Git: remote end hung up unexpectedly

I have read various posts for this problem and I can verify that using -t when executing a remote ssh command does indeed force the tty to be assigned and allow the command to complete. However, the problem I'm running into is that twelve hours before that moment, I had free access to this server. Now, with no known changes, I can no longer connect.

I can log into this server all day long with no problem. However, when I try to execute a remote command say ssh servername 'ls / var / tmp', the connection is disconnected without any logged error on the server.

So what has changed?

Here is my git config:

wwwin-svn-sjc:142> git config --list
receive.denynonfastforwards=false
user.name=joericks
user.email=joericks@cisco.com
http.postbuffer=52428800000

      

        
        
        
      

    

I came across http.postbuffer at an obscene level to address this as a potential issue. I can switch to a different account and clone these repos from that server using the same URLs without issue. Other administrators and users are also unaffected. When locally to the server and using the problematic account, I can add, commit, and push to remote servers all day long with no problem.

Outside of git, I can get remote ssh commands to execute using ssh -t, but it does work and my users just won't agree to work if I can't tell them why / how it happens or what caused it. I blew my .ssh and ssh config settings. An attempt at cloning without keys caused the required password prompt and the same failure.

I confirmed that the permissions were correct for the .ssh files and parent directories:

> ls -alrt  
total 712  
-rw-r--r--  1     58 Sep 15 17:02 config  
-rw-r--r--  1 681826 Mar  7 12:24 known_hosts  
-rw-------  1   1675 Mar  7 15:22 id_rsa  
-rw-r--r--  1    405 Mar  7 15:22 id_rsa.pub  
drwx------  2   4096 Mar  7 15:23 .  
-rw-r--r--  1    405 Mar  7 15:23 authorized_keys  
drwxr-xr-x 78  24576 Mar  7 15:25 ..  

      

        
        
        
      

    

I've intentionally kept the ssh config as simple as possible:

>cat config  
ForwardX11 yes   
ForwardAgent yes  
StrictHostKeyChecking no  

      

        
        
        
      

    

Using ssh -vvv I get back with this output. (truncated for brevity)

Disconnected connection to problem server:

debug2: channel 0: open confirm rwindow 0 rmax 32768  
debug2: channel 0: rcvd adjust 2097152  
debug2: channel_input_status_confirm: type 99 id 0  
debug2: exec request accepted on channel 0  
debug2: channel 0: rcvd eof  
debug2: channel 0: output open -> drain  
debug2: channel 0: obuf empty  
debug2: channel 0: close_write  
debug2: channel 0: output drain -> closed  

      

        
        
        
      

    

Nice function server call:

debug3: packet_send2: adding 48 (len 67 padlen 13 extra_pad 64)  
debug2: we sent a password packet, wait for reply  
debug1: Authentication succeeded (password).  
debug1: channel 0: new [client-session]  
debug3: ssh_session2_open: channel_new: 0  
debug2: channel 0: send open  
debug1: Entering interactive session.  
debug2: callback start  
debug2: client_session2_setup: id 0  
debug1: Sending command: ls  
debug2: channel 0: request exec confirm 1  
debug2: fd 3 setting TCP_NODELAY  
debug2: callback done  
debug2: channel 0: open confirm rwindow 0 rmax 32768  
debug2: channel 0: rcvd adjust 2097152  
debug2: channel_input_status_confirm: type 99 id 0  
debug2: exec request accepted on channel 0  
11:43  

      

        
        
        
      

    

At this point, I am really at a loss, and unfortunately I have at least one other user who is facing this same problem. Has anyone ever figured out what exactly is causing this issue (tty allocation fails unless explicitly forced) and upon restarting the knee system, the system found a hotfix to fix the problem?

John