How to add ISO image to your linux repositories and use it to install basic software packages

Summary:

In this post, we are going to briefly see how you can add your Linux ISO image to the repository and use this to install basic software packages. We shall use an example of installing VSFTPD package used for SFTP service. 

This example was carried out on openSUSE Leap 15.0

Problem or Goal: 

This technique is mostly used when you are working with a new installation and you don't have access to the internet which would otherwise allow you to enable and install packages from online Linux repositories (repos).

Cause: 

Working on new Linux installation with no access to the internet (online repos)

Solution: 

1. Use the command below to check which repos are currently configured for your new installation, at this point you should have no repos enabled. 

# zypper lr -E

2. Use a flash disk to transfer the ISO image file to your new installation and copy it under the /tmp directory. 

3. Use the command below to add the ISO image to the repo

# zypper ar -c -t yast2 "iso:///?iso=\ /tmp/Put_The_ISO_Image_File_Name_Here.iso" "The_Alias_Name_Here"

4. Use the command below to check whether the repo is added successfully. 

# zypper lr -E

5. Use the command below to install the software package:

# zypper -n in vsftpd

Problem Solved?

Yes, with this technique, you can install the basic software packages that come with the ISO image and when you get access to the internet, you can add online repos and install more packages.

Quick Check - Common WireShark Filters When Troubleshooting Network/Communication Problems

Here is a list of commonly used WireShark filters for network and systems engineers when analyzing packet dumps:

1) Filtering DNS packets according to their "return code" (rcode):

(dns.flags.rcode == x) //Filter dns packets whose rcode is equal to "x".

!(dns.flags.rcode == x) //Filter dns packets whose rcode is NOT equal to "x".

See common "return codes" and their description in the table below:

Return Message	RCODE	Description
NOERROR	0	DNS Query completed successfully
FORMERR	1	DNS Query Format Error
SERVFAIL	2	Server failed to complete the DNS request
NXDOMAIN	3	Domain name does not exist.
NOTIMP	4	Function not implemented
REFUSED	5	The server refused to answer for the query
YXDOMAIN	6	Name that should not exist, does exist
XRRSET	7	RRset that should not exist, does exist
NOTAUTH	8	Server not authoritative for the zone
NOTZONE	9	Name not in zone

2) Filtering packets based on Source and Destination IP Addresses:

(ip.src == 31.13.64.11) && (ip.dst == 10.210.0.10)

3) Filtering M3UA packets based on Origination Point Code (OPC): 

(m3ua.protocol_data_opc == 12345) //Filter M3UA packets whose OPC is equal to "12345".

4) Filtering M3UA packets based on Destination Point Code (DPC): 

(m3ua.protocol_data_dpc == 34567) //Filter M3UA packets whose DPC is equal to "34567".

5) Filtering M3UA packets based on OPC or DPC: 

(m3ua.protocol_data_opc == 12345) || (m3ua.protocol_data_dpc == 34567) //Filter M3UA packets whose OPC is equal to "12345" OR DPC is equal to "34567".

(m3ua.protocol_data_opc == 12345) && (m3ua.protocol_data_dpc == 34567) //Filter M3UA packets whose OPC is equal to "12345" AND DPC is equal to "34567".

6) Filtering GSM MAP packets based on TBCD (Telephony Binary-Coded Decimal) digits: 

(gsm_map.tbcd_digits == "1234567") //Filter GSM MAP packets whose TBCD is equal to "1234567".

7) Filtering TCP/UDP packets based on port number: 

(tcp.port == 8080) //Filter TCP packets whose source or destination port number is equal to "8080".

(tcp.srcport == 80) //Filter TCP packets whose source port number is equal to "80".

(tcp.dstport == 443) //Filter TCP packets whose source or destination port number is equal to "443".

(udp.port == 53) //Filter UDP packets whose source or destination port number is equal to "53".

(udp.srcport == 53) //Filter UDP packets whose source port number is equal to "53".

(udp.dstport == 53) //Filter UDP packets whose destination port number is equal to "53".

8) Filtering TCAP (Transaction Capabilities Application Part) packets based on OTID (Orig. Transaction Id)/DTID (Dest. Transaction Id): 

(tcap.otid == 1a:00:6a:15) //Filter TCAP packets whose otid is equal to "1a:00:6a:15".

(tcap.dtid == 1a:00:6a:15) //Filter TCAP packets whose dtid is equal to "1a:00:6a:15".

9) Filtering Voice packets based on E.164 international telephone numbering plan:   

(e164.calling_party_number.digits == "256723039294") //Filter voice packets whose calling party (A)is equal to "256723039294".

(e164.called_party_number.digits == "254782243131") //Filter voice packets whose called party (B)is equal to "254782243131".

10) Filtering S1AP (Signalling between eNodeB & MME):   

(s1ap.ENB_UE_S1AP_ID == 101) //During LTE Attach & Default EPS Bearer Establishment, The eNodeB allocates 101 as the ENB-UE-S1AP-ID for the new session.

(s1ap.MME_UE_S1AP_ID == 211) //During LTE Attach & Default EPS Bearer Establishment, The MME allocates 211 as the MME-UE-S1AP-ID for the session.

11) Filtering GTP (GPRS Tunneling Protocol) packets:   

(gtp.ptmsi == 0xc6eb8017) //Filter GTP packets whose P-TMSI(Packet Temporary Mobile Subscriber Identity) is equal to "0xc6eb8017".

Taking a TCPDUMP on Linux OS and Writing the packets to CAP file (Wireshark Readable)

Summary: 

In this post, we are going to quickly take a TCP Dump (Capture packets) on a linux server and write the output to a CAP file that is WireShark readable. 

Problem or Goal: 

Used to capture and analyze packets when troubleshooting network/communication problems. 

Cause: 

N/A

Solution (Examples): 

a) TCP Dump on a single physical interface to capture DNS (port 53) packets

#tcpdump -i eth0 -s 2000 -w /tmp/name_of_trace_file.cap port 53

b) TCP Dump on a bonded interface to capture HTTP (port 80) packets

#tcpdump -i bond1 -s 2000 -w /tmp/name_of_trace_file.cap port 80

Note: "-s 2000" limits the packet length to 2000bytes, default packet length is 1024bytes use "-s 0" to remove the packet length limit. 

Problem Solved?

You can then transfer the file by SFTP to your laptop and read it with Wireshark 

How to grep a zipped file (.gz) without having to first unzipp it

Summary:

In this post, we are going to quickly look at a command to help us grep a zipped file (.gz format) without having to first unzipp it.

Problem or Goal: 

Well in most cases, system logs quickly grow very big and they are archived as zipped files (.gz format). We tend to always find ourselves in situation where we have to search and analyze archived logs when troubleshooting or doing fault analysis or preparing RCA reports. 

This command saves us the time to quickly grep through .gz archives without having to gunzip them, it also saves on the system resources because unzipping a file will consume disk space and cpu.

Cause:

Let's quickly look at apache "access_log" on my webserver:

# cd /var/log/apache2

# ls -ltr

At the moment, the access_log is not archived and i can easily cat and grep

Now am going to archive this file using "gzip" command and attempt to "cat and grep", i will use the gzip command that keeps the original file and creates a compressed copy. 

# gzip -c access_log > access_log.gz

# ls -ltr

Now let's attempt the "cat and grep"

Oops! it's returning nothing!

Solution: 

The solution is to use the "zcat" magic command instead of "cat"

Problem Solved?

Let's see

Yes we have output, it did the magic, we can grep a zipped file without having to first unzip it. 

Cheers.

How to recursively search for a repeated character in a text file and replace it with a new line using SED (e.g comma-delimited text)

Summary: 

In this post, I am going to show you how to search for repeated character in a text file and replace it with a new line to create a list. 

Problem or Goal: 

Take for example the task below, I have a log file containing parameters separated by a comma (","), see snippet below: 

3gdt,aace_support,access_restriction,adc,add,apn_conversion_wg,apn_redirection,apn_resolution_extension,attach_iot_limit,authentication_stationary_subscriber,conversational_qos_class,detach_inactive_subscriber_da,dual_access_support,dual_transfer_mode,ebm,edge_support,enhanced_uplink_mbr,gb_over_ip,gsm_adaptive_paging,gtp_user_location,gtpprime,gw_failure_restoration_gsm,gw_failure_restoration_wcdma,highest_qos_imsi,imei_check,integrated_traffic_capture,ipsec_support,lawful_interception,national_roaming_restriction,nw_init_sec_pdpctxt,payload_limit,pdp=1575,ps_ho,qos_hsdpa_mbr,rim_transfer,s_cdr_cause_code_ext,sau=2200,sau_lte=1,secondary_context,selective_service_request,sgsn_pool,sms_limit,srns_relocation,ss7_over_ip,streaming_qos_class,subscription_restriction,ue_signalling_control,ue_trace_mme

These are Ericsson SGSN-MME features, my task is to present these features in a neat list :-)

Cause:

Most log files are comma-delimited (Comma-delimited is a type of data format in which each piece of data is separated by a comma. This is a popular format for transferring data from one application to another, because most database systems are able to import and export comma-delimited data.)

Solution: 

The SED command below will read the file recursively and spit out a very nice list of Ericsson SGSN-MME features. 

linux-v7yi:/home# cat SGSN01-ER-SER-AD-Features-Summary.txt | sed -E 's/,/\n/g'

3gdt

aace_support

access_restriction

adc

add

apn_conversion_wg

apn_redirection

apn_resolution_extension

attach_iot_limit

authentication_stationary_subscriber

conversational_qos_class

detach_inactive_subscriber_da

dual_access_support

dual_transfer_mode

ebm

edge_support

enhanced_uplink_mbr

gb_over_ip

gsm_adaptive_paging

gtp_user_location

gtpprime

gw_failure_restoration_gsm

gw_failure_restoration_wcdma

highest_qos_imsi

imei_check

integrated_traffic_capture

ipsec_support

lawful_interception

national_roaming_restriction

nw_init_sec_pdpctxt

payload_limit

pdp=1575

ps_ho

qos_hsdpa_mbr

rim_transfer

s_cdr_cause_code_ext

sau=2200

sau_lte=1

secondary_context

selective_service_request

sgsn_pool

sms_limit

srns_relocation

ss7_over_ip

streaming_qos_class

subscription_restriction

ue_signalling_control

ue_trace_mme

Problem Solved?

Yes, a simpler example to understand is below:

linux-v7yi:/home# echo a,b,c,d,e,f,g,h,i,j,k | sed -E 's/,/\n/g'

a

b

c

d

e

f

g

h

i

j

k

linux-v7yi:/home#

Crontab -l: crontab: command not found - Fedora 32

Summary: 

[root@localhost ~]# crontab -l

-bash: crontab: command not found

Problem or Goal: 

To be able to schedule jobs/tasks using crontab on Fedora.

For example scheduling the cacti poller script to run every 5 mins

*/5 * * * *    root /usr/bin/php /usr/share/cacti/poller.php > /dev/null 2>&1

Cause: 

Crontab is not installed by default on Fedora

Solution: 

1. Fedora uses with the following automated task utilities: 

• cron
• anacron
• at
• batch

2. To install Cron and Anacron, run the following command:

[root@localhost cacti]# dnf install cronie cronie-anacron

3. Start the crond service

[root@localhost cacti]# systemctl start crond

Problem Solved? 

Yes

[root@localhost cacti]# crontab -l

no crontab for root

Use "crontab -e" command below to add your jobs/tasks

[root@localhost cacti]# crontab -e

for example the cacti poller job has been added

[root@localhost cacti]# crontab -l

*/5 * * * *    root /usr/bin/php /usr/share/cacti/poller.php > /dev/null 2>&1