執行命令
[hduser@hadoop4 bin]$ kafka-run-class.sh kafka.tools.GetOffsetShell --broker-list 192.168.43.15:9092 --topic test --time -1
test:0:992786
test:1:1001516
test:2:1000773
test:3:1000894
test:4:1004034
[hduser@hadoop4 bin]$ kafka-run-class.sh kafka.tools.GetOffsetShell --broker-list 192.168.43.15:9092 --topic test --time -2
test:0:0
test:1:0
test:2:0
test:3:0
test:4:0
[hduser@hadoop4 bin]$ kafka-run-class.sh kafka.tools.GetOffsetShell
An interactive shell for getting topic offsets.
Option Description
------ -----------
--broker-list <String: hostname: REQUIRED: The list of hostname and
port,...,hostname:port> port of the server to connect to.
--max-wait-ms <Integer: ms> DEPRECATED AND IGNORED: The max amount
of time each fetch request waits.
(default: 1000)
--offsets <Integer: count> DEPRECATED AND IGNORED: number of
offsets returned (default: 1)
--partitions <String: partition ids> comma separated list of partition ids.
If not specified, it will find
offsets for all partitions (default:
)
--time <Long: timestamp/-1(latest)/-2 timestamp of the offsets before that.
(earliest)> [Note: No offset is returned, if the
timestamp greater than recently
commited record timestamp is given.]
(default: -1)
--topic <String: topic> REQUIRED: The topic to get offset from.
2、獲取條數
從上述的命令解析可以看到，
--time -1 表示獲取的最新位移值
--time -2 表示獲取的最早的位移值，可能由於最早的數據由於過期被刪除，所以最早的位移不一定是0
通過兩數相減，就可以知道當前分區的數據條數。