(with thanks to Wietse Muizelaar)
If, for whatever reason, you have to do work on a node or copy files
from one node to the other node, but the data and management
interfaces are unuseable, you can use the interconnect to access
the node from its partner.
(prerequisite)
The diag user is unlocked and has a password.
(security login unlock -username diag
security login password -username diag)
We have Node1 and Node2. Node2 has no managementnetwork anymore.
1. Node2
cl1::> net int show
(output skipped)
clus1 up/- 169.254.170.138/16 cl1-01
(output skipped)
cl1::> set advanced
cl1::*> systemshell -node cl1-02
login: diag
Password:*******
cl1-02% sudo bash
bash-3.2# vipw
(now change the usershell of the admin user from /sbin/ngsh to /usr/bin/bash)
to be able to access the node with ssh or scp, start ssh:
cl1-02% /etc/rc.d/sshd start
2. Node1
cl1::> set advanced
cl1::*> systemshell -node cl1-01
login: diag
Password:*******
cl1-01% sudo bash
scp /tmp/whateverfile admin@169.254.170.138:/tmp/
password:******
(don't forget to change back the shell of the admin-user to -ngsh
after you are done: login as admin...start ngsh....set advanced....go to systemshell as diag...
run vipw again.... 🙂 )
(another option is to run ngsh after you login as admin)