kudu启动失败

Question

首先：两个配置文件
master.gflagfile ：
–fromenv=rpc_bind_addresses
–fromenv=log_dir
–fs_wal_dir=/home/hadoop/data/kudu/kudu_master_data
–fs_data_dirs=/home/hadoop/data/kudu/kudu_master_data
tserver.gflagfile：
–fromenv=rpc_bind_addresses
–fromenv=log_dir
–fs_wal_dir=/home/hadoop/data/kudu/kudu_tserver_data
–fs_data_dirs=/home/hadoop/data/kudu/kudu_tserver_data
–tserver_master_addrs=hadoop000:7051
之后我看还需要启动ntpd：
sudo /bin/systemctl start ntpd.service
sudo systemctl enable ntpd.service
启动后还是报：
->sudo ./kudu-master start
Failed to start Kudu Master Server. Return value: 1 [FAILED]
老师求解，还有报错日志在哪里呢，我没有找到哪个log日志，在之前创建的那个log文件夹里是空的。

Answer 1

Michael_PK 2020-09-16 00:17:09

你先把kudu的进程都kill掉，然后去/home/hadoop/data/kudu这个目录下面找找，看看具体的日志信息。

Failed to start Kudu Master Server. Return value: 1 [FAILED] 这个信息看不出来，只有找到对应的日志才可以

• 提问者 慕九州3016327 #1

  目录都翻过了，没有日志文件

• Michael_PK 回复 提问者 慕九州3016327 #2

  你把这个卸载了，然后换个配置文件里面的配置换个目录，重新配一次

• 提问者 慕九州3016327 回复 Michael_PK #3

  找到了pk哥，居然在/var/log/kudu下。
  stdout:
  ntpdc 4.2.6p5@1.2349-o Tue Jun 23 15:38:20 UTC 2020 (1)
  localhost: timed out, nothing received
  ***Request timed out
  localhost: timed out, nothing received
  ***Request timed out
  localhost: timed out, nothing received
  ***Request timed out
  
  E0916 19:46:36.134727  2215 system_ntp.cc:91] /bin/chronyc -n tracking
  ------------------------------------------
  stdout:
  506 Cannot talk to daemon
  
  E0916 19:46:36.136970  2215 system_ntp.cc:91] /bin/chronyc -n sources
  ------------------------------------------
  stdout:
  506 Cannot talk to daemon
  
  F0916 19:46:36.137012  2215 master_main.cc:74] Check failed: _s.ok() Bad status: Service unavail
  able: Cannot initialize clock: Error reading clock. Clock considered unsynchronized
  报的这个错误，我看是时钟不同步，可是我是单机的，没有集群，怎么会不同步呢

点开查看后面4条评论