I am using ansible to do the deployment, according to the document or easy to deploy. 1. File system has used EXT4 format to modify mount parameters (increase nodelalloc and noatime)
First, check the file format of the system, the command is as follows
#vi/etc/fstab
In the row where you set the data store's mount point, the third section is the disk format, and in the fourth section set the parameters (note, comma-delimited).
A reboot is then required and the system reads the file mount disk when it starts. 2. Time Synchronization Service Sync issue
If the following error occurs in the deployment times, there are two reasons why the NTP service is not started on the server and the other is started but not synchronized in time.
TASK [Check_config_dynamic:preflight Check-fail when NTP service was not running or Ntpstat was not synchronised to NTP Server] *************************************************** fatal: [10.222.222.36]: failed! = = {"Changed": false, "cmd": "Ntpstat | Grep-w Synchronised | Wc-l "," Delta ":" 0:00:00.098635 "," End ":" 2018-03-02 13:33:31.529697 "," Failed_when_result ": True," RC ": 0," Start ":" 201 8-03-02 13:33:31.431062 "," stderr ":" "," Stderr_lines ": []," stdout ":" 0 "," stdout_lines ": [" 0 "]} fatal: [10.222.222.35]: failed! = = {"Changed": false, "cmd": "Ntpstat | Grep-w Synchronised | Wc-l "," Delta ":" 0:00:00.113097 "," End ":" 2018-03-02 13:33:31.635350 "," Failed_when_result ": True," RC ": 0," Start ":" 201 8-03-02 13:33:31.522253 "," stderr ":" "," Stderr_lines ": []," stdout ":" 0 "," stdout_lines ": [" 0 "]} fatal: [10.222.222.34]: failed! = = {"Changed": false, "cmd": "Ntpstat | Grep-w Synchronised | Wc-l "," Delta ":" 0:00:00.113614 "," End ":" 2018-03-02 13:33:31.644088 "," Failed_when_result ": True," RC ": 0," Start ":" 2018-03-02 13:33:31.530474 "," stderr ":" "," Stderr_lines ": []," stdout ":" 0 "," stdout_l Ines ": [" 0 "]}
For the first heavy issue, run the command:
$ sudo systemctl start ntpd.service
If prompted not to find Ntpd.service, install by command:
$ sudo yum install NTP
Check NTP status:
$ ntpstat
If normal, the following results will be returned, with emphasis on (Word synchronised-synchronization):
Synchronised to NTP server (85.199.214.101) @ Stratum 2 time
correct to within MS
polling server every 1024x768 s
3. Simple installation, required number of machines and memory and disk requirements
I used the virtual machine installed, created 4 virtual machines, each virtual machine allocated 2GB of memory, dynamic 30G disk.
4 Machine allocation: 1 installed TIDB server and pd,3 installation tikv.
30G of disk, will be used out of 23G multipoint.
(Other problems, welcome communication)