Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Somehow sshd stopped running and no amount of start, restart or onestart will make it go again. I normally ssh into it from a dual-boot laptop computer that shows up on the network as gabi-buntu when running Ubuntu Karmic, and as gabi-pc when running Windows XP Pro. Neither my Putty connection nor the Linux terminal can establish a ssh link anymore.

Upon rebooting the server, I am greeted with "/etc/rc: WARNING: run_rc_command: cannot run /usr/sbin/sshd". In addition, a message will appear saying things like

rpc.statd: failed to contact host gabi-buntu RPC: port mapper failure RPC: timed out

Everything else works fine. The FreeBSD 7.2 box runs a print server, a Samba server, and an Apache server for a wiki via https. It also serves up NFS shares for Linux clients.

Any suggestions?

share|improve this question
    
What does "sshd -d" give you? –  Geoff Fritz Jan 20 '10 at 18:38
1  
I would start from the warning "cannot run /usr/sbin/sshd". Check if that file is indeed there and if it is executable. You might also want to check the modification time. If the file exists, then "/usr/sbin/sshd -d" as Geoff suggested would be helpful. –  Marie Fischer Jan 20 '10 at 19:23

1 Answer 1

Geoff and Marie, I didn't get that far. I'm new here. I kept an eye on this question but it kept showing no answers, so I never actually looked at it -- which would have alerted me to your comments. I found a workaround.

First, my diagnostic:

Turns out the NFS server does care whether its clients are present. I didn't know. I assumed it just served whoever was there, looking to be served.

The "gabi-buntu" client was the Ubuntu Karmic half of a dual-boot laptop. Karmic does not suspend/hibernate on a laptop -- or at best it does, but it can't wake up, which is the same thing because it takes a cold restart. That event registers with the NFS server's rpc.statd daemon as a client crash. That's what caused the port mapper failure -- see the BUGS section of rpc.statd(8) in the FreeBSD man pages.

My workaround:

I don't really need an NFS server. Until Ubuntu figures out how to suspend/hibernate on laptops, I'll stick with Win XP and access my server share via Samba. That works fine.

So, all I had to do was fix ssh. The /usr/sbin/sshd file was there alright. I'm not sure why it didn't work, but it doesn't matter: "make -D WITH_OVERWRITE_BASE deinstall reinstall clean" of the "openssh-portable" port cured whatever ailed it and it works now.

share|improve this answer
1  
FYI, if you want to reinstall the base sshd and you have the source installed you can cd /usr/src/secure/usr.sbin/sshd ; make ; make install as root –  Chris S Apr 3 '10 at 3:20

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.