Search Results

If you download the software companion CD for Solaris 10 (link available on the same page you download the OS image from) it has a SFWsudo package that installs in /opt that will work.

From ALOM try this:

bootmode bootscript="-s cdrom"

From ALOM use the 'break' command to send a break during the OS boot process. That will drop you at the OK prompt. It will not work reliably during POST or OBP load, you must wait for the system to...

If you can boot from the hard drive(s) you should be able to use scadm to reset ALOM and eeprom to set OBP.

Have you tried to boot without hard drive(s) installed? If so- what happened?

I believe `pooladm` will list what you need.

GNU grep on Solaris is 'ggrep'

give me the output of `poolcfg -c info` please.

---------- Post updated at 05:32 PM ---------- Previous update was at 01:41 PM ----------

Also, do you have dynamic pools enabled? "svcs...

Edit the rules file and comment out or remove the line that references sample_host.

The man page for metainit has examples for concatenation. Additionally Sun's documentation of SVM has explicit examples for setting up a RAID 10 or 0+1.
Setting up s0 or s1 as the data slice then...

I believe Sun Management Console (/usr/sbin/smc) will give you the info you need. It ships with default Solaris installations.

closing? Do you mean clone? If so, u3 probably has it.

iostat(1M) (man Pages(1M): System Administration Commands) - Sun Microsystems (http://docs.sun.com/app/docs/doc/801-6680-1M/6i11qf503?a=view)

Native xntp on Solaris sucks. You can restart the service with `svcadm restart ntp` but it probably won't help. Try openntp.

"Stopping because process dumped core."

I honestly recommend reverting the SMF files back to normal and either writing a new one or using an rc script in rc3.d

I wonder if the LUNs are not being unmounted cleanly on reboot. Perhaps some processes were still using them. I also wonder why s2 is being referenced, are you mounting the whole disk?

Just stumbled upon what I was looking for (even though it's not needed in this case):
Solaris Fingerprint Database: An Identification Tool for Solaris Software and Files...

`man trap`

Basically trap will prevent the /etc/profile script from being canceled or killed.

First thing's first, run:

chmod -R o-rwx /usr/bin

I'll see if I can find the web page on sun.com that holds the database for file checksums and permissions.

Nope.


root@jumpstart# mount -F nfs localhost:/export/js /mnt
root@jumpstart# share -F nfs -o ro /mnt
share_nfs: Cannot share remote filesystem: /mnt
root@jumpstart#

kstat can be run without root privileges. Something as simple as `kstat -p|grep cpu` might work for your needs.

1: `w|head -1`
2: `ps -e|wc -l`
3&4: `vmstat 1 2|tail -1`
5: `ps -elfy|sort +4n -r|head -3`


EDIT:
Duke is right, `sar` is the best way to do this.

in your example d50 is the mirror, d51 and d52 are the sub mirrors.

d51 and d52 are meta-devices that reference the real slice on the disk. d50 is the meta-device that references both...

I wouldn't suggest using zfs or vxfs ;)

find / -fstype ufs -type f -name python -ls

If you're using a fully patched Solaris 10 then Zones are portable and can be cloned although the process is not as smooth and quick as VMWare's billion-dollar way of doing it (or so I'm told).