suspect mode last resort

Just type this baby in:

:g/^$/d


NOTE: This means that all the lines that just have a
carriage return on them (NO Spaces), will be removed.

Ok, so I have some of those lines too. How can I remove all of them as well?

:g/^ *$/d


NOTE: There is a space after the ‘^’ and before the ‘*’.

how to remove comments?

Let’s say that your comment lines start with the symbol #. You have a handy one-liner for sed that will get rid of them:

sed '/^\#/d' myFile > tt
mv tt myFile


sed from here
vi example from here

Usually if I repost, I try to copy the whole post, so it doesn’t get lost, but I’m feeling lazy today. I just wanted to save the link so I wouldn’t lose it (and yes, I know all about “bookmarks” in browsers…)

As an example, the connection going to bge1 is a VLAN tagged connection.
The vlan configured is 487, supporting network 192.19.20.0/24. The following
commands will initialize the new interface to support this:

/usr/sbin/ifconfig bge1 plumb
/usr/sbin/ifconfig bge1 up
/usr/sbin/ifconfig bge487001 plumb
/usr/sbin/ifconfig bge487001 192.19.20.54 netmask 255.255.255.0 broadcast + up

The first line simply creates the bge1 interface (assuming it is not
already available) with the second line bringing bge1 online, though
unconfigured. The third line creates a VLAN tagged interface on bge1
for VLAN 487. Line four configures the new interface (bge487001) to
IP address 192.19.20.54 with a class c netmask. The naming involved in
this interface is the determinable factor in which vlan the interface
is associated with. A break down of how this name is composed is:

driver-name + VLAN_ID * 1000 + device-instance

thus

bge + 487 * 1000 + 1 = bge487001

This restriction of interface naming in Solaris allows for one to
easily determine what the parent interface is (bge1) and to what VLAN
the interface functions on (487). Of note, in order to use the VLAN
tagged interface, the parent interface (bge1) must be online (up).

To bring a VLAN tagged interface up at boot time, it is handled the
same way that any other Solaris network interface would be managed.
In the above scenario, there would be a file at /etc/hostname.bge1 and
another at /etc/hostname.bge487001. The contents of hostname.bge1 would
essentially configure the interface to a null IP address and bring it up:

0.0.0.0
up

The contents of hostname.bge487001 would contain the following:

HOSTNAME
netmask 255.255.255.0 broadcast + up

named.ca is referred to as the “Root Hints” file

First, look at another host system’s root hints file, and pick a random IP address of one of the Root-Servers

In this example, we’ll use 192.36.148.17, which is currently the address of the host I.ROOT-SERVERS.NET

# cd /var/named
# dig @192.36.148.17 . ns >root_hints.new
#

Look at the file /var/named/root_hints.new that you’ve just created.
If it looks right, rename it or copy it to /var/named/named.ca.

from : here

I had occasion to expand a LUN on our spiffy new SAN. “No Problem” I thought – we’re using zfs, and after looking around, it has the “autoexpand” option. I’ll just expand the LUN, set autoexpand=on, and away I go. I even tested it on a Solaris box not in use. Worked like a champ.

So I bop over to the server in question, expand the LUN, set autoexpand=on and.. wait. what? “set” is not a valid command? ugh. looks like the zfs version included with the release of Solaris I’m running doesn’t support that command. And a reboot doesn’t fix it. and neither does zpool export/import

Enter the following…
copied from ZFS Pool Management

How to grow an existing zfs zpool from EMC LUN:

1) Perform LUN grow on array(see lun grow documentation)

2)  On server side, make sure any active zones are brought down.

3)  format(from O/S) disk with 'format' command.

Select disk.  Select 'type'.  Choose 0(auto config).

Zero out slice 6, adjust slice 0 to be full size of new disk, then relabel

4) Export and re-import pool

zpool export   Ex.  zpool export zfssanpool

5)  zpool import   Ex. zpool import zfssanpool

6)  type 'zfs list' to see that zpool is now of expanded size.

After googling for a bit, I ran across this post

Basically, it says that Solaris couldn’t mount a disk listed in the vfstab, so it’s gonna pout. Fix the vfstab, and reboot, and all should be good. Note that I’m not entirely sure that a reboot is needed, but it’s easy. Text of post below.

Yesterday, when I was working with my Solaris system, after reboot, I couldn’t ssh into the system, the ssh command failed with the “Connection refused” error, after some research (by logging into the serial console of the server and rebooting the system again through the serial console), I found the below system error message.

Therefore when you get ssh “Connection refused” error, chances are that some filesystem whose entry is in /etc/vfstab would have failed to mount, in which case check the system log, remove the line causing the problem from /etc/vfstab and ssh should work fine.

PowerShell

Use PowerShell to set the processor affinity for one or more running processes. There’s an example script below, setting the processor mask of calc.exe to the first 4 processors. I like this method because the script is simple, it would be easy to schedule, works on x86 and x64, supports multiple processes of the same name and at least partly because it highlights just how easy administration with PowerShell is.

Note that if you use factorial of a large number with calc.exe (n!) you’ll generate100% CPU which can be useful for testing. The mask below is 0xf = 1111 – a mask allowing use of only the first four processors:

$calcSet = Get-Process -ProcessName “calc”
foreach ($calc in $calcSet) {$calc.ProcessorAffinity=0xF}

EXEC sp_resetstatus 'yourDBname';
ALTER DATABASE yourDBname SET EMERGENCY
DBCC checkdb('yourDBname')
ALTER DATABASE yourDBname SET SINGLE_USER WITH ROLLBACK IMMEDIATE
DBCC CheckDB ('yourDBname', REPAIR_ALLOW_DATA_LOSS)
ALTER DATABASE yourDBname SET MULTI_USER