DEV Community: Mohammad Faisal

GitLab - skip pipeline on push

Mohammad Faisal — Wed, 28 Jul 2021 07:32:44 +0000

If you have configured a GitLab CI/CD pipeline well then on each push to the repository result in execution of the pipeline. Sometimes it is not desirable to launch the pipeline immediately on the push or you may want to execute the pipeline using a schedule, in such case you can pass additional push option to instruct GitLab to skip the pipeline execution. To do so, the command would be:

git push -o ci.skip

But if you are using any GUI client for git then you will not be able to pass this additional argument to the push command. In that case, you can configure the push.pushOption instead as:

git config --local --add push.pushOption ci.skip

^* Make sure to remove this configuration when you no longer want to skip the pipeline on push. To do so, use the --unset option:

git config --local --unset push.pushOption

If you prefer git CLI over GUI then you can create alias for such case:

git config --local --add alias.skipPipeline "push -o ci.skip"

Now you have choice:

to skip pipeline use command: git skipPipeline
normal push with pipeline execution: git push

I hope you may find this information useful. Feel free to leave your feedback/opinion in comments.

Find and remove docker image

Mohammad Faisal — Wed, 10 Feb 2021 17:15:13 +0000

Docker documentation mentions removing all stopped containers using command:

$ docker rm $(docker ps -a -q)

but at times we want to first find existing containers and then remove them. This can be achieved when you filter and limit the result of docker ps command.

Limit the result of `docker ps`

You can either use the --filter, --last or --latest options or pipe the result of docker ps to grep command to filter the results for matching PATTERN.

Here is an example using grep:

$ docker ps -a | grep PATTERN | \
  awk '{print $1}' | xargs -r docker rm

where the PATTERN could be part of container name, image name or any other details present in the result of docker ps command like mapped port number.

Feel free to share your feedback/opinion in comments.

Photo by Boxed Water Is Better on Unsplash

Solving CWE-606: Unchecked Input for Loop Condition

Mohammad Faisal — Mon, 11 May 2020 15:12:55 +0000

The CWE-606: Unchecked Input for Loop Condition is described as:

The product does not properly check inputs that are used for loop conditions, potentially leading to a denial of service because of excessive looping.

with an attached demonstrative example in C as:

void iterate(int n){
  int i;
  for (i = 0; i < n; i++){
    foo();
  }
}
void iterateFoo()
{
  unsigned int num;
  scanf("%u",&num);
  iterate(num);
}

Recently I came across this issue in my Java application when we scanned the project with Checkmarx's CxSAST. The code is something like:

List<String> records = ... // The records are read from a received file
for ( int i = 0; i < records.size(); i++){
  // some processing here
}

Now in order to solve this issue I identify two approaches:

define a max size of records and compare the size of the records to that value and if it is greater than that then either do not process any record and exit or process the records in chunks by splitting them by the max size.
rewrite this for-loop as for-each which do not check for the execution condition and process until it has record in there.

In the 1st solution it would be difficult to put a size restriction as we do not know the limit of number of records could be there and we have to process them all. If we go by a chunk processing then we could put some arbitrary value.

In the 2nd solution we may trick the checkmarx that we have not used the user input as loop condition but I am not sure if that will work or not?

What is your opinion on this?

Spring batch - changing default data-source

Mohammad Faisal — Fri, 24 Jan 2020 05:51:33 +0000

I had recently started learning Spring-Batch and followed the very first Getting started guide.
The default datasource used by Spring-batch is HSQL as an in-memory database, good enough for local development. But you may not only use that in production code. You can change that datasource to yours with the following configurations:

spring.datasource.url=<jdbc:url:for-your-database>
spring.datasource.username=<db-username>
spring.datasource.password=<db-password>

All spring.datasource.* properties are bound to the fields of org.springframework.boot.autoconfigure.jdbc.DataSourceProperties class.

Missing draft post

Mohammad Faisal — Fri, 24 Jan 2020 02:38:12 +0000

I had recently created a post and saved that as draft. I can no longer find that in my profile. Where does the draft posts goes?

Minimum Configuration for log4j

Mohammad Faisal — Wed, 08 Jan 2020 17:37:31 +0000

No appenders could be found for logger(log4j)

You may have encountered this issue with many applications (hibernate, spring, etc.) when trying to run a Java program from command line.
The error appears in console something like this:

log4j:ERROR Could not find value for key log4j.appender.CONSOLE
log4j:ERROR Could not instantiate appender named "CONSOLE".

log4j:WARN No appenders could be found for logger (com.company.module.ClassName).
log4j:WARN Please initialize the log4j system properly.
log4j:WARN See http://logging.apache.org/log4j/1.2/faq.html#noconfig for more info.

As the warning says itself: Please initialize the log4j system properly, you must check when running the program did the log4j.properties or log4j.xml is available on the classpath or either one of them do really exists?

A minimum configuration would be

log4j.properties

# Set root logger level to DEBUG and its only appender to CONSOLE.
log4j.rootLogger=DEBUG, CONSOLE

# CONSOLE appender is set to be a ConsoleAppender.
log4j.appender.CONSOLE=org.apache.log4j.ConsoleAppender

# CONSOLE appender uses PatternLayout.
log4j.appender.CONSOLE.layout=org.apache.log4j.PatternLayout
log4j.appender.CONSOLE.layout.ConversionPattern=%-4r [%t] %-5p %c %x - %m%n

log4j.xml

<!DOCTYPE log4j:configuration SYSTEM "log4j.dtd">
<log4j:configuration debug="true" xmlns:log4j='http://jakarta.apache.org/log4j/'>

  <appender name="console" class="org.apache.log4j.ConsoleAppender">
    <param name="Target" value="System.out"/>
    <layout class="org.apache.log4j.PatternLayout">
    <param name="ConversionPattern" value="%d{yyyy-MM-dd HH:mm:ss} %-5p %c{1}:%L - %m%n" />
    </layout>
  </appender>

  <root>
    <priority value ="debug"></priority>
    <appender-ref ref="console"></appender>
  </root>
</log4j:configuration>

You can name anything for the appender name which I had chosen CONSOLE/console as this is more relevant if you want to see the logs in console rather than appended in some file.

Application specific log configuration

log4j.logger.domain.company.name=<LOG_LEVEL>