Tutorial CentOS 32-bit/64-bit complete dedicated server guide, chroot, startup script and newest libstdc++

Thanks for posting I just added some days ago on Steam someone having the exact same error telling me he's on a Centos submitted by OVH canada and it seems after a few tests we have made the OS differs and indeed:

A CentOS VM I have installed from scratch minimal iso here http://mirror.ovh.net/ftp.centos.org/6.5/isos/x86_64/
Aswell as the CentOS on my dedicated server both returns

Code:

[root@frpar02-games /]# uname -i
x86_64
[root@frpar02-games /]# uname -m
x86_64
[root@frpar02-games /]# uname -a
Linux frpar02-games 2.6.32-431.el6.x86_64 #1 SMP Fri Nov 22 03:15:09 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

What is this build 3.2.13-grsec-xxxx-grs-ipv6-64 ???

Can you type rpm --version ? mine is

RPM version 4.8.0

 

[root@blastoise: /] # rpm --version
RPM version 4.8.0

That's just the kernel build line, which should have very little if anything to do with RPM installations. It's the default OVH CentOS 6 image, however the packages and everything else are the same, just a more secure kernel (since they provide the image).

This is the first time I've seen this type of issue, though I've been able to run the server on Cent6 using another method.

=====EDIT=====
It would seem that I have found the problem. in my attempts to troubleshoot I changed chroot-starbound (after my initial post) to chstar to allieviate any concerns that the - in the directory would be causing issues.

However, it appears that, for some strange and peculiar reason, the rpm command is looking for more than just /var/lib/rpm. but rather theses as well

Code:

chroot("/chstar/")                      = 0
rt_sigprocmask(SIG_BLOCK, ~[RTMIN RT_1], [], 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
stat("/etc", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/", {st_dev=makedev(8, 2), st_ino=1703938, st_mode=S_IFDIR|0777, st_nlink=3, st_uid=0, st_gid=0, st_blksize=4096, st_blocks=8, st_size=4096, st_atime=2013/12/23-15:12:10, st_mtime=2013/12/23-15:11:15, st_ctime=2013/12/23-15:12:10}) = 0
stat("/etc/pki", 0x7fffd2820000)        = -1 ENOENT (No such file or directory)
stat("/etc", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/etc/pki/rpm-gpg", 0x7fffd2820000) = -1 ENOENT (No such file or directory)
stat("/etc/pki", 0x7fffd2820000)        = -1 ENOENT (No such file or directory)
stat("/etc", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/etc", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/etc/rpm", 0x7fffd2820000)        = -1 ENOENT (No such file or directory)
stat("/etc", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/etc", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/etc/yum.repos.d", 0x7fffd2820000) = -1 ENOENT (No such file or directory)
stat("/etc", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/usr/share/doc", 0x7fffd2820000)  = -1 ENOENT (No such file or directory)
stat("/usr/share", 0x7fffd2820000)      = -1 ENOENT (No such file or directory)
stat("/usr", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/usr/share/doc/centos-release-6", 0x7fffd2820000) = -1 ENOENT (No such file or directory)
stat("/usr/share/doc", 0x7fffd2820000)  = -1 ENOENT (No such file or directory)
stat("/usr/share", 0x7fffd2820000)      = -1 ENOENT (No such file or directory)
stat("/usr", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)
stat("/usr/share/doc", 0x7fffd2820000)  = -1 ENOENT (No such file or directory)
stat("/usr/share", 0x7fffd2820000)      = -1 ENOENT (No such file or directory)
stat("/usr", 0x7fffd2820000)            = -1 ENOENT (No such file or directory)

And then again farther down in the strace:

Code:

lstat("/etc/issue", 0x7fffd2820b80)    = -1 ENOENT (No such file or directory)
lstat("/etc/issue.net", 0x7fffd2820b80) = -1 ENOENT (No such file or directory)
lstat("/etc/system-release-cpe", 0x7fffd2820b80) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Base.repo", 0x7fffd2820b80) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Debuginfo.repo", 0x7fffd2820b80) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Media.repo", 0x7fffd2820b80) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Vault.repo", 0x7fffd2820b80) = -1 ENOENT (No such file or directory)
chroot(".")                            = 0
chdir("/")                              = 0
open("/chstar/var/tmp/rpm-tmp.jwPcEG", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/share/locale/en_US.UTF-8/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/share/locale/en_US.utf8/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/share/locale/en_US/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/share/locale/en.UTF-8/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/share/locale/en.utf8/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/usr/share/locale/en/LC_MESSAGES/libc.mo", O_RDONLY) = -1 ENOENT (No such file or directory)

It would appear that something in the OVH image is causing this. I've still got to crawl the other strace outfile, but this is a concern since OVH provides a cheap, reliable server.

 

Yeah basically should work but you need to find an alternative way to create a quick chroot, that problematic if you have to set a chroot manually that's just boring

The other guy I know told me he has 3.10.9-xxxx-grs-ipv6-64 and is failing at the same steps

Code:

rpm -i --root=/chroot_starbound --nodeps http://mirror.centos.org/centos/6/os/$(uname -i)/Packages/centos-release-6-5.el6.centos.11.1.$(uname -m).rpm
warning: /chroot_starbound/var/tmp/rpm-tmp.uo1sKB: Header V3 RSA/SHA1 Signature, key ID c105b9de: NOKEY
error: open of /chroot_starbound/var/tmp/rpm-tmp.uo1sKB failed: No such file or directory
/chroot_starbound/var/lib/rpm: No such file or directory

The NOKEY warning is normal everyone get it but the 2 others not, and even in -v verbose mode it does not output much

Anyway if you find how to workaround let me know I m looking too on the web what could be a worjkaround so I could add this in the FAQ because OVH servers are a lot used too

Even if for you it is the same you see why not following the standard is problematic, almost why I left OVH long time ago, always reinventing the wheel...

 

Will do. I have various communication methods (primarily IRC and jabber), PM me if you want private contact information. Currently I'm not seeing a way to get around this, but I will test it on another box and see what I can find that's different.

Also, the nokey goes away if you download the file locally and then rpm -i it. Something in the package since it's in a tmp file causes RPM to give that -I believe-. I could be wrong.

 

Error is here me I get

Code:

lstat("/etc/issue", 0x7fff8fef1960)    = -1 ENOENT (No such file or directory)
lstat("/etc/issue.net", 0x7fff8fef1960) = -1 ENOENT (No such file or directory)
lstat("/etc/system-release-cpe", 0x7fff8fef1960) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Base.repo", 0x7fff8fef1960) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Debuginfo.repo", 0x7fff8fef1960) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Media.repo", 0x7fff8fef1960) = -1 ENOENT (No such file or directory)
lstat("/etc/yum.repos.d/CentOS-Vault.repo", 0x7fff8fef1960) = -1 ENOENT (No such file or directory)
chroot(".")                            = 0
chdir("/")                              = 0
open("/chroot-test/var/tmp/rpm-tmp.rz42Ez", O_RDONLY) = 21
fcntl(21, F_SETFD, FD_CLOEXEC)          = 0
read(21, "\355\253\356\333\3\0\0\0\0\1centos-release-6-5.el6"..., 96) = 96
read(21, "\216\255\350\1\0\0\0\0\0\0\0\7\0\0\4\204", 16) = 16
read(21, "\0\0\0>\0\0\0\7\0\0\4t\0\0\0\20\0\0\1\f\0\0\0\7\0\0\0\0\0\0\2\30"..., 1268) = 1268
read(21, "\0\0\0\0", 4)                = 4
fstat(21, {st_mode=S_IFREG|0600, st_size=20816, ...}) = 0
read(21, "\216\255\350\1\0\0\0\0\0\0\0009\0\0\22l", 16) = 16
read(21, "\0\0\0?\0\0\0\7\0\0\22\\\0\0\0\20\0\0\0d\0\0\0\10\0\0\0\0\0\0\0\1"..., 5628) = 5628
rt_sigprocmask(SIG_BLOCK, ~[RTMIN RT_1], [], 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0

attaching the complete strace if you want to easy compare with yours in a tool, just used -o switch to output to file

 

Good day dev. I have a problem, the application downloads via steamcmd two days .... New update on windows it 200 + mb, but in CentOS 1GB O_O
Download application 10 kb \ s, the actual speed of my server channel 1GB \ s (server located in the data center, germany).
Why so different dimensions patch for windows and linux?
why application download so slow?

p.s.: maybe need open any special ports for download?

 

similar problem in the download method.
When I used the method that is specified in the guide ". / Steamcmd.sh + login login password + force_install_dir / starbound / + app_update 211820 validate + quit"
speed does not exceed 10kb \ s.
But when I did it in turn:
. / steamcmd.sh
login login password
force_install_dir / starbound /
app_update 211820
without "validate" and "quit", it was a good download speed.

I test the next patch and say is true or not.

p.s. Sorry my ENG, use google translate.

 

With validate is slower at start of download but then the speed grows, you can also try changing the steam download server region for this you need to edit /Root/Steam/config/config.vdf and change the "CurrentCellID" "14" to match the one you have set in your real client, for example "CurrentCellID" "14" = download server France Paris.

But my recommendation is just let SteramCMD to do its job and don't read at the current number...

 

I'm on an old version of Centos (5) and I managed to get through most of it, even changing the install to centos 5 at the beginning but steamcmd still wouldn't run (dependency issues still). Had to manually put files on there. That still didn't run either. Looks like I'm playing the waiting game.

 

what is the error you get with steamcmd ? because as far I see steamcmd is 32bit and ships with its library near the steamcmd binary

 

Hello,
i am getting the following error message:

Code:

[root@ns4009078 /]# rpm -i -v --root=/chroot-starbound --nodeps http://mirror.centos.org/centos/6/os/x86_64/Packages/centos-release-6-5.el6.centos.11.1.x86_64.rpm
Retrieving http://mirror.centos.org/centos/6/os/x86_64/Packages/centos-release-6-5.el6.centos.11.1.x86_64.rpm
warning: /chroot-starbound/var/tmp/rpm-tmp.jcalRs: Header V3 RSA/SHA1 Signature, key ID c105b9de: NOKEY
Preparing packages for installation...
error: open of /chroot-starbound/var/tmp/rpm-tmp.jcalRs failed: No such file or directory
/chroot-starbound/var/lib/rpm: No such file or directory

i tried running both commands, 1 being the one from the guide, second one is by replacing the rpm package link with the one you suggested on page 1
Sorry if i missed something stupid

 

updated the FAQ

Q: Help, I'm getting the following error when creating the chroot

Code:

warning: /chroot-starbound/var/tmp/rpm-tmp.jcalRs: Header V3 RSA/SHA1 Signature, key ID c105b9de: NOKEY
error: open of /chroot-starbound/var/tmp/rpm-tmp.jcalRs failed: No such file or directory

A: You are not running an official image of CentOS, but a modified build provided by OVH. To know if you are running an official CentOS build, type uname -a you should see something similar to 2.6.32-431.1.2.0.1.el6.x86_64 and not 3.10.9-xxxx-grs-ipv6-64 or 3.2.13-grsec-xxxx-grs-ipv6-64. Unfortunately at this time no workaround is known (I hope you understand why OVH is a bad hosting company)

 

Do you have any good alternatives for ovh, they do seem quite bad for me but i dont have any decent other cheap solutions

 

The cheapest I know is online.net they are raffling all OVH customers, French based they have true dedicated server at good hardware, unlimited bandwidth and debian ubuntu centos windows fedora ubuntu xubuntu kubuntu lubuntu

But they have high demand, took the last LT 2014 yesterday but you can run the game on the Classic too

Me i have a cheap single proc dedicated SC-GEN (€9.99) here for small services like email, dns, http and a better LT 2014 for a good gaming server

 

Sadly those are all french datacenters, our playerbase mainly consists out of american players, do you happen to know some good american ones ?

 

yup unfortunately in america you will get I think 150ms minimal ping

No I don't really know dedicated servers market in America, OVH does not offer you to choose from a variety of operating systems really ? Here we have a nice control panel where we can reformat the box under the os you want, as much as you want, that's strange they are forcing you to use an unofficial build, I don't get it

 

Thanks for the guide, it was really helpful!

I had an error with the installation, I received the following error message:

After a quick search, this page assisted to resolve: http://community.playstarbound.com/...ibraries-for-starbound_server-32-64bit.39126/

And I was able to start the server, with those additional steps.

However, I was not able to get your scripts working. I had issues with the chroot /chroot-starbound command. It kept on saying 'Directory not found'. So I referred back to your instructions, and thought that I may have missed a step.

Now I am concerned I have made a total mess of things.... When I try to start the server with Jansen's instuctions, I get:

Maybe I should start again. Apologies for my confusion, I don't know enough about linux, but it seems I know enough be dangerous. 8)

Thanks for your help.

 

Starting to think I did something incorrect... I tried to reinstall gcc... and now I am getting this error in the process.

Thanks for your assistance.

 

You cannot get the libstdc++.so.6 errors if you follow the guide because the purpose of this guide is to demonstrate how you can properly get ride of this error without modifying the game binaries

Then if you choose to mix up the "bspatch" method which I already mention to be bad (because it is modifying the binary) in the description of this guide with this guide then deal with it, it is not supported.

The second error shows you haven't installed gcc, so you did not followed the steps again because this is in the first steps after creating the chroot.

Off to bed now good luck.

 

Thanks for the quick response. Obviously I did something incorrectly during your instructions, because I received the libstdc errors.

Any recommendations to begin a clean install?