Tower Login fail at console and no web management access

[Camel]

I try to login to my tower that I just built and it just keeps throwing me back to the tower login prompt on the console and there is no connection via the ip;

 

Tower login:

 

  I put root info in, it asks for password, after I enter the password it goes to "Linux 2.6.32.9-unRAID" then "Welcome to Linux 2.6.32.9-unRIAD (tty1)" and back to "Tower login:"

 

What the heck is going on? I try admin and user creds with no luck. I was formatting the drives overnight and I cannot get into the console or web management via ip. If I force a reboot by killing the power it will reboot and I can once again login, but I cannot continually keep doing that and to have the connection drop like that is not acceptable since no data will be moving.

 

Intel D845ebg2 mobo with no bells or whistles, everything in bios is disabled except for what is necessary.

 

I am totally besides myself with this one and do not know where to begin. Please help, thanks,

Camel

[Joe L.]

I try to login to my tower that I just built and it just keeps throwing me back to the tower login prompt on the console and there is no connection via the ip;

 

Tower login:

 

  I put root info in, it asks for password, after I enter the password it goes to "Linux 2.6.32.9-unRAID" then "Welcome to Linux 2.6.32.9-unRIAD (tty1)" and back to "Tower login:"

 

What the heck is going on? I try admin and user creds with no luck. I was formatting the drives overnight and I cannot get into the console or web management via ip. If I force a reboot by killing the power it will reboot and I can once again login, but I cannot continually keep doing that and to have the connection drop like that is not acceptable since no data will be moving.

 

Intel D845ebg2 mobo with no bells or whistles, everything in bios is disabled except for what is necessary.

 

I am totally besides myself with this one and do not know where to begin. Please help, thanks,

Camel

Sounds to me like the errors in the system log used all available RAM and the kernel out-of-memory process is killing off what it thinks are processes that have been idle the longest.  The web-interface is usually the first to go, then the samba shares, then even login shells.    This happens to me if I attempt to compile something big and did not define a swap space, as I only have 512 Meg of ram in my older server. 

 

Unfortunately, the only way to get back control is to reboot, then to monitor the system log for whatever is filling it.  (unless it is some add-on you mis-configured that is filling RAM with its own stuff)   

 

Typing

tail -f /var/log/syslog

in a system console or telnet window will follow the system log.  You'll see what is filling the system log that way.

 

Joe L.

[Camel]

Thanks Joe,

 

I can keep this telnet session open until it crashes and refer to it to see what the issue is (monitoring)?

 

I think you are correct in use of available ram as I only have 512M. I hate to add more mem, can I define a swap space with another small older drive (20Meg or so)? Or should I seriously consider more mem?

 

Much appreciated and thanks for your quick reply.

 

Best,

Camel

[Camel]

I should also mantion since this happened in a middle of a format I have in my main screen;

 

 

          Stopped. Invalid configuration. Too many wrong and/or missing disks!

 

How do I start over? I assume I should just reassign all the disks since this has been a new install with no data stored yet.

 

AND, I guess I need to resolve this issue before I move forward with installing data...?

 

Thanks again,

Camel

[Camel]

I ran the "tail -f /var/log/syslog" and I am getting some errors that I am not sure about. Also my disk 1 in the main page is red with all others green. I had to run the 'initconfig' to get all the drives back to blue.

 

After I started the array I have disk 1 red and the errors in the log. How do I stop the tail -f ?

 

I did preclears on all the disks with no issues, however, I am not feeling to good about a cable or two and possibly a sata card. But I would appreciate if someone could educate me on how to decipher the log file.

 

Thanks very much,

Camel

tail_-f.zip

[dgaschk]

Enter control-c to stop tail. Unassign disk 1 and the get a SMART report.