LinuxQuestions.org - Problem with vServer, SSH doesn't start

- Linux - Newbie (https://www.linuxquestions.org/questions/linux-newbie-8/)

- - Problem with vServer, SSH doesn't start (https://www.linuxquestions.org/questions/linux-newbie-8/problem-with-vserver-ssh-doesnt-start-686434/)

Problem with vServer, SSH doesn't start

Hi,

I've got Ubuntu 8.04 LTS running on a vServer and last week my server hoster has updated the kernel and (hence rebooted the server) and since then my server doesn't boot correctly anymore.

That means:
Ping works.
SSH does not. (Port is closed)
Apache does not. (Port is closed)
POP3 and Imap ports are open and I can connect to pop3 via telnet (didn't test anything else).

I've got no access to the server, there is only a "rescue console" with which I can access and edit the files, check the log files and restart the server.

The only strange thing I could find in the log files is:

Code:

Nov 26 19:43:28 ... postfix/tlsmgr[8130]: warning: cannot open entropy device /dev/urandom: No such file or directory

Nov 26 19:43:28 ... postfix/tlsmgr[8130]: exiting to reopen external entropy source dev:/dev/urandom

Nov 26 19:43:29 ... postfix/tlsmgr[8151]: warning: cannot open entropy device /dev/urandom: No such file or directory

Nov 26 19:43:29 ... postfix/tlsmgr[8151]: exiting to reopen external entropy source dev:/dev/urandom

Nov 26 19:43:30 ... postfix/tlsmgr[9483]: warning: cannot open entropy device /dev/urandom: No such file or directory

Nov 26 19:43:30 ... postfix/tlsmgr[9483]: exiting to reopen external entropy source dev:/dev/urandom

I think it's probably related to why SSH doesn't work.

Any ideas? Is there a possibility to make sshd log why it doesn't start?

first off check to see if any port blocking has been added to iptable. this can be done by typing in the following.

Code:

iptables -L

Also post your syslog, and what happens when you try and start the ssh dameon. And last ask you server hoster what the hell they were doing touching your vm unless when you signed up they told you that is something they would do.

Well... the problem is that I don't have any access to the server *because* ssh isn't starting! Hence, I can't try to start ssh and also I can't check iptables.

The "rescue console" only works while the server is not running. I've got access to the files but to test the changes I've got to quit the rescue mode and restart the server.

The quote above is from the syslog. There is nothing from SSHD. How can I make ssh write something helpful into the logs? Currently, it only logs login attempts.

Quote:

And last ask you server hoster what the hell they were doing touching your vm unless when you signed up they told you that is something they would do.

Well, they've upgraded the kernel on their host OS... actually that should not change anything for my VM. Probably there already was something bad in my VM and it was the restart that made it go havoc.

Quote:

Originally Posted by goldmar (Post 3357135)

How can I make ssh write something helpful into the logs? Currently, it only logs login attempts.

If the rescue console allows you to modify sshd_config, you might consult its man page . I particularly direct your attention to the keyword LogLevel. Please be aware that the arguments to keywords are case sensetive.

Since you have access to the files, can't you just 'mknod -m 644 /dev/random c 1 8; mknod -m 644 /dev/urandom c 1 9; chown root:root / dev/random /dev/urandom'?