search

krisgeus / SEBC-London

0 stars 0 forks source link

Installation Lab #1

Open krisgeus opened 7 years ago

krisgeus commented 7 years ago

Check vm.swappiness 

sysctl vm.swappiness
Setting it with `sysctl vm.swappiness=1` doesn't survive a reboot so  need to set it in a more permanant way.  
sudo sh -c 'echo "vm.swappiness = 1" >> /etc/sysctl.d/00_swappiness.conf'
krisgeus commented 7 years ago

Show mount attributes. 

mount
krisgeus commented 7 years ago

Show the reserve space of ext based volumes

No ext based volumes available. If there were i'd use the tune2fs <device> command i guess

krisgeus commented 7 years ago

Disable transparent hugepages 

sudo sh -c 'echo never > /sys/kernel/mm/transparent_hugepage/enabled'
sudo sh -c 'echo never > /sys/kernel/mm/transparent_hugepage/defrag'

Added these commands to /etc/rc.d/rc.local to make it permanent (surviving a reboot)

krisgeus commented 7 years ago

List network interface configuration 

ifconfig -a
krisgeus commented 7 years ago

List forward and reverse lookups using getent or nslookup 

getent hosts $(hostname -i)
getent hosts $(hostname -f)

to make sure we get back an ipv4 address use: 

getent ahostsv4 $(hostname -f)
krisgeus commented 7 years ago

Show nscd is running 

systemctl status nscd

!! Its not running BTW.

krisgeus commented 7 years ago

Show ntpd is running 

systemctl status ntpd

!! It's not running BTW

krisgeus commented 7 years ago 
systemctl status chronyd
● chronyd.service - NTP client/server
   Loaded: loaded (/usr/lib/systemd/system/chronyd.service; enabled; vendor preset: enabled)
   Active: active (running) since Mon 2017-06-05 09:18:50 EDT; 57min ago
  Process: 495 ExecStartPost=/usr/libexec/chrony-helper update-daemon (code=exited, status=0/SUCCESS)
  Process: 478 ExecStart=/usr/sbin/chronyd $OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 486 (chronyd)
   CGroup: /system.slice/chronyd.service
           └─486 /usr/sbin/chronyd

Jun 05 09:18:50 ip-172-31-5-114.eu-west-1.compute.internal systemd[1]: Starting NTP client/server...
Jun 05 09:18:50 ip-172-31-5-114.eu-west-1.compute.internal chronyd[486]: chronyd version 2.1.1 starting (+CMDMON +NTP +REFCLOCK +RTC +PRIVDROP +DEBUG +ASYNCDNS +IPV6 +SECHASH)
Jun 05 09:18:50 ip-172-31-5-114.eu-west-1.compute.internal chronyd[486]: Frequency 19.083 +/- 0.270 ppm read from /var/lib/chrony/drift
Jun 05 09:18:50 ip-172-31-5-114.eu-west-1.compute.internal systemd[1]: Started NTP client/server.
Jun 05 09:18:56 ip-172-31-5-114.eu-west-1.compute.internal chronyd[486]: Selected source 52.209.118.149
Jun 05 09:18:56 ip-172-31-5-114.eu-west-1.compute.internal chronyd[486]: System clock wrong by 0.982776 seconds, adjustment started
krisgeus commented 7 years ago

Used ansible to set the vm.swappiness and the disabled transparent huge pages on all nodes

krisgeus commented 7 years ago

Setup mysql and replication with ansible. Satus of replication is in installation/labs/2_replica_working.md

krisgeus commented 7 years ago

Had some problems with setting up the cluster. SELinux was not disabled and seems like the culprit. Also had to open up a lot of ports for internode communication in the aws security group. Ended up allowing all internode communication on all ports to get the smoothest install.

After disabling SELinux and throwing away the half-backed cluster i was able to install CDH. Not sure if setting /var/lib/zookeeper (and more directories like it to 777 helped or it was really SELinuxx??). Need to find that out later by starting blank and let ansible do the magic again