Skip to main content
mike_bronson7
mike_bronson7
Gold ⭐️⭐️⭐️
July 6, 2020
Solved

HDFS replica + and min data nodes number in the HDFS cluster

we have Hadoop cluster with only 2 data nodes machines

 

in HDFS configuration we defined the Block replication to 3

 

so

Block replication=3

is it OK? to defined Block replication=3 , when we have only two data nodes in the cluster?

 

from my understanding when we defined Block replication=3 while we have 2 data nodes machines in HDFS cluster

 

its means that one machine should have 2 replica .  and the other machine one replica , am I correct here?

vsundaram
vsundaramAnswer
Gold ⭐️
July 14, 2020

@mike_bronson7 It is recommended to have minimum 3 data nodes in the cluster to accommodate 3 healthy replicas of a block as the default replication factor is 3. HDFS will not write replicas of same blocks on the same data node. In this scenario there will be under replicated blocks and 2 healthy replicas will be placed on the available 2 data nodes.

View in thread

1 reply

vsundaram
Gold ⭐️
vsundaramAnswer
Gold ⭐️
July 14, 2020

@mike_bronson7 It is recommended to have minimum 3 data nodes in the cluster to accommodate 3 healthy replicas of a block as the default replication factor is 3. HDFS will not write replicas of same blocks on the same data node. In this scenario there will be under replicated blocks and 2 healthy replicas will be placed on the available 2 data nodes.

Translate
    Terms & ConditionsAccessibility statement