当前位置：网站首页>Connect error: no route to host (errno:113)

Connect error: no route to host (errno:113)

2022-06-29 03:09:00 【Feng, crazy】

One 、 problem

#yyds Dry inventory #ceph File mount failed Connect error: No route to host(errno:113)_linux

      
      
       
       // Failure reports error 
       
       
Connect error: No route to host(errno:113)
      
      
      
      
       
       1.
       
       2.

1.1 Check the network

      
      
       
       1. Check the firewall 
       
       
systemctl status firewalld
       
       
systemctl stop firewalld
       
       
2. Check selinux
       
       
getenforce        // If disabled  It is closed , If enforce  It's a forced pattern  
       
       
setenforce 0   // Temporarily Closed 
       
       
 // I began to think it was a network problem , But after investigation , There is no firewall on all servers , also 
       
       
 telnet ceph01:6789// No problem  (yum install telnet -y)
      
      
      
      
       
       1.
       
       2.
       
       3.
       
       4.
       
       5.
       
       6.
       
       7.
       
       8.

Two 、 resolvent

2.1 ceph Status view

      
      
       
       // see ceph state 
       
       
  ceph status    Show  HEALTH_OK,
      
      
      
      
       
       1.
       
       2.

2.2 mds Status view

      
      
       
        see mds state 
       
       
    ceph mds stat
       
       
     Show ：
       
       
    cephfs-0/0/1 up
       
       
    
       
       
 Check it again ceph-mds state ,
       
       
    systemctl status [email protected]
       
       
    // Different nodes hostname Dissimilarity   Find out mds No running, There's something wrong with the State 
      
      
      
      
       
       1.
       
       2.
       
       3.
       
       4.
       
       5.
       
       6.
       
       7.
       
       8.

2.3 mds restart

      
      
       
        restart （ Failure ）
       
       
 systemctl start [email protected]    
       
       
 Report errors keyring file does not exist , To avoid unnecessary trouble , Decide to reload mds
      
      
      
      
       
       1.
       
       2.
       
       3.

2.4 mds reinstall

      
      
       
       1、 stop it mds process .
       
       
        ceph stop mds
       
       

       
       
2、 take mds Service identification is invalid .
       
       
    ceph mds fail 0
       
       

       
       
3、 Delete cephfs file system .
       
       
    ceph fs rm cephfs --yes-i-really-mean-it
       
       

       
       
4、 see ceph State of the cluster .
       
       
    #ceph -s 
       
       
    // Show mds No longer exists 
       
       

       
       
5、 establish MDS
       
       
    ceph-deploy admin ceph01 ceph02 ceph03
       
       

       
       
    ceph-deploy mds create ceph01 ceph02 ceph03
       
       

       
       
    ceph mds stat
       
       
    // Show cephfs-1/1/1 up  {0=ceph01=up:active}, 2 up:standby
      
      
      
      
       
       1.
       
       2.
       
       3.
       
       4.
       
       5.
       
       6.
       
       7.
       
       8.
       
       9.
       
       10.
       
       11.
       
       12.
       
       13.
       
       14.
       
       15.
       
       16.
       
       17.
       
       18.
       
       19.
       
       20.

2.4 Remount the file system

      
      
       
       //mount  and  umount
       
       
    // mount 
       
       
    mount -t ceph 192.168.103.49:6789,192.168.103.50:6789,192.168.103.51:6789:/ /mnt/cephfs/ -o name=admin,secret=AQBOJqdhzwUGKhAARGeIdl57JF7lcFnx602K/w==
       
       

       
       
    // uninstall 
       
       
    umount /mnt/cephfs
      
      
      
      
       
       1.
       
       2.
       
       3.
       
       4.
       
       5.
       
       6.