Centos on Kenno's Open Note

How to Install AWS CLI on RockyLinux

Tue, 31 Jan 2023 12:02:59 +1100

The awscli package is available from the EPEL repository, which is not enabled by default on RockyLinux.

Enable the EPEL repo by installing epel-release pacakge.

$ sudo dnf install epel-release

Once the EPEL repo is successfully enabled, we can install awscli package.

$ sudo dnf install awscli

Enable and verify that the awscli is installed on the system.

$ aws --version
aws-cli/1.18.156 Python/3.6.8 Linux/4.18.0-425.10.1.el8_7.x86_64 botocore/1.18.15

As a bonus, here is how to configure the AWS credential.

Unable to Use tar to Decompress tar.bz2 File

Sun, 25 Jul 2021 17:18:43 +1000

I just setup a new server with a minimal installation of AlmaLinux 8. AlmaLinux is a variant of free Red Hat Enterprise Linux 8. It’s like a CentOS 8, but will have a full support until the end of RHEL 8 life cycle.

This server is to serve as a new proxy for all VMs inside my home lab. I want to use tinyproxy which is need to be compiled from source as the RPM package is not available as of the time writing. After downloading the source code, and try to decompress it, I was presented with the following error message:

Systemd List Failed Units and Services

Tue, 15 Dec 2020 16:49:08 +1100

This is a quick note to list failed services on systemd system.

[root@zilla ~]# systemctl status
● zilla
    State: degraded
     Jobs: 0 queued
   Failed: 1 units
    Since: Fri 2020-09-04 02:20:47 AEST; 3 months 11 days ago
   CGroup: /
           ├─1 /usr/lib/systemd/systemd --system --deserialize 22
           └─system.slice
             ├─httpd.service
             │ ├─21860 /usr/sbin/httpd -DFOREGROUND
             │ ├─21861 /usr/sbin/httpd -DFOREGROUND
             │ ├─21862 /usr/sbin/httpd -DFOREGROUND
             │ ├─21863 /usr/sbin/httpd -DFOREGROUND
             │ ├─21864 /usr/sbin/httpd -DFOREGROUND
             │ └─21865 /usr/sbin/httpd -DFOREGROUND
...

[root@zilla ~]# systemctl list-units --failed
  UNIT         LOAD   ACTIVE SUB    DESCRIPTION
● ipmi.service loaded failed failed IPMI Driver

LOAD   = Reflects whether the unit definition was properly loaded.
ACTIVE = The high-level unit activation state, i.e. generalization of SUB.
SUB    = The low-level unit activation state, values depend on unit type.

1 loaded units listed. Pass --all to see loaded but inactive units, too.
To show all installed unit files use 'systemctl list-unit-files'.

Configure iSCSI Target on CentOS 8

Wed, 25 Nov 2020 19:04:38 +1100

I need to have an iSCSI storage to be used by my oVirt lab. The last time I had to create an iSCSI target was during the preparation of the RHCE 7 exam. Therefore, I think this is an opportunity to practice doing this again.

First, install the required package. Then enable the target service, otherwise the target won’t work on the next reboot.

[root@utility ~]# dnf install targetcli -y

[root@utility ~]# systemctl enable target

Run the targetcli command to create an iSCSI target/storage using the storage block (/dev/mapper/data_iscsi LVM volume).

How to Enable Persistent Logging for Systemd Journal

Sat, 29 Aug 2020 11:01:31 +1000

On RHEL 7/8, CentOS 7/8 and even Ubuntu (??), by default the journal log data is stored only in memory (/run/log/journal/ directory).

There are 2 ways to retain the journal log messages. The first one is to set the variable Storage to persistent in the /etc/systemd/journald.conf.

[Journal]
Storage=persistent

Then restart the systemd-journald service.

Another solution is simpler and it looks like it’s the recommended way of achieving this. All we have to do is to create a directory, /var/log/journal (with correct ownership and permission), and journald will automatically store the log messages there.

SELinux Error ValueError: Port already Defined

Mon, 27 Jul 2020 14:44:53 +1000

Today, I want to run the httpd server on a custom port, 8002 on a CentOS 7 server. Of course, all my system has SELinux in enforcing mode.

In order for the httpd to bind on port 8002, we need to add this port to the httpd_port_t port type list.

First let’s confirm that port 8002 is not yet in the http_port_t list:

[root@servera ~]# semanage port -l | grep ^http
http_cache_port_t              tcp      8080, 8118, 8123, 10001-10010
http_cache_port_t              udp      3130
http_port_t                    tcp      80, 81, 443, 488, 8008, 8009, 8443, 9000

We know from the manual semanage-port(8), to add a port to an SELinux port type list, we run the following command:

How to Install Ansible on Centos 8

Sat, 09 Nov 2019 10:06:03 +1100

On most Linux distribution, Ansible can be installed either using python3-pip or ansible package.

I’m going to show how to get Ansible installed using the second method on CentOS 8.

First add the EPEL repository to the CentOS 8 server.

# yum localinstall https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm

Next, install ansible package:

# yum install ansible
Last metadata expiration check: 0:00:49 ago on Sun 10 Nov 2019 09:33:22 PM AEDT.
Dependencies resolved.
=======================================================================================================================
 Package                           Arch                  Version                        Repository                Size
=======================================================================================================================
Installing:
 ansible                           noarch                2.8.5-2.el8                    epel                      15 M
Installing dependencies:
 python3-babel                     noarch                2.5.1-3.el8                    AppStream                4.8 M
 python3-jinja2                    noarch                2.10.1-2.el8_0                 AppStream                538 k
 python3-jmespath                  noarch                0.9.0-11.el8                   AppStream                 45 k
 python3-markupsafe                x86_64                0.23-19.el8                    AppStream                 39 k
 python3-pyasn1                    noarch                0.3.7-6.el8                    AppStream                126 k
 libsodium                         x86_64                1.0.18-2.el8                   epel                     162 k
 python3-bcrypt                    x86_64                3.1.6-2.el8.1                  epel                      44 k
 python3-pynacl                    x86_64                1.3.0-5.el8                    epel                     100 k
 sshpass                           x86_64                1.06-9.el8                     epel                      27 k
Installing weak dependencies:
 python3-paramiko                  noarch                2.4.3-1.el8                    epel                     289 k

Transaction Summary
=======================================================================================================================
Install  11 Packages

Total download size: 21 M
Installed size: 104 M
Is this ok [y/N]:

Let’s confirm that the installation went ok by checking the version of ansible:

Connect to a WPA2-Enterprise Connection Using nmcli

Mon, 27 Aug 2018 14:11:38 +0000

This is like a re-post of a question from askubuntu.com. To get to that post, I usually had to google it. So, I’ll post it here so I know where to look for when I need it again.

At work, we use WPA2-Enterprise for wireless connection. To create a new connection on my laptop with wlp3s0 as Wifi device, follow this step:

# nmcli con add type wifi ifname wlp3s0 con-name work-wifi ssid work-ssid

# nmcli con edit id work-wifi
nmcli> set ipv4.method auto
nmcli> set 802-1x.eap peap
nmcli> set 802-1x.phase2-auth mschapv2
nmcli> set 802-1x.identity myusername
nmcli> set 802-1x.password mypassword
nmcli> set wifi-sec.key-mgmt wpa-eap
nmcli> save
nmcli> activate

That’s it if you know how to do it.

How to Generate SELinux Policy Manual Page

Sun, 26 Aug 2018 23:21:33 +0000

It seems that by default, SELinux policy manpage is not available. Here’s how to make it available.

First install selinux-policy-devel package
Then generate the man page by running:
# sepolicy manpage -a -p /usr/share/man/man8<br/>
-a : all domains
-p : path to store the manpage (by default it’s /tmp)
/usr/share/man/man8 – section 8 of manpage is for system admin command
Create or update manual page index caches
# mandb
Benefit
# man -k _selinux (or e.g. ”man httpd_selinux“)

RHEL Recovering the Root Password

Sun, 12 Aug 2018 14:24:05 +0000

Early last year I started a blog post about the preparation for Red Hat Certified System Administrator (RHCSA) exam. My intention back then was to have more posts follow suit about the exam topics. Like many things in life, things don’t always go according to plan. That was the only post, but I passed the RHCSA exam.

Today, I’m going to start another new post about the preparation for the Red Hat Certified System Engineer (RHCSE) exam. I already booked an exam which is taking place on September 12. The exam is 3.5 hours long and cover a broader range than before.

Install MATE Desktop on CentOS 7

Wed, 13 Jun 2018 01:26:00 +0000

Tonight I started to review the learning material for Red Hat System Administration III to prepare for RHCSE exam. I need a machine with GUI, and all my CentOS 7 VMs are headless. So let’s install the GUI on it, shall I?

First install epel repo, if it’s not yet installed:

# yum install epel-release -y

Then install the 2 groups:

# yum groupinstall "X Window system" -y
# yum groupinstall "MATE Desktop" -y

Tip, if “I” can’t remember the names of the groups, I can run this command to list all available groups:

grep: error while loading shared libraries: libpcre.so.0

Fri, 01 Jul 2016 14:14:13 +0000

I have a server running CentOS 6 successfully upgraded to CentOS 7.

The only problem I encountered so far is that ‘grep’ doesn’t work.

[root@zilla ~]# grep
grep: error while loading shared libraries: libpcre.so.0: cannot open shared object file: No such file or directory

[root@zilla ~]# ldd /usr/bin/grep
        linux-vdso.so.1 =>  (0x00007ffc0b639000)
        libpcre.so.0 => not found
        libc.so.6 => /lib64/libc.so.6 (0x00007f6be1c6c000)
        /lib64/ld-linux-x86-64.so.2 (0x00007f6be2038000)

First thing I tried was to reinstall pcre package which provides libpcre.so*. It still didn’t work. Fortunately, someone has posted a solution as the following:

ISC Dhcpd Duplicate UID Lease

Thu, 25 Jun 2015 14:10:51 +0000

Someone asked this question in one of his blog posts:

Ever see this?

dhcpd: uid lease 192.168.1.150 for client xx:xx:xx:xx:xx:xx is duplicate on 192.168.1/24

Yes, I’m seeing something like that now in the my dhdcp logs. Here is the actual message:

DHCPREQUEST for 192.168.1.109 from 08:00:27:6e:61:b8 via eth0
DHCPACK on 192.168.1.109 to 08:00:27:6e:61:b8 via eth0
DHCPINFORM from 192.168.1.109 via eth0
DHCPACK to 192.168.1.109 (08:00:27:6e:61:b8) via eth0
uid lease 192.168.1.224 for client 08:00:27:6e:61:b8 is duplicate on 192.168.1.0/24
DHCPREQUEST for 192.168.1.109 from 08:00:27:6e:61:b8 via eth0
DHCPACK on 192.168.1.109 to 08:00:27:6e:61:b8 via eth0

I think I know why this happened. Basically, the machine with MAC address of ’08:00:27:6e:61:b8′ used to have grab an IP address

YUM what provides this file, huh?

Wed, 24 Jun 2015 17:08:20 +0000

This post is primarily written to remind myself, an Ubuntu user, to figure out how to find what RPM package provides a file — for example a header file.

Today, I’m trying to compile and install R 3.1.3 on CentOS 6 from source code. This R installation is one of the requirements to install Tessera stack. The instruction from the installation page is:

sudo yum install blas blas-devel lapack lapack-devel
wget http://cran.r-project.org/src/base/R-3/R-3.1.3.tar.gz
tar zxvf R-3.1.3.tar.gz
cd R-3.1.3
./configure --with-blas --with-lapack --with-x --enable-R-shlib
make
sudo make install

Unfortunately, the process failed with ./configure with the complain about missing readline library. To fix it, I had to install readline-devel package.

Good permission for /tmp

Mon, 16 Feb 2015 12:44:18 +0000

I have been having issues getting X2Go running on a CentOS 6.5 box. It took me a while to figure out that the problem laid with the permission of /tmp directory. I’ll write how to solve the X2Go problem in another post.

The recommended permission for /tmp is 1777. Notice that the forwarding ‘1’ is the sticky bit which only allows the owner of the file within /tmp (or the root user) to delete/rename the file.

Can't find the 'libpq-fe.h header

Mon, 03 Feb 2014 12:59:10 +0000

I need to install pg 0.17.1 gem on a server running CentOS. This is the error message I saw:

# gem install pg -v '0.17.1'
Building native extensions.  This could take a while...
ERROR:  Error installing pg:
        ERROR: Failed to build gem native extension.

/usr/bin/ruby extconf.rb
checking for pg_config... no
No pg_config... trying anyway. If building fails, please try again with
 --with-pg-config=/path/to/pg_config
checking for libpq-fe.h... no
Can't find the 'libpq-fe.h header
*** extconf.rb failed ***
Could not create Makefile due to some reason, probably lack of
necessary libraries and/or headers.  Check the mkmf.log file for more
details.  You may need configuration options.
...

First thing came to my mind was I must have forgotten to install the -dev package.

ZFS on CentOS

Tue, 08 Oct 2013 07:55:46 +0000

First we need to add the ZFS on Linux repository to our system by installing a zfs-release package as shown below:

$ sudo yum localinstall --nogpgcheck http://archive.zfsonlinux.org/epel/zfs-release-1-3.el6.noarch.rpm
$ sudo yum install zfs

Once, zfs has been installed, we can create a storage pool. There’re many RAID configuration to choose from, and I’m not going to get into it. However, if you want to learn more, you can read this ZFS Administration article which explains in great detail about ZFS RAID(Z).

Install crontab on CentOS 6

Mon, 27 May 2013 14:33:26 +0000

When you install CentOS 6 using minimal CD, a lot of packages aren’t installed by default; cron included. To install it,

# yum install cronie

After the installation, you can verify that crond is running:

# chkconfig --list crond
crond 0:off 1:off 2:on 3:on 4:on 5:on 6:off

CentOS: lsb_release: command not found

Sun, 21 Oct 2012 22:53:07 +0000

First let me declare that I’m new to CentOS. The folowing is a note to self which demonstrates how we search for a package that provides some commands and intall it using yum on CentOS.

When I ran ‘lsb_release -a’, the following message is returned:

-bash: lbs_release: command not found

To fix it, we need to install the package which provides lsb_release command.

# yum provides */lsb_release
...
redhat-lsb-4.0-3.el6.centos.i686 : LSB base libraries support for CentOS
Repo        : base
Matched from:
Filename    : /usr/bin/lsb_release

redhat-lsb-4.0-3.el6.centos.x86_64 : LSB base libraries support for CentOS
Repo        : base
Matched from:
Filename    : /usr/bin/lsb_release

dkms-2.2.0.3-2.el6.noarch : Dynamic Kernel Module Support Framework
Repo        : epel
Matched from:
Filename    : /usr/lib/dkms/lsb_release

So, let’s install redhat-lsb package: