey)));
byte[] trimmedBytes = new byte[24];
Buffer.BlockCopy(keyArray, 0, trimmedBytes, 0, 24);
return trimmedBytes;
}
PerformEncryption is used as a helper method to perform encryption/decryption of a string. A secret key is also supplied for either operation.
It is used in a web API application that is consumed by a mobile app in Android & iOS devices.
Bad Data error is occurring with large portion of the users that are on Android, with a much smaller occurance of this error on iOS. Any tests I have conducted on similar mobile devices do not produce the issue.
Only way I can reproduce error is if I modify the string value in a unit test after its encrypted.
The web API uses Async/Await so I'm not sure if this has something to do with it?
Is there anything I miss with the code above that I have left out or is bad practice?
I don't have access to the raw request being sent to the server so I can't determine if the encrypted value has its content appended with dodgy characters from Android/iOS in the request??
My other thoughts are:
should I switch from using UTF8 getbytes() to ASCII equivalent helper class if that is causing issues with iOS and android environments
Should I just switch to using a diff algorithm completely like AES.
3DES should not be used for new code, instead use AES.
Do not use ECB mode, it is insecure, see ECB mode, scroll down to the Penguin. Instead use CBC mode with a random IV, just prefix the encrypted data with the IV for use in decryption.
PBKDF2 is more secure than a simple hash or even a salted hash. The main difference is iteration in order to take more time to calculate the key,100,000 iterations is common for PBKDF2.
UTF-8 is to be preferred over ASCII which is too limited.
I'm trying to encrypt and decrypt the string using objective c and C#. both are working fine in native code, but when I was try to decrypt string in c# was encrypted in iOS. I get some error.
This was the code I used in the objective c
- (NSData *)AES256EncryptWithKey:(NSString *)key Data: (NSData *) data
{
char keyPtr[kCCKeySizeAES256+1]; // room for terminator (unused)
bzero(keyPtr, sizeof(keyPtr)); // fill with zeroes (for padding)
[key getCString:keyPtr maxLength:sizeof(keyPtr) encoding:NSUTF8StringEncoding];
NSUInteger dataLength = [data length];
NSData *iv = [#"abcdefghijklmnopqrstuvwxyz123456" dataUsingEncoding:NSUTF8StringEncoding];
size_t bufferSize = dataLength + kCCBlockSizeAES128;
void *buffer = malloc(bufferSize);
size_t numBytesEncrypted = 0;
CCCryptorStatus cryptStatus = CCCrypt(kCCEncrypt, kCCAlgorithmAES128, kCCOptionPKCS7Padding,
keyPtr, kCCKeySizeAES256,
[iv bytes] /* initialization vector (optional) */,
[data bytes], dataLength, /* input */
buffer, bufferSize, /* output */
&numBytesEncrypted);
if (cryptStatus == kCCSuccess)
{
return [NSData dataWithBytesNoCopy:buffer length:numBytesEncrypted];
}
free(buffer); //free the buffer;
return nil;
}
In want to know how to decrypt in C#, I give blocksize is 256, ivsize to 32 and used "RijndaelManaged()". I'm not using salt & password.
Error: something like "Padding is invalid and cannot be removed."
I tried to set padding too like PKCS7, none, zero but nothing help to decrypt.
can any one help this?
Edit:
My C# code here
public string DecryptString(string encrypted)
{
string result = null;
_encoder = new UTF8Encoding();
if (!string.IsNullOrWhiteSpace(encrypted) && (encrypted.Length >= 32))
{
var messageBytes = Convert.FromBase64String(encrypted);
using (var rm = new RijndaelManaged())
{
rm.BlockSize = _blockSize;
rm.Key = _encoder.GetBytes("mykey_here");
rm.IV = _encoder.GetBytes("abcdefghijklmnopqrstuvwxyz123456"); ;
rm.Padding = PaddingMode.Zeros;
var decryptor = rm.CreateDecryptor(rm.Key, messageBytes.Take(_ivSize).ToArray());
result = _encoder.GetString(Transform(messageBytes.Skip(_ivSize).ToArray(), decryptor));
}
}
return result;
}
protected byte[] Transform(byte[] buffer, ICryptoTransform transform)
{
byte[] result;
using (var stream = new MemoryStream())
using (var cs = new CryptoStream(stream, transform, CryptoStreamMode.Write))
{
cs.Write(buffer, 0, buffer.Length);
cs.FlushFinalBlock();
result = stream.ToArray();
}
return result;
}
iOS (Common Crypto) explicitly specifies all encryption parameters, the C# code implicitly determines many parameters. These implicit parameters while simplifying usage are problematic when trying to achieve interoperability.
The C# class RijndaelManaged allows explicitly specifying parameter, change your code to use these, in particular BlockSize (128), KeySize (128), Mode (CipherMode.CBC) and Padding (PaddingMode.PKCS7). The defaults for mode and Padding are OK. See RijndaelManaged Documentation
AES and Rijndael are not the same, in particular AES uses only a block size of 128 bits (16 bytes) and Rijndael allows several block sizes. So one needs to specify a block size of 128 bits for Rijndael. Thus the iv is also 128 bits (16 bytes).
Both support encryption keys of 128, 192 and 256 bytes.
You would probably be better off using the AESManaged class than the RijndaelManaged class. See AesManaged Documentation
The C# side expects the data to be Base64 encoded, the iOS side does not show that encoding operation, make sure that is being done on the iOS side.
Since you are using an iv make sure you are using CBC mode on both sides. In Common Crypto CBC mode is the default, make sure CBC mode is being used on the C# side.
Make sure the C# side is using PKCS#7 or PKCS#5 padding, they are equivalent. It appears that PKCS#7 is the default on the C# side so this should be OK.
It is best to use a key of exactly the size specified and not rely on default padding. In Common Crypto the key size is explicitly specified and null padded if the supplied key is to short. The C# looks like it is determining the key size by the supplied key, in this case the key is 10 bytes so the decryption key probably defaults to 128 bits and the key is being internally padded with nulls. On iOS you are explicitly specifying a key size of 256 bits. This is a mis-match that needs to be fixed. Supply a key that is the exact size specified on the iOS side.
Finally there is the iv, the C# code expects the iv to be prepended to the encrypted data but the iOS code is not providing that. The solution is to change the iOS code to prepend the iv to the encrypted code. Change the iv to be 16 bytes, the AES block size.
Finally provide hex dumps of the test data in, data out, iv and key just prior to and after the encryption call if you need more help.
Mysql :
SELECT AES_ENCRYPT('Test','pass')
AES_ENCRYPT() and AES_DECRYPT() enable encryption and decryption of data using the official AES (Advanced Encryption Standard) algorithm, previously known as “Rijndael.” Encoding with a 128-bit key length is used, but you can extend it up to 256 bits by modifying the source. We chose 128 bits because it is much faster and it is secure enough for most purposes.
http://dev.mysql.com/doc/refman/5.5/en/encryption-functions.html#function_aes-encrypt
I was trying to convert that Encrypted string into Decryped Strig in C#.net but i don't get the results as i expect.
http://msdn.microsoft.com/en-us/library/system.security.cryptography.rijndael.aspx#Y0
C#
static string DecryptStringFromBytes(byte[] cipherText, byte[] Key, byte[] IV)
In this method I pass ciphertext,Key value which i usedfrom Mysql and
Rijndael.Create().IV for byte[] IV
I use the code but i don't get expected result.
Review the code and comment Idk where made a mistake
What you are doing is following a road of pain. Either decrypt/encrypt on MySQL and use an encrypted connection to the database (if that matters) or encrypt/decrypt on your .NET application, storing the encrypted data in a suitable column.
Mixing AES implementations is prone to mistakes and things can break more easily if you change versions of .NET or MySQL.
Now, to know what exactly is wrong we need to know if the IV is compatible between MySQL and .NET, or else find out what is MySQL's implementation IV and supply that.
And the other potential source of problems is how you have generated the byte arrays (we are not seeing that in your example). You have to consider character encoding issues in generating the arrays if the key is textual.
In the comments of this MySQL docs link there is information about the missing parameters.
After a long hours, I found a solution to this issue.
Couple of FYI's:
MySQL as a default for AES_Encrypt uses 128 bit, with ECB mode, which does not require an IV.
What padding mode they use is not specified, but they do say they pad it. For padding I use PaddingMode.Zeros.
In C#, use AesManaged, not RijndaelManaged since that is not recommended anymore.
If your Key is longer than 128 bits (16 bytes), then use a function below to create the correct key size, since the default MySQL AES algorithm uses 128 bit keys.
Make sure you play around with the correct Encoding and know exactly what type of character encoding you will receive back when translating the bytes to characters.
For more info go here: https://forums.mysql.com/read.php?38,193084,195959#msg-195959
Code:
public static string DecryptAESStringFromBytes(byte[] encryptedText, byte[] key)
{
// Check arguments.
if ((encryptedText == null || encryptedText.Length <= 0) || (key == null || key.Length <= 0))
{
throw new ArgumentNullException("Missing arguments");
}
string decryptedText = null;
// Create an AES object with the specified key and IV.
using (AesManaged aesFactory = new AesManaged())
{
aesFactory.KeySize = 128;
aesFactory.Key = AESCreateKey(key, aesFactory.KeySize / 8);
aesFactory.IV = new byte[16];
aesFactory.BlockSize = 128;
aesFactory.Mode = CipherMode.ECB;
aesFactory.Padding = PaddingMode.Zeros;
// Create a decryptor to perform the stream transform.
ICryptoTransform decryptor = aesFactory.CreateDecryptor();
// Create the streams used for decryption.
using (MemoryStream stream = new MemoryStream())
{
using (CryptoStream decryptStream = new CryptoStream(stream, decryptor, CryptoStreamMode.Write))
{
decryptStream.Write(encryptedText, 0, encryptedText.Length);
}
decryptedText = Encoding.Default.GetString(stream.ToArray());
}
}
return decryptedText.Trim();
}
public static byte[] AESCreateKey(byte[] key, int keyLength)
{
// Create the real key with the given key length.
byte[] realkey = new byte[keyLength];
// XOR each byte of the Key given with the real key until there's nothing left.
// This allows for keys longer than our Key Length and pads short keys to the required length.
for (int i = 0; i < key.Length; i++)
{
realkey[i % keyLength] ^= key[i];
}
return realkey;
}
Here is some working code for achieving the same encryption via C# as MySQL:
public byte[] AESEncrypt(byte[] plaintext, byte[] key) {
/*
* Block Length: 128bit
* Block Mode: ECB
* Data Padding: Padded by bytes which Asc() equal for number of padded bytes (done automagically)
* Key Padding: 0x00 padded to multiple of 16 bytes
* IV: None
*/
RijndaelManaged aes = new RijndaelManaged();
aes.BlockSize = 128;
aes.Mode = CipherMode.ECB;
aes.Key = key;
ICryptoTransform encryptor = aes.CreateEncryptor();
MemoryStream mem = new MemoryStream();
CryptoStream cryptStream = new CryptoStream(mem, encryptor,
CryptoStreamMode.Write);
cryptStream.Write(plaintext, 0, plaintext.Length);
cryptStream.FlushFinalBlock();
byte[] cypher = mem.ToArray();
cryptStream.Close();
cryptStream = null;
encryptor.Dispose();
aes = null;
return cypher;
}
For details see MySQL Bug # 16713
EDIT:
Since the above is relying on officially non-documented information (though it is working) I would recommend to avoid it and use one of the options described in the answer from Vinko Vrsalovic .
If you run SELECT AES_ENCRYPT('Test','pass')
your are sending the pass over the network unencrypted so any one can unencrypted the data.
The AES_ENCRYPT is used to store data so if the database gets hacked your data is safe, not to transmit data.
if you want data encryption over the net work connect to your mysql server using the ssl socket
I am trying to encrypt data with a password and store it inside a ASN.1 encoded CMS message (using C# and BouncyCastle 1.4)
The code I have seems to have two problems:
the data does not seem to be signed with a HMAC, so when I tamper with the encodedData (by enabling the commented out line), the decryption still succeeds.
when I decrypt the data I have tampered with, I get beck corrupted plain text. However only a two blocks of plaintext data are corrupted. This seems to suggest that the encryption does not actually use CBC mode.
(edit: disregard the second point, this is exactly how CBC is supposed to work)
This is what I am testing with:
public void TestMethod1()
{
byte[] data = new byte[1024]; // plaintext: a list of zeroes
CmsEnvelopedDataGenerator generator = new CmsEnvelopedDataGenerator();
CmsPbeKey encryptionKey = new Pkcs5Scheme2PbeKey("foo", new byte[] { 1, 2, 3 }, 2048);
generator.AddPasswordRecipient(encryptionKey, CmsEnvelopedDataGenerator.Aes256Cbc);
CmsProcessableByteArray cmsByteArray = new CmsProcessableByteArray(data);
CmsEnvelopedData envelopeData = generator.Generate(cmsByteArray, CmsEnvelopedDataGenerator.Aes256Cbc);
byte[] encodedData = envelopeData.GetEncoded();
// encodedData[500] = 10; // tamper with the data
RecipientID recipientID = new RecipientID();
CmsEnvelopedData decodedEnvelopeData = new CmsEnvelopedData(encodedData);
RecipientInformation recipient = decodedEnvelopeData.GetRecipientInfos().GetFirstRecipient(recipientID);
byte[] data2 = recipient.GetContent(encryptionKey);
CollectionAssert.AreEqual(data, data2);
}
What am I doing wrong? What would be the correct way to write this?
To add an HMAC to a CMS message, you would have to use a AuthenticatedData-structure.
I am not especially familiar with Bouncy Castle, but from a cursory look at the API, I would say that it does not support AuthenticatedData. In fact, it looks like it only supports SignedData for authentication.
So your options seems to be:
Use another library (or write your own code) to handle the AuthenticatedData-structure.
Calculate the HMAC and provide it in a non-standard way (in a proprietary Attribute or out-of-band).
Use SignedData with an RSA key pair instead.