Author Archives: Kaven G.

avatar

About Kaven G.

System Engineer / Network Administrator

Linux : How to show interface IP address at boot on the console with CentOS/RHEL7?

Particularly useful when building appliances or when you ship pre-build systems to a remote location where technical knowledge of people at the remote site is limited, you can easily make the system “print” the IP acquired by DHCP or even the static assigned one to the system console login.

You will need to create two scripts in order to have this working.

1. Create the script that will get the information from the “ifconfig” output :

2. Append the following (this example assume that the interface is eth0) :

3. Change the permissions :

4. Create the second script that will invoque the first one and then send the information to the “issue” file :

5. Append the following :

6. Change the permissions :

You can now reboot your system, it should print the IP below the system information.

Windows : Time synchornization

There might be a few situations where you need more flexibility than what the clock UI can provide. In that case, you will need some help from the command line and registry.

To manually synchronize the clock with a time server using the command line :

NOTE : Replace <ntp_server> with the server IP or FQDN.

Example :

To change the time server list/source :

HKEY_LOCAL_MACHINE \ SYSTEM \ CurrentControlSet \ Services \ W32Time \ Parameters

NOTE : Edit <peerlist>, using space between the server IP or FQDN. Make sure to append “,0x1” at the end of any FQDN record.

To set the pooling interval :

HKEY_LOCAL_MACHINE \ SYSTEM \ CurrentControlSet \ Services \ W32Time \ TimeProviders \ NtpClient \ SpecialPollInterval

NOTE : Edit <period>, value should be set in seconds.

VMware : The range of ephemeral TCP ports on this system is 1025 to 65534

Upgrading from vCenter 5.5 to 6.0? Then you might have run into the following issue during the setup :

The range of ephemeral TCP ports on this system is 1025 to 65534. This range of ports is too large and the system must be reconfigured to use a smaller ephemeral port range before the install can continue.
We recommend using the default range of 49152 to 65535 (For example, by running the command “netsh.exe int ipv4 set dynamicportrange tcp 49152 16384”

And / Or :

Installation of component VCSServiceManager failed with error code ‘1603’

Obviously, VMware find out the range of ephemeral ports to be too large. You can easily scale it down to the recommended values with the step below.

Open the Command Prompt with elevated privileges (Run as Administrator…) and enter the following command :

You should get a simple “OK” output, then you can attempt to press next or start the upgrade wizard again.

Windows : Exchange POP3 “Your e-mail Server Rejected Your User Name”

If you encountered the following error trying to connect to a POP3 account on your Microsoft Exchange mail server, this might indicate the following issues :

– The POP3 protocol isn’t enabled for this account ;
– The login type you are trying to use isn’t currently supported by the server.

This usually happen when trying to use a non-SSL connection. By default, Exchange will only accept encrypted connections. If you know what you are doing and would like to allow non-encrypted connection, simply do the following :

Warning! Applying the command(s) suggested below mean that users can authenticate sending their active directory password in clear text.

1. Open the Microsoft Exchange Shell Console with elevated privileges.

2. Enter the following command to allow unencrypted login :

NOTE : The command above signify that encryption isn’t required for authentication on port 110.

Alternatively, you can set the following :

NOTE : The command above signify that encryption isn’t required for authentication on port 110 but basic authentication is only permitted on port that use TLS or SSL encryption.

The default setting is the following :

The POP3 Exchange service must be restarted after applying any of these parameters.

Windows : magicJack cannot access the internet after upgrade from Windows 7 to 10

If you own the 1st generation magicJack and you have upgraded your computer from Windows 7 to 10, you may have noticed the following error message while starting the magicJack application :

magicJack cannot access the Internet. Please click here to start magicFix.

This message show up even though your computer is connected to the Internet. It appear that the software require an upgrade for the newer version of Windows to work.

Just get the newer magicJack client version at the following address :

http://upgrades.magicjack.com/upgrade/magicJackSetup.exe

Category: PC

Linux : perl: warning: Setting locale failed

You probably noticed the following output on your console after logging to your newly installed CloudLinux server :

perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
LANGUAGE = (unset),
LC_ALL = (unset),
LC_CTYPE = “UTF-8”,
LANG = “en_US.UTF-8”
are supported and installed on your system.
perl: warning: Falling back to the standard locale (“C”).
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
LANGUAGE = (unset),
LC_ALL = (unset),
LC_CTYPE = “UTF-8”,
LANG = “en_US.UTF-8”
are supported and installed on your system.
perl: warning: Falling back to the standard locale (“C”).

This does not represent any major issue, but can be quite annoying over time. To get rid of these notifications, do the following procedure.

1. Edit the following file using your favorite text editor :

2. Append the following :

NOTE : The file should be blank by default. You may adjust the above to meet your requirements if using different language or encoding.

Linux : CloudLinux installation error “You have third-party kernel module rpm installed”

Ending up with the following error trying to install CloudLinux?

You have third-party kernel module rpm installed.
It’s not compatible with CloudLinux kernel.
Please contact CloudLinux support at https://helpdesk.cloudlinux.com

In my case, this was because of the “kmod-vmware-tools-vmxnet3” module, since it was on a VMware guest machine.

Since removing it isn’t an option and I don’t see any good reason why this couldn’t work with CloudLinux, I decided to find a workaround to bypass the check for custom kernel modules.

Using the following installation parameter did the trick :

So let’s assume we are installing CloudLinux with a IP based license, the command would look like this :

Windows : The following error occurred attempting to join the domain “domain.tld”: The request is not supported.

Ending up with the following error trying to join a server or workstation to an existing Active Directory domain?

The following error occurred attempting to join the domain “domain.tld”:
The request is not supported.

Additionally, if you have a look at the EventViewer, you can see the following log entries under the following logging facility :

Windows Logs > System

The machine <ComputerNetBIOSname> attempted to join the domain <domain.tld> but failed. The error code was 50.
The machine<ComputerNetBIOSname> attempted to join the domain <domain.tld> but failed. The error code was 1332.

In my case, the machine was a secondary domain controller that did not deployed properly and hanged for ages on “dcpromo”. I had to force reboot the machine and of course it caused some troubles.

To resolve this matter, I did the following steps :

1. Disjoin the domain

2. Delete the computer account on the domain in “dsm.msc”

3. Open the registry editor and do the steps below.

WARNING! Do a full registry backup before doing anything. I also highly suggest that you do also backup the individual keys you are about to delete or modify.

3.1. Find the string values below and delete their content (which should match your domain name) :

3.2. Find the keys below and delete them entirely. Create new empty keys with the same name.

NOTE : In my case, these were the only one I had to remove. I used the “Find Next” function to search for my domain name in the registry, I would suggest you do the same in case you still have some left over.

4. Reboot the server.

4.1. OPTIONAL (In my case, remember I was having this issue deploying a SDC, if the same thing happened to you, then remove the Active Directory Domain role service and reboot again).

5. You may now join the computer back again on the domain.

NOTE : This procedure has been performed on Windows Server 2012 R2

JunOS : Warning aggregated-ether-options link-speed no kernel value

You probably have seen the following on your Juniper JunOS device :

dcd[15676]: ae60 : Warning: aggregated-ether-options link-speed no kernel value! default to 0
dcd[1103]: ae60 : aggregated-ether-options link-speed set to kernel value of 1000000000

This mean that your aggregated interface (ae) does not have any link-speed statement. By the book, this statement should be declared in your configuration such as :

So let’s say we have a bond composed of 2 x 10 Gbps and the interface name is ae60 :

You may now commit your configuration, the notification should disappear from the system logs.

Mac : How to uninstall LogMeIn

LogMeIn remote control software cannot be removed without traces by just removing the .app icon/folder from the Application directory like you used to do with the majority of software (it will leave the finder tray icon, will attempt to start on boot, etc.).

There is an easy way using the uninstaller script to achieve this task.

1. Open the terminal :

Go > Utilities > Terminal

2. Run the following command :

NOTE : You will be prompted for your administrator’s credentials, just enter your password and hit enter.

Category: Mac