It's not often two things I love so much come together to cause me so much annoyance (besides my kids). I've written a Haskell program at work that uses libraries like text, xml-enumerator, attoparsec-text, etc. I have it working properly on my Windows machine at work, my Ubuntu virtual machine at work (32-bit), my Ubuntu desktop (32-bit again) and an EC2 instance running Ubuntu (64-bit).
Our client is running CentOS 5.3, 64-bit. I can't for the life of me get this executable to run properly. I tried creating a static executable using:
ghc --make myprog.hs -optl-static -optl-pthread
But when I try to run that executable on the CentOS server, I get an error message:
openFile: invalid argument (Invalid argument)
I'm assuming this is related to the bug described here. I've tried compiling from both 32 and 64 bit Ubuntu, tried static and shared builds, nothing works (though I occasionally get segfaults instead of the above error message). I can try downloading CentOS 5.3 and creating a virtual machine for it, but it will take a while to download, and I'm not sure which version of GHC will work on it (I tried getting GHC 7 on their server, but I ran into a libc issue).
At this point, I've come up with a few possible approaches, but I'd like to avoid these if at all possible:
Rewrite in a different language (the thought of doing this in Java makes me queasy, though it could be a nice time to try out Cal/OpenQuark).
Maybe try out an alternate compiler, like jhc. But I'm not quite certain how to get started installing all the dependencies for this program in jhc; if people have experience and know that text/attoparsec/etc work in jhc, I'd love to hear it.
Hack of all hacks: build a Windows executable, install wine on their server and run it that way.
As a total aside, these are the situations where I really wish we had a JVM backend for GHC. I suppose I could try out LambdaVM as well. But I'd love to hear community advice on what to do here.
This simple example "works for me":
$ cat A.hs
main = print "yes"
$ ghc -O2 --make -static -optc-static -optl-static A.hs -fvia-C -optl-pthread
$ ldd A
not a dynamic executable
$ ./A
"yes"
(and I've used this process, via .cabal, to ship executables for clients in the past couple of years).
I think the best bet is to file bugs, and get this working. The IHG can also fund work like this, but I'm fairly sure the GHC team would consider this a high priority, if you're trying to ship products.
It is related to the old glibc library in CentOS. You have to compile with the same version of glibc as installed on CentOS.
I had exactly the same problem. Haskell executable compiled on arch (or ubuntu) won't run on CentOS. In my case though i was lucky, because our admin just removed CentOS and installed Arch for application server.
I found out the problem. It seems that the link to the Biohaskell page is accurate: this is a problem loading iconv. This occurs when calling openFile, but not when calling openBinaryFile. Since xml-enumerator uses the latter, it worked just fine. Switching over the rest of the code to use openBinaryFile instead (via Data.Enumerator.Binary.enumFile) got everything to work.
This is a good workaround for my use case, but the bug still exists.
Related
I'm about to release a library (.so) to my client.
However, my client's platform is a very old debian(9.1 released on 2017).
My libray only works with >= glibc-2.27.
I managed to run program by the following tricks
// copy libm-2.27.so from my computer to the old debian
ln -sf ./libm-2.27.so libm.so.6
gcc ./test.c -o ./test -lmylib -L ./ -lm
LD_LIBRARY_PATH=`pwd` ./test
But my client don't accept this solution.
Is it possible to link to an older version of math library ?
For instance, the client gives me the math library and I link my library against it in my computer.
thanks!
My libray only works with >= glibc-2.27.
Is it because you actually require functionality that was added to GLIBC-2.27, or because you system just happens to have that version installed and you don't know how to build for an older system?
From the comments, it appears that you don't actually need GLIBC-2.27.
In that case, the simplest solution is to install a VM with Debian 9.1 in it, and build you library inside that VM.
This has an added advantage that you can test your library before you ship it to your client, in the environment that matches that of the client.
If you do not want a VM, other solutions are listed here.
I managed to run program by the following tricks ... But my client don't accept this solution.
Your client is smart to reject that solution: it can not generally work, and running in such environment would expose your client to a multitude of potential undefined behaviors and crashes. To understand why this is so, read this answer.
I'm learning F#, never having used any DotNet nor Mono before. I like the language, but I'm having a hard time with the myriad runtime versions and how to tell the various tools which version to use.
Summary: it appears DLL Hell on Ubuntu is a thing. How can I tell FSAC to run its auto-complete magic against the latest runtime version, namely Microsoft.NETCore.App 2.1.5, instead of whatever older incarnation it decided to use?
Longish explanation.
My platform is Ubuntu 18.04 amd64. I have installed the latest DotNet Core, specifically the packages dotnet-sdk-2.1 (version 2.1.403-1) and its dependencies dotnet-host, dotnet-runtime-2.1, etc. (which are all version 2.1.5-1). I also have Mono installed, because many tools apparently rely on it.
When I try to use some features that were recently added to DotNet Core, it feels like I'm uncovering a Pandora Box.
Take this example, which uses the new ProcessStartInfo.ArgumentList:
#r "System.Diagnostics.Process.dll"
open System.Diagnostics
let execute exe args =
let proc = new Process()
proc.StartInfo.FileName <- exe
for arg in args do proc.StartInfo.ArgumentList.Add arg
proc.Start() |> ignore
proc.WaitForExit()
execute "echo" ["foo"; "baz"]
If I naively try to run it with /usr/bin/fsharpi (which exec's /usr/bin/mono /usr/lib/mono/fsharp/fsi.exe) I get a "ArgumentList is not defined" error. Fair enough, I guess?
If I run it with /usr/bin/dotnet /usr/share/dotnet/sdk/2.1.403/FSharp/fsi.exe it works! It took me a long time to figure out I could even do it, but at last I'm on the right path.
But if I load the file in any IDE, such as VSCode with the Ionide plugin, or Sublime Text with the FSharp plugin, they still flag ArgumentList as invalid. Sublime's plugin also flags the #r line as invalid, but doesn't tell me why.
Focusing on Sublime's plugin and digging into the code, it appears that it relies on a tool called FsAutoComplete aka. fsac and on Linux explicitly invokes it with mono.
I tried running the FSAC tool with DotNet Core, but it fails to find some library or other:
$ cd .config/sublime-text-3/Packages/FSharp/fsac/fsac/
$ dotnet fsautocomplete.exe
A fatal error was encountered. The library 'libhostpolicy.so' required to execute the application was not found in '/home/tobia/.config/sublime-text-3/Packages/FSharp/fsac/fsac/'.
Failed to run as a self-contained app. If this should be a framework-dependent app, add the /home/tobia/.config/sublime-text-3/Packages/FSharp/fsac/fsac/fsautocomplete.runtimeconfig.json file specifying the appropriate framework.
If I take a filesystem approach and search for System.Diagnostics.Process.dll, I'm baffled by the results:
/usr/lib/mono/4.5/Facades/System.Diagnostics.Process.dll
/usr/lib/mono/4.7.1-api/Facades/System.Diagnostics.Process.dll
/usr/lib/mono/xbuild/Microsoft/Microsoft.NET.Build.Extensions/net461/lib/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/2.1.403/Microsoft/Microsoft.NET.Build.Extensions/net461/lib/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/microsoft.netcore.app/2.1.0/ref/netcoreapp2.1/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/netstandard.library/2.0.3/build/netstandard2.0/ref/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/lib/net46/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/lib/net461/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/ref/net46/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/ref/net461/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/ref/netstandard1.3/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/ref/netstandard1.4/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/runtimes/linux/lib/netstandard1.4/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/runtimes/osx/lib/netstandard1.4/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/runtimes/win/lib/net46/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/runtimes/win/lib/net461/System.Diagnostics.Process.dll
/usr/share/dotnet/sdk/NuGetFallbackFolder/system.diagnostics.process/4.3.0/runtimes/win/lib/netstandard1.4/System.Diagnostics.Process.dll
/usr/share/dotnet/shared/Microsoft.NETCore.App/2.1.5/System.Diagnostics.Process.dll
Plus another dozen in my home directory, under ~/.nuget. That's insane.
To begin with, the entire /usr/share/dotnet/sdk/NuGetFallbackFolder subtree does not seem to be owned by any Deb package, which is worrisome enough. The others belong to these packages:
mono-devel: /usr/lib/mono/4.5/Facades/System.Diagnostics.Process.dll
mono-devel: /usr/lib/mono/4.7.1-api/Facades/System.Diagnostics.Process.dll
msbuild: /usr/lib/mono/xbuild/Microsoft/Microsoft.NET.Build.Extensions/net461/lib/System.Diagnostics.Process.dll
dotnet-sdk-2.1: /usr/share/dotnet/sdk/2.1.403/Microsoft/Microsoft.NET.Build.Extensions/net461/lib/System.Diagnostics.Process.dll
(*) dotnet-runtime-2.1: /usr/share/dotnet/shared/Microsoft.NETCore.App/2.1.5/System.Diagnostics.Process.dll
Running a simple Grep shows that only the last one marked with (*) knows anything about ArgumentList.
So all these other libraries are older / legacy versions, including the one in dotnet-sdk-2.1 itself.
I can only presume that the SDK, aka. the compilers and interpreters, are built against an older version of the runtime / system libraries than the one they build code against. Which is an interesting choice, to say the least.
Taking a process approach, lsof tells me that the FSAC process has only opened DLL files from inside /usr/lib/mono and from its own local directory. This is not promising.
Is there a way to tell FSAC to use the Microsoft.NETCore.App 2.1.5 runtime / system libraries for its autocomplete and linting work?
(Note: names of the binary and binary and library below are obfuscated to protect the innocent. ;-) The app is proprietary under NDA but the behavior may not depend on it.)
I have a Linux binary which prints the following error when run:
binary: error while loading shared libraries: libshared.so: cannot open shared object file: No such file or directory
Which is confusing on its own since libshared.so is in the LD_LIBRARY_PATH. However,
The library is found correctly when running ldd binary (i.e., the ldd output points to the file location)
The library is found correctly when running strace binary, so that the program manages to print its usage information!
I have never seen an application which behaves differently when run on its own vs in strace, but I figure maybe someone else has seen this happen before? Any ideas how to resolve this?
I don't have the source so I can't rebuild. Running the app in production under strace is probably a non-starter. The OS is RHEL 6.2.
(Old question, but hopefully this will help somebody else)
Under new Linux installations, LD_LIBRARY_PATH is not used by the standard system runtime linker for programs with SUID set. It appears that strace, gdb and friends work differently, and do use LD_LIBRARY_PATH.
For suid programs, all libraries must be found in the system library cache. Check (as root) whether your "missing" library is present using
ldconfig -p | grep <my_library_name>
and, if anything's missing, add it to a new file in /etc/ld.so.conf or ld.so.conf.d/ as appropriate, and then rebuild using
ldconfig -v
Or remove the SUID bit if it's not required, of course.
This really helped me a lot!
I was having a similar problem where libraries were not being picked up properly from the LD_LIBRARY_PATH, even when the ldd command showed all libraries were satisfied. However, when troubleshooting using strace, it was working. In my case, however, the problem was that SGID (set group ID sticky bit being on). When the files were installed, the sysadmin did a recursive chmod that set it on all files (including the executables). Once the SGID was removed on the executables, the libraries were now found without strace and everything worked as it should using the LD_LIBRARY_PATH.
I have a a problem sounds pretty wierd. I compiled a program in Qt and it ran successfully.
But when I navigated to the release folder in the konsole and ran the program (even with super user), it reported :
If 'HelloWorld' is not a typo you can use command-not-found to lookup the package that contains it, like this:
cnf HelloWorld
I checked the permissions, all users can read it.The path is right. I'm out of way.
Please help me!
Note: I use Qt 4.74 on OpenSuse 12.1. Just switched from windows few days ago
I think this ain't relate to the source code, since even a simple HelloWorld program can't run either
If the program is in the current directory, make sure you run it with a prefix indicating so:
./HelloWorld
All,
My host supports Perl CGI scripts, how do I use a compiled CGI script on the host?
I tried setting execute permissions via chmod, but when I try and run it via the browser, I get a server error.
Thanks in advance for all help.
It's possible with a few restrictions:
You're compiling statically or don't have any external dependencies
You're using a glibc that's no newer than theirs
If you're compiling under gcc, you might also need to provide libgcc_s.so which might mean you need a LD_LIBRARY_PATH in which case you'd probably run your binary through a shell script.
You can check the dependencies on your binary with ldd. My recommendation is to compile statically with no dependencies on a system with a glibc at least as old as theirs.
NOTE: (1) above isn't strictly a requirement if you can set LD_LIBRARY_PATH and can run everything through a wrapper script, but things get much more complicated if you need such functionality.
perhaps i'm missing something, but the fact that your host supports "perl cgi" doesn't mean that they support "compiled cgi" (which i would presume is a C CGI program).
Some details about what the server error was would be helpful.