Problem at reboot 2

[yesti]

Hello, I am very new to Solaris. I had to power down two systems (Solaris 5.6 and 5.7) to install a UPS. I became superuser and typed 'shutdown -y -g0 -i0' and the systems powered down. Now when I turn the power back on one hangs at the line 'Setting default interface for multicast: add net 224.0.0.0: gateway hostname' and the other one line later at 'mount: nonexistent mount point: /tmp/opt'. What did I do wrong? Thanks.

 

[Annihilannic]

What kind of systems are they?

If they have a console keyboard attached, try using the Stop-A key combination while it's booting to drop to the "OK" prompt, then boot -s to boot into single user mode. Enter the root password when it prompts you.

If you succeed, please post the contents of your /etc/vfstab file. It sounds like you may have some rogue entries in there. /tmp/opt is not a normal mount point. Annihilannic.

 

[yesti]

Hello,

Thank you for the reply. On the machine that hung on the multicast line, I put in some echo statements and the vfstab had mount points to the machine with the /tmp/opt error. The machine with the /tmp/opt error had mount points to other machine! Classic chicken and egg. I sent an email to my boss asking why they are configured such but he is on a trip and will return this afternoon.

Is there any way for such a configuration to work? Thanks again.

 

[julianbarnett]

This is a case for using the background mount (-o bg)option for an NFS mount. A normal mount (foreground) will sit there waiting until a mount has been satisfied - hence the hanging system. A soft mount carries on and retries in the background up to the retry=n limit specified in the command line, letting the server get on with booting. See man mount_nfs.

Not a good idea to create mount points in /tmp since I thought it got cleared at reboot (it's a virtual filesystem).

JB