SSH hangs; -bash at 100% of CPU

C

canreo

Verified User
Joined
Mar 5, 2007
Messages
18
Recently I haven't been able to access the shell via putty. I am able to successfully enter username/password, but after that the system just hangs.

When I view the process monitor via directadmin, it shows the following:

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
10377 root 25 0 6597 1340 1100 R 98.9 0.0 0:10.74 -bash

Note that multiple attempts at logging in will continue to cause the CPU to rise until the whole server starts to slow down substantially. The only way to return to a normal load average is to kill the process via the process monitor. Also, restarting SSHD has no effect.

The only thing I've done recently was to update CURL. I don't recall if I have attempted to login after I updated CURL.
 
Are you sure the only thing that did get an update was curl? Not other updates at system level?

Are your passwords secure?

Also, not sure if is related but i see "-bash", are you sure isnt that just a bad script named -bash not related at all with the bash command interpreter?

Regards
 
SeLLeRoNe said:
Are you sure the only thing that did get an update was curl? Not other updates at system level?

Are your passwords secure?

Also, not sure if is related but i see "-bash", are you sure isnt that just a bad script named -bash not related at all with the bash command interpreter?

Regards
Click to expand...

I'm pretty sure nothing else was updated and the PWs are only known by myself. I know the "-bash" fires whenever I try to login and multiple login attempts will create more "-bash" entries on the process monitor. I had 5 going at one point and the load jumped to 14 (it's normally less than 0.5)

So far I can't SSH to update anything, or even change the root password. Any help is appreciated!
 
SeLLeRoNe said:
Are you able to access to the server not from ssh?

Regards
Click to expand...

I don't have physical access to it. The datacenter would have to do that. I was hoping it was just a configuration issue that I could fix myself without getting them involved. Note that I can access the control panel via :2222
 
Follow up on this:

I had my datacenter do a hard reboot with no effect. One of the engineers used a restore disk and noticed that the timestamp on BASH was off by a day. So he replaced the corrupted file and everything works normally now. We're not sure why it was corrupted. Does the custom build script to update CURL also update BASH? If so, that could be the culprit. Kudos to Jordan at GIPNetworks for the help.

Thanks SeLLeRoNe for trying to help me. I appreciate it!
 
You must log in or register to reply here.
Back
Top