* Support ED25519 PuTTY keys.
Fix#659
* PuTTYKeyFile: Use net.schmizz.sshj.common.Buffer instead of own KeyReader.
A tiny refactoring made in order to allow usage of other utility methods which require Buffer.
* Support ECDSA PuTTY keys.
* Some code cleanup
Co-authored-by: Jeroen van Erp <jeroen@hierynomus.com>
* Loop through security providers to check for BC
Instead of only counting BouncyCastle as being registered if it
is set as the explicit security provider used by SSHJ, count it as
registered if it is available as a provider.
This commit improves Android compatibility, which requires not
specifying an explicit provider.
* Generify BC-specific curve specifiers
The ECNamendCurveGenParameterSpec is a BC-specific workaround for
missing curve tables in Java 1.4 and earlier. For the sake of Android
compatibility, where Conscrypt can't deal with this custom spec class,
replace it with the standard ECGenParameterSpec and update the curve
names to the standard identifiers.
* Implement AES-GCM cipher support
Fixes#217.
A port of AES-GCM cipher support from Apache MINA-SSHD, based on https://github.com/apache/mina-sshd/pull/132.
Included tests for decoding SSH packets sent from Apache MINA-SSHD and OpenSSH (Version 7.9p1 as used by Debian 10).
Manual tests also done on OpenSSH server 7.9p1 running Debian 10 with its available ciphers, including 3des-cbc, aes128-cbc, aes192-cbc, aes256-cbc, aes128-ctr, aes192-ctr, aes256-ctr, aes128-gcm@openssh.com and aes256-gcm@openssh.com.
* Changes per PR feedback
- Fixed variable/statement whitespaces and add back missing braces per coding standard requirement
- Moved Buffer.putLong() and Buffer.getLong() into GcmCipher.CounterGCMParameterSpec since it's the only user
- Moved BaseCipher.authSize into GcmCipher since it is the only cipher that would return a non-zero. BaseCipher will keep return 0 instead
- Made BaseCipher.cipher protected instead of making it publicly accessible
- Combined the three decoding modes in Decoder.decode() into one single method, to reduce code duplication
- Added integration test for the ciphers, along with the newly implemented AES-GCM ciphers
In some cases, current code will leak parts or even the whole ssh key if it's slightly malformed.
One example of that malformation will be a key, where all newlines are replaced by other character, thus turning a multiline key to a single big string.
Then that whole line will be leaked to exception message.
* Fix RSA certificate key determination.
Fixes#599.
* Correct serialization of RSA certificates with unlimited dates.
* The test for connecting with RSA certificate.
* Remove redundant change in TransportImpl.java
* Add forgotten test keys.
* Make net.schmizz.sshj.common.KeyType.CertUtils.epochFromDate readable.
Co-authored-by: Vladimir Lagunov <vladimir.lagunov@jetbrains.com>
Co-authored-by: Jeroen van Erp <jeroen@hierynomus.com>
* Fix matching of pubkeys to key algorithms
Allow all configured key algorithms for pubkey authentication, even if
these algorithms are not supported as host key algorithms by the
server.
Preference is given to the modern rsa-sha2-* signature algorithms if
the server indicates support for them as host keys signature
algorithms.
* Replace Boolean with primitive boolean
* Add integration tests for ecdsa-sha2-nistp384/521
* Remove redundant import
* Clean up Transport interface
Co-authored-by: Jeroen van Erp <jeroen@hierynomus.com>
SignatureECDSA.encode() does not correctly handle signatures longer
than 128 bytes, which affects signatures using the nistp521 curve.
This commits fixes the issue by replacing the ad-hoc ASN.1 DER
parsing with a use of ASN1InputStream.
Some SSH servers will not honor the negotiated rsa-sha2-256 algorithms
if the client does not indicate support for SSH_MSG_EXT_INFO messages.
Since we only need to accept these messages, but are free to ignore
their contents, adding support amounts to sending "ext-info-c" with our
kex algorithm proposal.
* Don't specify JcaPEMKeyConverter provider as null
If no provider is set in the `SecurityUtils`, no named provider should be set for the `JcaPEMKeyConverter` as this would cause a `missing provider` exception.
* Don't specify JcePEMDecryptorProviderBuilder provider as null
If no provider is set in the `SecurityUtils`, no named provider should be set for the `JcePEMDecryptorProviderBuilder` as this would cause a missing provider exception. This currently breaks `PKCS8KeyFile` if `SecurityUtils.setSecurityProvider(null)` and `SecurityUtils.setRegisterBouncyCastle(false)` is used.