Fail to start queue when run "grakn server start" (version 1.2.0)


#1

I’m trying grakn version recently, both on mac and CentOS. It’s ok using brew to install and start server, but when I tried it on CentOS, it fail to run “grakn server start” fail at step “starting queue…”

  1. environment
    CentOS release 6.5 (Final)
    using root to install
  2. install steps
    download "grakn-dist-1.2.0.tar.gz"
    tar xzf "grakn-dist-1.2.0.tar.gz"
    cd grakn-dist-1.2.0
    modify /etc/profile and add GRAKN_HOME into it then source /etc/profile.
    run "grakn server start"
    it shows below information:

Starting Storage…SUCCESS
Starting Queue…FAILED!
Unable to start Queue
Please run ‘grakn server status’ or check the logs located under ‘logs’ directory.

  1. I check the logs, seems show nothing useful.
    The cassandra.log file content in gist:
    cassandra.log
    The grakn-postprocessing.log shows nothing.
    The grakn.log shows few contents:

2018-07-18 07:37:46,711 [main] INFO ai.grakn.engine.GraknConfig - Project directory in use: /home/xiaoq/grakn-dist-1.2.0
2018-07-18 07:37:46,716 [main] INFO ai.grakn.engine.GraknConfig - Configuration file in use: /home/xiaoq/grakn-dist-1.2.0/conf/grakn.properties
2018-07-18 07:37:46,719 [main] INFO ai.grakn.engine.GraknConfig - Project directory in use: /home/xiaoq/grakn-dist-1.2.0
2018-07-18 07:37:46,719 [main] INFO ai.grakn.engine.GraknConfig - Configuration file in use: /home/xiaoq/grakn-dist-1.2.0/conf/grakn.properties
2018-07-18 07:37:46,721 [main] INFO ai.grakn.engine.GraknConfig - Project directory in use: /home/xiaoq/grakn-dist-1.2.0
2018-07-18 07:37:46,721 [main] INFO ai.grakn.engine.GraknConfig - Configuration file in use: /home/xiaoq/grakn-dist-1.2.0/conf/grakn.properties
2018-07-18 07:37:47,229 [main] INFO ai.grakn.engine.GraknConfig - Project directory in use: /home/xiaoq/grakn-dist-1.2.0
2018-07-18 07:37:47,229 [main] INFO ai.grakn.engine.GraknConfig - Configuration file in use: /home/xiaoq/grakn-dist-1.2.0/conf/grakn.properties

  1. ps -ef | grep redis-server

root 6827 1 0 09:38 ? 00:00:52 redis-server *:6379
root 13397 11995 0 15:52 pts/2 00:00:00 grep redis-server

  1. ls -la /tmp/*.pid

-rw-r–r-- 1 root root 5 7月 18 15:37 /tmp/grakn-storage.pid

  1. jps

13464 Jps
12558 CassandraDaemon


Please report any error "java.lang.RuntimeException: Couldn’t get PID of Grakn. Received '"
#2

We’ve observed that some people are having a problem with CentOS, which has an old library depended by Redis.

Can you try to start Redis manually and see if it runs:

cd grakn-dist-1.2.0
./services/redis/redis-server-linux ./services/redis/redis.conf
echo $? # check the status code of redis 

#3

Thanks for your suggestion. I run the command, and it shows below:

./services/redis/redis-server-linux: /lib64/libc.so.6: version GLIBC_2.14' not found (required by ./services/redis/redis-server-linux) ./services/redis/redis-server-linux: /lib64/libc.so.6: versionGLIBC_2.17’ not found (required by ./services/redis/redis-server-linux)

It’s a GLIBC issue.And then I installed GLIBC_2.17 to resolve this issue. The queue successfully boot up. Another issues I’d like to mention is that, the grakn integrated redis in it, but I haven’t notice it, I thought it can use the redis I already have on the server before I install grakn, but it’s not, so it might conflict with my prior installed one on the port. Do we have a installation guide on other components integrated in grakn, something we need to pay attention to?