Hangs on boot..but boots up all the way!

[AdamCoombs]

hi there,

i have a v240 that looks like it boots up normally, starts all the services, gets to "system is ready", next thing it SHOULD display is the "console login:" prompt, but instead it get
%$$£%^$^$ sort of thing and it hangs..kills the server.
it boots up fine in single user mode...

its not a hardware issue, i think all that has been covered..we have stopped the services in rc dirs but that doesn't change it..i even patched it hoping it may help..

anyone seen anything like this before?

Adam

 

[AdamCoombs]

this is solairs 9 by the way!
pls help!

 

[marrow]

Has it worked before, or is this a new system?
Might be worth running fsck on the non-mounted file systems whilst in single-user.

 

[AdamCoombs]

yer, its worked for a couple of years..
we are moving offices so on saturday it was shutdown, moved to the new building, racked and turned on...
never came back up properly...
i have done all the fsck, format-analyze-read, metasync etc etc..
no joy...

 

[marrow]

So you now have a new IP address, netmasks & default router plus a new network. Boots OK in single_user because it's not on a network. Is this the only system that's moved and failed. Can you ping the router and /or your host when you see %$$£%^$^$ message

 

[AdamCoombs]

no, its all on the same IP, subnet, default router etc etc with a link back to the current office...no network changes..
in single user mode i can ping it, as soon as i bring it up to run level 3 it dies, no network, nothing...console freezes, falls off the network..total failure...

 

[AdamCoombs]

this is the only system that failed yes..
i moved 9 others, and they all came backup without a fuss!
jsut so happens that the one that has the problem is the really important on...so surprise there then..!

 

[marrow]

check these thread60-1194077 thread60-1178740 thread60-873136

Also have you tried at boot prompt have you tried "sync

 

[AdamCoombs]

unfortunatly yes we have!
i have to say, i have now has sun engineers on site since saturday trying to fix it with no joy...

 

[Annihilannic]

What kind of console is it, graphical or serial port or other (not familiar with V240s)?

Annihilannic.

 

[AdamCoombs]

serial port...via a laptop...

 

[Annihilannic]

Have you tried disabling services in /etc/rc2.d and /etc/rc3.d one-by-one until the system boots up fine? Or probably quicker to disable most of them, and enable them or start them manually one-by-one until it fails...

Annihilannic.

 

[AdamCoombs]

yup, tried that too....made no difference,
and if you start them all up manually when in single user mode they all start fine...

 

[cndcadams]

AdamCoombs;

A couple questions?

Does /etc/hostname.bge0 look correct compared to /etc/hosts, and /etc/nodename. Did you double check /etc/inittab and make sure that looks good, this is the last line;

co:234:respawn:/usr/lib/saf/ttymon -g -h -p "`uname -n` console login: " -T sun
-d /dev/console -l console -m ldterm,ttcompat


Also when you lose your network does the link light go out on the bge port?

Thanks

CA

 

[cndcadams]

AdamCoombs;

Can you post your output from printenv for me?

I want to double check a couple things against my v240.

Are you using the alom port at all?

Thanks

CA

 

[cndcadams]

AdamCoombs;

I found this;

http://www.sunmanagers.org/pipermail/sunmanagers/2005-September/037764.html.

If the system brd has not already been replaced I would have the Sun people replace it.

I did see you booted the system single user and all seemed fine, I just think you are actually running into a hardware issue.

You would not happen to have another v240? If yes have you tried moving your boot drives into that and seeing if it boots fine?

I will wait for your responses.

Thanks

CA

 

[AdamCoombs]

hi ca..

we ruled out a hardware issue when we moved the boot disks to another v240 and it did the same thing..
i am going down to the office this morning so i can check out your other suggestion..

thanks all for your time...
i will keep you posted!
Adam

 

[ac325ci]

make sure your /etc/group memberships are correct.. also check gid's agains login / console etc..