Hi all,

For the simple smoke test, on AArch64 for AmazonLinux2 (OpenSSL 1.0.2k) and
Ubuntu 20.04 (OpenSSL 1.1.1f) everything loads ok with the current Jar that
Gary posted.

-Geoff

AL2 output:
java -cp commons-crypto-1.1.0-20200824.190246-21.jar
org.apache.commons.crypto.Crypto
Apache Commons Crypto 1.1.0-SNAPSHOT
Native code loaded OK: 1.1.0-SNAPSHOT
Native name: Apache Commons Crypto
Native built: Aug 18 2020
OpenSSL library loaded OK, version: 0x100020bf
OpenSSL library info: OpenSSL 1.0.2k-fips  26 Jan 2017
Random instance created OK:
org.apache.commons.crypto.random.OpenSslCryptoRandom@54bedef2
Cipher AES/CTR/NoPadding instance created OK:
org.apache.commons.crypto.cipher.OpenSslCipher@13221655
Additional OpenSSL_version(n) details:
1: not available
2: compiler: gcc -I. -I.. -I../include  -fPIC -DOPENSSL_PIC -DZLIB
-DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DKRB5_MIT
-DL_ENDIAN -Wall -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions
-fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches
-Wa,--noexecstack -DPURIFY -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM
3: built on: reproducible build, date unspecified
4: platform: linux-aarch64
5: OPENSSLDIR: "/etc/pki/tls"

Ubuntu output:
java -cp commons-crypto-1.1.0-20200824.190246-21.jar
org.apache.commons.crypto.Crypto
Apache Commons Crypto 1.1.0-SNAPSHOT
Native code loaded OK: 1.1.0-SNAPSHOT
Native name: Apache Commons Crypto
Native built: Aug 18 2020
OpenSSL library loaded OK, version: 0x1010106f
OpenSSL library info: OpenSSL 1.1.1f  31 Mar 2020
Random instance created OK:
org.apache.commons.crypto.random.OpenSslCryptoRandom@65b54208
Cipher AES/CTR/NoPadding instance created OK:
org.apache.commons.crypto.cipher.OpenSslCipher@119d7047
Additional OpenSSL_version(n) details:
1: compiler: gcc -fPIC -pthread -Wa,--noexecstack -Wall -Wa,--noexecstack
-g -O2 -fdebug-prefix-map=/build/openssl-9j6sUa/openssl-1.1.1f=.
-fstack-protector-strong -Wformat -Werror=format-security
-DOPENSSL_TLS_SECURITY_LEVEL=2 -DOPENSSL_USE_NODELETE -DOPENSSL_PIC
-DOPENSSL_CPUID_OBJ -DOPENSSL_BN_ASM_MONT -DSHA1_ASM -DSHA256_ASM
-DSHA512_ASM -DKECCAK1600_ASM -DVPAES_ASM -DECP_NISTZ256_ASM -DPOLY1305_ASM
-DNDEBUG -Wdate-time -D_FORTIFY_SOURCE=2
2: built on: Mon Apr 20 11:53:50 2020 UTC
3: platform: debian-arm64
4: OPENSSLDIR: "/usr/lib/ssl"
5: ENGINESDIR: "/usr/lib/aarch64-linux-gnu/engines-1.1"

On Thu, Aug 27, 2020 at 10:05 PM Matt Sicker <boa...@gmail.com> wrote:

> For a library with as many vulnerabilities as OpenSSL, I’m surprised macOS
> keeps such an ancient version! It’s not like they ship a trimmed down and
> audited version of LibreSSL, either.
>
> On Thu, Aug 27, 2020 at 20:19 Gary Gregory <garydgreg...@gmail.com> wrote:
>
> > The issue for me is that it was a PITA to override macos' baked in
> >
> > (ancient) LibreSSL.
> >
> >
> >
> > Gary
> >
> >
> >
> > On Thu, Aug 27, 2020, 20:03 Alex Remily <alex.rem...@gmail.com> wrote:
> >
> >
> >
> > > Interesting.  If I understand correctly, you did get it to run
> >
> > > successfully to completion, but only after placing a compatible
> >
> > > libcrypto in the directory of execution, probably the first place
> >
> > > dlopen looks for it.  Would you agree then that the error was caused
> >
> > > by loading an incompatible libcrypto?  I'm inclined to think this is a
> >
> > > configuration issue that should be well documented, as opposed to one
> >
> > > that should be addressed through code.  Like you, I also tried setting
> >
> > > the LD_LIBRARY_PATH environment variable with no success.  I was able
> >
> > > to symlink the libcrypto in the usr/local/lib directory, though, which
> >
> > > fixed the issue, but I agree this is a limitation.  A user should be
> >
> > > able to run more than one instance of libcrypto on the same host.  I'm
> >
> > > unsure as to the best way to proceed.
> >
> > >
> >
> > >
> >
> > > On Thu, Aug 27, 2020 at 6:41 PM Gary Gregory <garydgreg...@gmail.com>
> >
> > > wrote:
> >
> > > >
> >
> > > > On Mon, Aug 24, 2020 at 7:28 PM Alex Remily <alex.rem...@gmail.com>
> >
> > > wrote:
> >
> > > >
> >
> > > > > Gary,
> >
> > > > >
> >
> > > > > Can you check that your libcrypto.dylib is symlinked to the
> libcrypto
> >
> > > > > for OpenSSL 1.1.1.g?  Mine wasn't, and I was getting different
> output
> >
> > > > > from the main function than from the unit test output.  I'm not
> >
> > > > > confident that this is the root of the problem, but it may at least
> >
> > > > > eliminate a possibility.
> >
> > > > >
> >
> > > > > On my machine I had to set /usr/local/lib/libcrypto.dylib -->
> >
> > > > > /usr/local/Cellar/openssl@1.1/1.1.1g/lib/libcrypto.1.1.dylib.  The
> > JNI
> >
> > > > > libraries use dlopen to find and load libcrypto, and dlopen looks
> for
> >
> > > > >
> >
> > > >
> >
> > > > That did not work for me. The only thing that works is copying the
> > dylib
> >
> > > > file to the current dir. Hack!
> >
> > > >
> >
> > > > Gary
> >
> > > >
> >
> > > >
> >
> > > > > it in /usr/local/lib/, among other places.
> >
> > > > >
> >
> > > > >
> >
> > > > >
> >
> > >
> >
> https://developer.apple.com/library/archive/documentation/System/Conceptual/ManPages_iPhoneOS/man3/dlopen.3.html
> >
> > > > >
> >
> > > > > If that doesn't work I'm going to need to step through the code.
> My
> >
> > > > > output:
> >
> > > > >
> >
> > > > > WARNING in native method: JNI call made without checking exceptions
> >
> > > > > when required to from CallStaticObjectMethod
> >
> > > > > WARNING in native method: JNI call made without checking exceptions
> >
> > > > > when required to from CallObjectMethod
> >
> > > > > Apache Commons Crypto 1.1.0-SNAPSHOT
> >
> > > > > Native code loaded OK 1.1.0-SNAPSHOT
> >
> > > > > Native Name Apache Commons Crypto
> >
> > > > > Native Built Aug 24 2020
> >
> > > > > OpenSSL library loaded OK, version: 0x1010107f
> >
> > > > > OpenSSL library info OpenSSL 1.1.1g  21 Apr 2020
> >
> > > > > Random instance created OK
> >
> > > > > Cipher instance created OK
> >
> > > > > Additional OpenSSL_version(n) details:
> >
> > > > > 1: compiler: clang -fPIC -arch x86_64 -O3 -Wall -DL_ENDIAN
> >
> > > > > -DOPENSSL_PIC -DOPENSSL_CPUID_OBJ -DOPENSSL_IA32_SSE2
> >
> > > > > -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m
> >
> > > > > -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DKECCAK1600_ASM -DRC4_ASM
> >
> > > > > -DMD5_ASM -DAESNI_ASM -DVPAES_ASM -DGHASH_ASM -DECP_NISTZ256_ASM
> >
> > > > > -DX25519_ASM -DPOLY1305_ASM -D_REENTRANT -DNDEBUG
> >
> > > > > 2: built on: Tue Apr 21 13:29:43 2020 UTC
> >
> > > > > 3: platform: darwin64-x86_64-cc
> >
> > > > > 4: OPENSSLDIR: "/usr/local/etc/openssl@1.1"
> >
> > > > > 5: ENGINESDIR: "/usr/local/Cellar/openssl@1.1
> > /1.1.1g/lib/engines-1.1"
> >
> > > > >
> >
> > > > > Alex
> >
> > > > >
> >
> > > > > On Sun, Aug 23, 2020 at 9:50 PM Gary Gregory <
> garydgreg...@gmail.com
> > >
> >
> > > > > wrote:
> >
> > > > > >
> >
> > > > > > I do have LibreSSL but I used homebrew to install OpenSSL 1.1.1g
> >
> > > which I
> >
> > > > > > put first on the PATH. Maybe something is off in my setup...
> >
> > > > > >
> >
> > > > > > Gary
> >
> > > > > >
> >
> > > > > > On Sun, Aug 23, 2020, 21:46 Alex Remily <alex.rem...@gmail.com>
> >
> > > wrote:
> >
> > > > > >
> >
> > > > > > > Gary,
> >
> > > > > > >
> >
> > > > > > > I'll have a look.  I did the 1.1 support stuff and I'm familiar
> >
> > > with
> >
> > > > > > > that class and that error, although I don't recall seeing that
> >
> > > > > > > specific error in that class.  The JNI libraries check the
> > OpenSSL
> >
> > > > > > > version at runtime, but maybe a compile time dependency got
> >
> > > through.
> >
> > > > > > >
> >
> > > > > > > Out of curiosity, I assume you also have LibreSSL installed?  I
> >
> > > have
> >
> > > > > > > run into issues on my Mac with which librypto gets loaded by
> the
> >
> > > JNI
> >
> > > > > > > libraries during the dlsym.  I wonder if the runtime is
> referring
> >
> > > to
> >
> > > > > > > one version and the JNI library is loading another.
> >
> > > > > > >
> >
> > > > > > > Anyway, I'll poke around and see what I can figure out.  I'll
> try
> >
> > > to
> >
> > > > > > > get to it with the rest of the testing this week.
> >
> > > > > > >
> >
> > > > > > > Alex
> >
> > > > > > >
> >
> > > > > > > On Sun, Aug 23, 2020 at 11:18 AM Gary Gregory <
> >
> > > garydgreg...@gmail.com>
> >
> > > > > > > wrote:
> >
> > > > > > > >
> >
> > > > > > > > I wondering if anyone can confirm the following issue and/or
> > help
> >
> > > > > explain
> >
> > > > > > > > it, on MacOS 10.15.6 with OpenSSL 1.1.1g, running:
> >
> > > > > > > >
> >
> > > > > > > > mvn package
> >
> > > > > > > >
> >
> > > > > > > > then:
> >
> > > > > > > >
> >
> > > > > > > > java -Xdiag -Xcheck:jni -cp target/classes
> >
> > > > > > > > -Dcommons.crypto.lib.tempdir=target/
> >
> > > org.apache.commons.crypto.Crypto
> >
> > > > > > > > WARNING in native method: JNI call made without checking
> >
> > > exceptions
> >
> > > > > when
> >
> > > > > > > > required to from CallStaticObjectMethod
> >
> > > > > > > > WARNING in native method: JNI call made without checking
> >
> > > exceptions
> >
> > > > > when
> >
> > > > > > > > required to from CallObjectMethod
> >
> > > > > > > > Apache Commons Crypto 1.1.0-SNAPSHOT
> >
> > > > > > > > Native code loaded OK: 1.1.0-SNAPSHOT
> >
> > > > > > > > Native name: Apache Commons Crypto
> >
> > > > > > > > Native built: Aug 22 2020
> >
> > > > > > > > Exception in thread "main" java.lang.UnsatisfiedLinkError:
> >
> > > > > > > OpenSSL_version
> >
> > > > > > > >   at
> org.apache.commons.crypto.OpenSslInfoNative.OpenSSL(Native
> >
> > > > > Method)
> >
> > > > > > > >   at org.apache.commons.crypto.Crypto.main(Crypto.java:144)
> >
> > > > > > > >
> >
> > > > > > > > I wonder if we have issues on 1.1.x vs 1.0.x.
> >
> > > > > > > >
> >
> > > > > > > > My versions:
> >
> > > > > > > >
> >
> > > > > > > > openssl version
> >
> > > > > > > > OpenSSL 1.1.1g  21 Apr 2020
> >
> > > > > > > >
> >
> > > > > > > > mvn -version
> >
> > > > > > > > Apache Maven 3.6.3 (cecedd343002696d0abb50b32b541b8a6ba2883f)
> >
> > > > > > > > Maven home: /opt/apache-maven-3.6.3
> >
> > > > > > > > Java version: 1.8.0_265, vendor: AdoptOpenJDK, runtime:
> >
> > > > > > > >
> >
> > > > >
> > /Library/Java/JavaVirtualMachines/adoptopenjdk-8.jdk/Contents/Home/jre
> >
> > > > > > > > Default locale: en_US, platform encoding: UTF-8
> >
> > > > > > > > OS name: "mac os x", version: "10.15.6", arch: "x86_64",
> > family:
> >
> > > > > "mac"
> >
> > > > > > > >
> >
> > > > > > > > Thank you,
> >
> > > > > > > > Gary
> >
> > > > > > > >
> >
> > > > > > > >
> >
> > > > > > > > On Sat, Aug 22, 2020 at 7:48 PM Gary Gregory <
> >
> > > garydgreg...@gmail.com
> >
> > > > > >
> >
> > > > > > > wrote:
> >
> > > > > > > >
> >
> > > > > > > > > Hi all,
> >
> > > > > > > > >
> >
> > > > > > > > > I intent on creating a release candidate for Commons Crypto
> >
> > > soon.
> >
> > > > > > > > >
> >
> > > > > > > > > I pushed a snapshot today which contains native binaries
> for
> >
> > > > > Windows 32
> >
> > > > > > > > > and 64, Linux 32 and 64, Mac 64, and ARM and ARM HF.
> >
> > > > > > > > >
> >
> > > > > > > > > Please help testing these on whatever platforms you may
> have
> >
> > > > > access to.
> >
> > > > > > > > >
> >
> > > > > > > > > Gary
> >
> > > > > > > > >
> >
> > > > > > > > >
> >
> > > > > > > > >
> >
> > > > > > >
> >
> > > > > > >
> >
> > > ---------------------------------------------------------------------
> >
> > > > > > > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> >
> > > > > > > For additional commands, e-mail: dev-h...@commons.apache.org
> >
> > > > > > >
> >
> > > > > > >
> >
> > > > >
> >
> > > > >
> ---------------------------------------------------------------------
> >
> > > > > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> >
> > > > > For additional commands, e-mail: dev-h...@commons.apache.org
> >
> > > > >
> >
> > > > >
> >
> > >
> >
> > > ---------------------------------------------------------------------
> >
> > > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
> >
> > > For additional commands, e-mail: dev-h...@commons.apache.org
> >
> > >
> >
> > >
> >
> > --
> Matt Sicker <boa...@gmail.com>
>

Reply via email to