Not work on CentOS

[jianbingfang]

• code-server version: code-server-1.31.0-20-x86_64-linux
• OS Version: CentOS release 6.10 (Final)

Steps to Reproduce

$ ./code-server --help
./code-server: /lib64/libc.so.6: version `GLIBC_2.15' not found (required by ./code-server)
./code-server: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by ./code-server)
./code-server: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.14' not found (required by ./code-server)
./code-server: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.18' not found (required by ./code-server)
./code-server: /usr/lib64/libstdc++.so.6: version `CXXABI_1.3.5' not found (required by ./code-server)
./code-server: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.15' not found (required by ./code-server)

[kylecarbs]

Your GLIBC version is out of date. We might be able to build for older targets. Will take a look at this, thanks for the report!

[jianbingfang]

The machine is not under my control, so I'm not able to upgrade it 😢.
Thanks for your time @kylecarbs !

[ly2314]

I also have this problem too, I have a RHEL 7.6 machine but the latest version of libstdc++ I have is GLIBCXX_3.4.19, and code-server requires GLIBCXX_3.4.21.

[CCv5]

Can I rebuild it by myself on my centos machine? are there a build guildline ?

[uddmorningsun]

@jianbingfang @ly2314 @CCv5 FYI ：

• Install development tools

[yancy@localhost ~]$ docker run -ti --rm --network host centos:7.4.1708 /bin/bash

[root@localhost /]# curl -L -O  https://repo.continuum.io/archive/Anaconda3-5.0.0-Linux-x86_64.sh
[root@localhost /]# yum install -y git bzip2 gcc gcc-c++ make libX11-devel.x86_64 libxkbfile-devel.x86_64 libsecret-devel.x86_64
[root@localhost /]# bash /Anaconda3-5.0.0-Linux-x86_64.sh -b -f -p /root/conda3

[root@localhost /]# vi /root/conda3/condarc
auto_update_conda: false
channel_priority: false
override_channels_enabled: true
show_channel_urls: true
notify_outdated_conda: false

[root@localhost /]# source /root/conda3/bin/activate
(root) [root@localhost /]# conda install -y nodejs=8
... ...
nodejs:       8.12.0-he6710b0_0 defaults
libgcc-ng:    7.2.0-hcbc56d2_1  defaults --> 8.2.0-hdf63c60_1 defaults
libstdcxx-ng: 7.2.0-h24385c6_1  defaults --> 8.2.0-hdf63c60_1 defaults
... ...
(root) [root@localhost /]# conda install -y -c conda-forge yarn
... ...

• Get project and install project all dependencies using yarn and package.json

[root@localhost /]# git clone https://github.com/codercom/code-server /tmp/code-server

(root) [root@localhost scripts]# pwd
/tmp/code-server/scripts
(root) [root@localhost scripts]# yarn install
yarn install v1.13.0
[1/4] Resolving packages...
[2/4] Fetching packages...
... ...

(root) [root@localhost scripts]# bash -x build.sh
+ set -e
+ yarn task build:server:binary
yarn run v1.13.0
$ ts-node -r tsconfig-paths/register build/tasks.ts build:server:binary
... ...

• Check with ldd or use CentOS to start it ! For examples: https://img.vim-cn.com/6a/d3f5cf3b9f350eb873fa276038d08de0b5dd36.png

[root@localhost server]# pwd
/tmp/code-server/packages/server
[root@localhost server]# ./cli-linux-x64 --version
server/0.0.0 linux-x64 node-v8.12.0

[seuliang]

• CentOS

Thank you！
Could you explain the last step in detail？
When I run ./cli-linux-x64 --version , i get segmentation fault (core dumped).
How can i fix this?

[uddmorningsun]

@seuliang Maybe you should:

• Use gdb your/command and then print stack info(if enable debug), then reserch related error
• Or LD_DEBUG=libs your/command see which library cannot load successfully

[vinayak-gadkari]

The build indeed worked on my RHEL system. Thanks!
System: Host: ***** Kernel: 3.10.0-693.2.2.el7.x86_64 x86_64 bits: 64 Console: tty 0
Distro: Red Hat Enterprise Linux Server release 7.6 (Maipo)

[activatedgeek]

I wonder why the binary does not respect LD_LIBRARY_PATH?