TEAM TASK NO 1

TEAM TASK NO 1

!! ?????????? ?????????????????????? !!

?? TASK DESCRIPTION:

?? Whenever the client uploads the file ( for ex - f.txt) of size 50MB and the replication is 3.

?? Does the client takes the entire data to the master or does the master provides the IP addresses of Datanodes so that client can upload the file to the Datanodes?

?? Question: Whom is the one uploading the file?

?? Answer: The client gets the IP from the Master and uploads the file to DataNode.

SOLUTION :

First, we have to go inside /etc/Hadoop directory to configure Hadoop configuration files.

No alt text provided for this image

Here in the slave hdfs-site.xml file, we have to put our dn directory to share the storage with the master.

No alt text provided for this image

For master, we have configured the same file also.

No alt text provided for this image

Now we configure core-site.xml in all nodes including the master, client, and slave.

No alt text provided for this image

In slave and client, give master public IP to do configuration.

No alt text provided for this image

In the master node, give 0.0.0.0 to allow all nodes within the cluster to contact the master.

No alt text provided for this image

Now we have to format the name node directory.

No alt text provided for this image

Then we have to start the namenode.

No alt text provided for this image

After starting the name node now we have to start the datanode also

No alt text provided for this image

From the client, the file is uploading.

No alt text provided for this image

Then we can run tcpdump command on all the nodes as well as in the client to capture all the packets.

No alt text provided for this image

In the client packets file, Here client first connects to the master to know which slave is available to use as the master has all slave nodes metadata.

No alt text provided for this image

Now the client can directly connect to slave 1 to upload the first block.

No alt text provided for this image

Again the client connects to slave 1 to upload the second block.

No alt text provided for this image

For the final block, once again client contact slave 1.

No alt text provided for this image

In the slave 1 packets file, slave 1 is first contacted by the client, and the file is uploaded in three blocks in slave 1 through the client.

No alt text provided for this image

Then slave 1 connect slave 2 to create a replica.

No alt text provided for this image

In slave 2 packets file, as shown slave 2 is contacted by slave 1, and a replica is created here.

No alt text provided for this image

Then slave 2 create another replica in slave 3.

No alt text provided for this image

Slave 3 gets data from slave 2 to upload and hence the third replica is also created in slave 3.

No alt text provided for this image

Here we successfully completed our Team_Task_No_1

Thank You For Reading !!!

Onkar Naik

DevOps @Forescout ?? | Google Developer Expert | AWS | DevOps | 3X GCP | 1X Azure | 1X Terraform | Ansible | Kubernetes | SRE | Platform | Jenkins | Tech Blogger ??

4 年

Great work Anushka Visapure and team ????

Harshal Thakare

ATSE@RedHat || Openshift || 3x RedHat Certified || DevOps(Docker??, Kubernetes?, Jenkins????) || Ansible || Cloud Computing ?(AWS) |||

4 年

Greatly done Anushka Visapure

要查看或添加评论,请登录

Anushka Visapure的更多文章

社区洞察

其他会员也浏览了