New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
That's probably why.
CentOS 6 and OpenVZ was a horror story for us so we ended up ditching C6 and went back to Centos 5.
I would not be surprised if your VPS suddenly stopped working considering it was running on CentOS 6.
Unreliable, random crashes, data corruption...we got lucky and found these problems out before we actually deployed the servers to clients.
Interesting problem. Can you run these commands below, give the screen details please
du -h /
df -h
vmstat
uname -ar
On another note, your ps -aux should not have these:
root 1448 0.0 0.0 0 0 ? Zs 22:09 0:00 [sys]
root 1473 0.0 0.1 11232 484 ? Ss 22:10 0:00 vzctl: pts/0
root 1474 0.0 0.2 2548 1344 pts/0 Ss 22:10 0:00 -bash
Seems like your the Hostnode is accessing your vps container - even after reboot you are seeing "vzctl: pts/0" in your ps -aux ?
Cancel your vps with them. I'd already do it.
I'm accessing the vps trough openvz console.
They are not admiting that is they fault. And from a company from top3 lowendbox I was just not thinking about something like that.
Is it possible that the vps was hacked ?
Uh, from the top 3...
Hacking Is possible, but I am more inclined to the Centos 6 stuff.
@dexterx whats your df -h ? also not able to get that command. What i am suspecting is that it could possibly be the system assume that your disk is full ...somewhere or somewhat might have choke the whole filesystem - syslogs and access log errors. Thats why nothing seems running or loading on reboots. It happens when your main filesystem has no more diskspace. thats why the system cant even load basic command libraries on reboot.
My last ticket from 2 minutes ago:
@dexterx okay thats not fun at all ...seems like something else is fishy with the boot up. Next risky move is to change your runlevel on boot up ... just let me remind u, its risky but since damage is done, no harm trying that to see if you can get the basic system up with proper bash / sh running.
@dexterx ah, okay ...filesystem ...that suck. I might not be much of a help if its true due to centos 6 downgrade to centos 5 issue because i havent tried centos 6. I am not a fan of centos dist to be honest =P
Is this ubuntu?
Did you update to 11.10?
Francisco
I don't think that there is another way but reinstallation.
Don't worry, changing the settings in the second time won't take too much time.
@Francisco cannot into reading thread
@Francisco: No. It's centos, and i did nothing. No new installs or upgrades or configs on the vps in the last month. Just happend today
@dexterx are you able to get the container backup from the provider and see if we can do something about it on another node? Well just to keep some hope afloat and possibilities open, better to have that backup. The backup should be the one before they actually upgraded to centos 6 at least ?
Back up your config files if you can, and imo change host.
CentOS 6 working just fine on two of my OpenVZ vps. Seems like you might have accidentally done 'rm -rf /', as that often does trigger many of the command errors. If that is the case, you might want to move all the files off, as SSHD is still running.
afaik you cannot do 'rm -rf /' with new coreutils.
Lol, I have a VPS I don't use anymore, so I gave it a try.
The VPS doesn't have ssh running
Thanks again for the help.
I just started to reinstall everything.
One more question.
How can I backup the whole vps and then restor if needed?
I mean all the files. Not just the public_html. So I can restore installed programs and configs if something like this happens again. I should mention that i have 2 more vps with alot of space for backup.
If your VPS provider has the backup feature enabled in SolusVM, you could backup the whole system, But why backup the whole system if the core files are messed up ? Just save the configs (Unless the configs themselves are the ones which caused you this problem), reinstall the programs on the new install of the OS, so that you have a 100% clean install, which reduces the chances of this happening again.
By configs I mean installed programs. Like kloxo, ffmpeg, cron, curl, mysql tweaks and so on. For the backup from the provider I have to pay a monthly fee. The vps is not getting me any money so investing more money in it is not an option. I was thinking there is another method to backup the whole system.
Find a provider with free backups.
OR
read http://www.thegeekstuff.com/2010/09/rsync-command-examples/ or anything else from google://rsync backup
Then u will have to take note of the tweaks that u do or do a manual rsync script to another location of that specific paths and libraries.
This may be pretty obvious but I'll say it anyway
While reinstalling, keep notes (in Notepad or whatever), of what software you install (including version number), and which config files you modify. When everything is done and working correctly, copy the config files somewhere off-server. Then boil your notes down to a step-by-step recipe.
Then in a worse case scenario you can rebuild your environment using your recipe & config files in a remarkably short time
There was an option for yum (or rpm) to generate a list of installed packages.
I use git for that
Name and shame your provider.
Even if i didn't like their attitude about all this, I prefer not to mention their name. Even if I don't know how, but there is a small chance for this to be my fault (hacked vps or something like that).
Anyway many thanks to all of you who tried to help me.
Probably is unsupported anyway....
@Francisco did answer in this thread so it's probably not him. That just leaves two.
@breton thanks for the link