hi all, i’d like to ask a question about how hadoop handle unequal disk size
i have hadoop cluster :
node1
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 20G 6.2G 13G 33% /
/dev/sda1 240G 79G 149G 35% /data
node2
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 20G 6.2G 13G 33% /
/dev/sda1 87G 72G 11G 87% /data
all stored in /data
this is my question :
1. since the disk on every cluster have different size, can i adjust the priority to store on node1 more than node2 (2:1 or 3:1) ?
2. if node2 getting full first (well, it should be first), what will happen to the running system ? does node2 will says “i’m full, don’s store at me”. then the system will store only at node1 ?
thanks for help