My notes, things I have learned, and documentation I have created.

As usual with Microsoft, it seems that have come up with a “better” idea for how to install the Remote Server Administration Tools for Windows 10 versions later than the October 2018 Update.

In the search, enter “Manage optional features” and click “Add a feature.”

And then the brilliant engineers at Microsoft, thought it would be so much better if we had install each component one at a time. So, pick and chose at your will. Click an RSAT option and then click Install. At least, they will queue. To get a status, just go back to the previous screen.

You need to reboot once you have all the features installed.

To install (or reinstall) the ActiveDirectory module, I think the only feature required it the “RSAT: Active Directory Domain Services and Lightweight Directory Services Tools”. It was not in “RSAT: Server Manager.”

Then, from a administrator Powershell windows, install the module:
PS> Install-Module ActiveDirectory

I had to re-install RSAT and the ActiveDirectory module after update to version 1909.

I have yet to come across the powershell equivalent to just install the tools. I will update this if I find it.

This is the best way to install the powershell ActiveDirectory module, because it actually works:

$JGZSession = New-PSSession -ComputerName domaincontroller
Export-PSsession -Session $JGZSession -Module ActiveDirectory -OutputModule RSATADModule
Remove-PSSession -Session $JGZSession
Import-Module RSATADModule

Install the Microsoft repository into your yum configuration:

# curl https://packages.microsoft.com/config/rhel/7/prod.repo > /etc/yum.repos.d/mssql-release.repo

Disable the repository:

vi /etc/yum.repos.d/mssql-release.repo
…
enabled=0
…

Remove the unixODBC packages if applicable:

# yum remove unixODBC-utf16 unixODBC-utf16-devel

Install the driver and command line tools (if wanted):

# yum –enablerepo packages-microsoft-com-prod install msodbcsql17
# yum –enablerepo packages-microsoft-com-prod install mssql-tools

Add the tools directory to your PATH variable as required:

vi ~/.bash_profile ~/.bashrc
…
export PATH=”$PATH:/opt/mssql-tools/bin”
..

Add the tools to your current session:

export PATH=”$PATH:/opt/mssql-tools/bin”

Test with sqlcmd:

sqlcmd -U username -P password -S server -d database

I did a little exercise to figure out a way to decode the encoded subject of some email messages.

Here is the decoded subject I was using:

Alaska is cool, go whale watching and glacier gazing (from 75%-off)

Here is the fully encoded subject of that message:

=?UTF-8?B?QWxhc2thIGlzIGNvb2wsIGdvIHdoYWxlIHdhdGNoaW5nIGFuZCBnbGFjaWVyIGdhemluZyAoZnJvbSA3NSUtb2ZmKQ==?=

Here is the command I used to decode the string:

$ echo QWxhc2thIGlzIGNvb2wsIGdvIHdoYWxlIHdhdGNoaW5nIGFuZCBnbGFjaWVyIGdhemluZyAoZnJvbSA3NSUtb2ZmKQ== | base64 -d
Alaska is cool, go whale watching and glacier gazing (from 75%-off)

Note: the leading and trailing did need to be stripped off for this to work successfully. It is perfectly within the standards to encode the email subjects like this. Unfortunately, spammers have known this for many years.

To determine a view definition:

SELECT DEFINITION FROM sys.sql_modules WHERE object_id = OBJECT_ID(‘viewschema.viewname‘)

To list all the tables in a database:

SELECT * FROM INFORMATION_SCHEMA.TABLES

To list all the columns and tables in a database:

SELECT * FROM INFORMATION_SCHEMA.COLUMNS

OUTER JOIN example:

SELECT xx.column1, yy.column2
FROM table1 xx
LEFT OUTER JOIN DB2.schema1.table2 yy ON xx.column1 = yy.column2

Change the time format:
Some CONVERT statements to the time format from DATETIME to something else:

DECLARE @SOMETIME DATETIME
SET @SOMETIME = ‘2020-01-31 11:00:00 AM’
SELECT CONVERT(VARCHAR(20), @SOMETIME, 22)
01/31/20 11:00:00 AM

This one has the four digit year:

DECLARE @SOMETIME DATETIME
SET @SOMETIME = ‘2020-01-31 11:00:00 AM’
SELECT CONVERT(VARCHAR(20), DATEADD(Hour, -8, @RECEIPTDATE), 101) + ‘ ‘ + CONVERT(VARCHAR(20), DATEADD(Hour, -8, @RECEIPTDATE), 8)
01/31/2020 11:00:00

Selecting records based on time example:
This will returns records (in this case just a bunch of time variables), if the the @SOMEDATETIME variable is between now and one hour ago:

DECLARE @SOMEDATETIME DATETIME
DECLARE @CURRENTDATETIME DATETIME
DECLARE @CURRENTDATETIMEMINUS DATETIME
SET @SOMEDATETIME = ‘2020-01-31 15:30:00.000’
SET @CURRENTDATETIME = CONVERT(VARCHAR(20), GETDATE(), 101) + ‘ ‘ + CONVERT(VARCHAR(20), GETDATE(), 8)
SET @CURRENTDATETIMEMINUS = CONVERT(VARCHAR(20), DATEADD(Hour, -1, @CURRENTDATETIME), 101) + ‘ ‘ + CONVERT(VARCHAR(20), DATEADD(Hour, -1, @CURRENTDATETIME), 8)

SELECT @SOMEDATETIME AS SOMEDATETIME
, @CURRENTDATETIMEMINUS AS CURRENTDATETIMEMINUS
, @CURRENTDATETIME AS CURRENTDATETIME
WHERE @SOMEDATETIME
BETWEEN @CURRENTDATETIMEMINUS AND @CURRENTDATETIME

OS: CentOS7

You should have your TXT record set in your DNS prior to implementing this on your server:
“v=spf1 mx ip4:aaa.bbb.ccc.ddd ip4:eee.fff.ggg.hhh -all”

Here is link I found useful to understand the options for the DNS record:
https://support.dnsimple.com/articles/spf-record/

You need to have the EPEL repository enabled to install the pypolicyd-spf package using the methodology I have outlined here.

To install and enable the EPEL repository:

# yum install https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm

Install the package Python package:

# yum install pypolicyd-spf

Modify the main.cf to add SPF verifcation to postix:
The master.cf and main.cf need to modified:

# cd /etc/postfix/
# cp -p master.cf master.cf.20190716
# vi master.cf
…
policy unix – n n – 0 spawn user=nobody argv=/bin/python /usr/libexec/postfix/policyd-spf
…

# cp -p main.cf main.cf.20190716
# vi main.cf
…
smtpd_recipient_restrictions =
permit_sasl_authenticated,
reject_unauth_destination
check_policy_service unix:private/policy
…

Note: the policy defined in the master.cf does not need to be named “policy”. However, if you change that, you need to also change it in your “check_policy_service” statement in the main.cf.

Restart postfix to implement changes:
# systemctl restart postfix

To add a permanent static route in MacOS Mojave, open up a terminal session.

List all of your available network services. Services really just seem to be aliases for network interfaces.

$ sudo networksetup -listallnetworkservices
An asterisk (*) denotes that a network service is disabled.
USB-Serial Controller D
iPad USB
iPhone USB
Wi-Fi
Bluetooth PAN
Thunderbolt Bridge

You will be adding the route to a service, therefore, you will need to know which service to apply the route to.

In my case, I wanted to add the route to my wireless interface:

$ sudo networksetup -setadditionalroutes Wi-Fi network-destination destination-mask gateway-to-use
$ sudo networksetup -setadditionalroutes Wi-Fi 10.1.1.0 255.255.255.0 192.168.0.254

List all of your current routes:

$ sudo networksetup -getadditionalroutes Wi-Fi
10.1.1.0 255.255.255.0 192.168.0.254

To remove static routes from a network service just remove the parameters from the command you used to set the route:

$ sudo networksetup -setadditionalroutes Wi-Fi

OS Version: Windows Server 2016

The only situation where you would want to do this would be if you already had an up to date and supported malware detection application installed. If you do not know or are unsure, I definitely do NOT recommend this.

Logged in as Administrator, bring up Powershell.

PS> Uninstall-WindowsFeature -Name Windows-Defender

PS> Restart-Computer -ComputerName “localhost”

I noticed a lot of the following events in a Windows event log:

Event ID: 16398 Bits-Client
A new BITS job could not be created.

Using Powershell is the best way to deal with this:

PS> Import-Module BitsTransfer

To list the entries (This should be 60 is you are getting the 16398 event, and the default has not been changed.):

PS> Get-BitsTransfer

To remove all the entries:

PS> Get-BitsTransfer -AllUsers | Remove-BitsTransfer

Using the legacy bitsadmin command:
This will display all the queue entries:

C:\> bitsadmin /List /AllUsers

You should see 60, if you are getting this message in the logs, since that is the default.

C:\> bitsadmin /AllUsers /Reset

Do a list again to verify that all have been canceled.

If that does not work, you do the following:

C:\> net stop Bits
C:\> cd C:\Users\All Users\AppData\Microsoft\Network\Downloader

Delete all the qmgr*.dat files in the directory:

C:\> del qmgr*.dat

Start the Background Intelligent Transfer Queue service:

C:\> net start Bits

Remote server: Windows 2016

I had some issues using Get-WmiObject to work on a remote server in Powershell. I kept getting: “The RPC server is unavailable.” I verified that the “Windows Management Instrumentation” service was running.

It turns out it was the firewall blocking the request.

I used the following to open up ports for the program to access the server information:

C:>netsh advfirewall firewall set rule group=”Windows Management Instrumentation (WMI)” new enable=yes

CentOS 7:

As the user on the machine where the keys are that you want to migrate:
Export the public keys:

gpg -a –export >mypublickey.asc

Export the private keys:

gpg -a –export-secret-keys > myprivatekey.asc

Export you trust database:

gpg –export-ownertrust > mytrust.db

Copy the mypublickeys.asc, myprivatekeys.asc and mytrust.db to the new machine.

Log into the new machine as the user you want the keys for:
Import the private key file:

gpg –import myprivatekey.asc

Import the public key file:

gpg –import mypublickey.asc

List the secret keys to verify:

gpg -K

List the public keys to verify:

gpg -k

Import the trustdb:

gpg –import-ownertrust mytrust.db