Service Management

Description of Services

abiquo-tomcat

• What it does: It is a servlet container responsible for the management of Abiquo modules.
• When it is required: When there are Abiquo modules on the server.

• How to check its status:

  # ps aux | grep tomcat
  

• Example output:

  root     18348  9.6 26.9 2457384 833320 ?      Sl   Feb29 782:38 /usr/java/default/bin/java
   -Djava.util.logging.config.file=/opt/abiquo/tomcat/conf/logging.properties -server -Xms1024m
   -Xmx1500m -XX:MaxPermSize=512m -XX:+CMSClassUnloadingEnabled -Djava.awt.headless=true -
  Djava.util.logging.manager=org.apache.juli.ClassLoaderLogManager -Djava.endorsed.dirs=/opt/abiquo/tomcat/endorsed -
  classpath /opt/abiquo/tomcat/bin/bootstrap.jar -Dcatalina.base=/opt/abiquo/tomcat -Dcatalina.home=/opt/abiquo/tomcat -
  Djava.io.tmpdir=/opt/abiquo/tomcat/temp org.apache.catalina.startup.Bootstrap start
  

• How to start or stop it:

  # service abiquo-tomcat start
  # service abiquo-tomcat stop
  # service abiquo-tomcat restart
  

mysqld

• What it does: MySQL daemon that runs the MariaDB database that contains most of the information that Abiquo uses
• When it is required: By default, it is installed on the Abiquo Server. It can be installed on a separate server or an already existing MySQL server.

• How to check its status:

  # service mysql status
  

• How to start or stop it:

  Caution!

  Stopping the MySQL database server before the Abiquo Server may harm the environment and cause data loss.

  # service mysql start
  # service mysql stop
  

redis

• What it does: The Redis server manages event subscription on nodes and delivers it to Abiquo. The Monitor manager or VSM remote service detects state changes in the virtual systems. So for example, when you add a virtual machine, the VSM creates a monitor for that virtual machine, and when the virtual machine is powered on, the state change is detected and notified by the VSM.
• Where is it required: On Remote Services servers that contain the Monitor Manager (VSM) Remote Service. Redis can also be installed on a separate server for the remote services. Redis is also installed on the Abiquo Server. In a monolithic configuration, Abiquo uses only one Redis instance.

• How to check its status:

  # redis-cli ping
  

  Alternatively, type the following commands:

  # redis-cli
  redis 127.0.0.1:6379> set mykey somevalue
  OK
  redis 127.0.0.1:6379> get mykey
  "somevalue"
  

• How to start or stop it:

  Caution

  Always stop the Abiquo-tomcat services first before stopping Redis.

  # redis start
  # redis-cli shutdown
  

rabbitmq-server

• What it does: RabbitMQ is used for communications between Remote Services and the central Abiquo Server and its API
• When it is required: On Abiquo main server installations

• How to check its status:

  Caution

  Always stop the Abiquo-tomcat services first before stopping RabbitMQ.

  # service rabbitmq-server status
  

• How to start or stop it:

  # service rabbitmq-server start
  # service rabbitmq-server stop
  

v2v-diskmanager

• What it does: Script used for automating VM Template conversions to supported formats
• When it is required: For disk format conversions. For performance reasons, it should be installed as a separate V2V Server.

• How to check its status: This is not a daemon; it is a service initiated by Abiquo. So the best way to know if it is currently working on a conversion is using ps aux:

  # ps aux | grep v2v
  root     32046  0.2  0.0  63892  1200 ?        S    16:44   0:00 /bin/bash /usr/bin/v2v-diskmanager VMDK_STREAM_OPTIMIZED L29wdC92bV9yZXBvc2l0b3J5LzEvcnMuYmNuLmFiaXF1by5jb20vYWJpcG9ydDkwMDAvb3ZmLzcyL2ZpbGUtYnVpbGQtYm90
  LXJoZWw1LWRpc2sxLnZtZGs= VMDK_FLAT L29wdC92bV9yZXBvc2l0b3J5LzEvcnMuYmNuLmFiaXF1by5jb20vYWJpcG9yd
  DkwMDAvb3ZmLzcyL2Zvcm1hdHMvZmlsZS1idWlsZC1ib3QtcmhlbDUtZGlzazEudm1kay1WTURLX0ZMQVQ=
  root     32095  0.0  0.0  61224   756 pts/1    R+   16:44   0:00 grep v2
  

• How to start or stop it: You will only need to stop it or to start it when there is an issue during conversions. For example, if you want to test why a conversion failed. You can check the /opt/abiquo/tomcat/logs/bpm-async.log to find the command issued by Abiquo:

  2012-03-23 16:44:07,142 DEBUG c.a.a.b.s.u.ExecUtils [ExecUtils.java:46] Executing OS command: v2v-diskmanager VMDK_STREAM_OPTIMIZED L29wdC92bV9yZXBvc2l0b3J5LzEvcnMuYmNuLmFiaXF1by5jb20vYWJpcG9ydDkwMDAvb3ZmLzcyL2ZpbGUtYnVpbGQtYm90LX
  JoZWw1LWRpc2sxLnZtZGs= VMDK_FLAT L29wdC92bV9yZXBvc2l0b3J5LzEvcnMuYmNuLmFiaXF1by5jb20vYWJpcG9ydDkwM
  DAvb3ZmLzcyL2Zvcm1hdHMvZmlsZS1idWlsZC1ib3QtcmhlbDUtZGlzazEudm1kay1WTURLX0ZMQVQ=
  

  Copy a command found in the log and run it again to see more detail from the error message. For example:

  $ v2v-diskmanager VMDK_STREAM_OPTIMIZED L29wdC92bV9yZXBvc2l0b3J5LzEvcnMuYmNuLmFiaXF1by5jb20vYWJpcG9
  ydDkwMDAvb3ZmLzcyL2ZpbGUtYnVpbGQtYm90LXJoZWw1LWRpc2sxLnZtZGs= VMDK_FLAT L29wdC92bV9yZXBvc2l0b3J5LzE
  vcnMuYmNuLmFiaXF1by5jb20vYWJpcG9ydDkwMDAvb3ZmLzcyL2Zvcm1hdHMvZmlsZS1idWl
  sZC1ib3QtcmhlbDUtZGlzazEudm1kay1WTURLX0ZMQVQ=
  

  To stop it, use the pkill command, or from the top application, pressing k to kill the process.

mechadora

• What it does: Creates a persistent machine from a VM Template that will run on an iSCSI disk
• When is required: This is a remote service that would usually be installed on V2V Servers.

• How to check its status:

  # ps aux | grep mechadora
  root      2521  0.0  0.0  63892  1160 ?        S    18:05   0:00 /bin/bash /usr/bin/mechadora
  deploy 10.60.13.20 iqn.2008-03.com.abiquo.localhost.localdomain:8654e6ac-cb3f-4751-9600-4bc642c327c5 
  /opt/vm_repository/1/abiquo-repository.s3.amazonaws.com/ubuntu-11.04/formats/ubuntu-11.04.qcow2-RAW 
  ip-10.60.13.20:3260-iscsi-iqn.2008-03.com.abiquo.localhost.localdomain:8654e6ac-cb3f-4751-9600-
  4bc642c327c5-lun-1
  

• How to start or stop it: The procedure is the same as for v2v-diskmanager.

nfs

• What it does: It is the service that manages the NFS shares from our system when we export them. In Abiquo we use it to share the vm_repository
• When it is required: Only when we are sharing our vm_repository, usually in small installations.

• How to check its status:

  # service nfs status
  

• How to start or stop it: Use the service command.

  # service nfs start
  # service nfs stop
  

smb

• What it does: Manages Samba shares in Linux environments. On the Abiquo platform, it is used to share the vm_repository origin with Hyper-V Hypervisors.
• When it is required: On the vm_repository origin, from where it will also be exported using NFS.

• How to check its status:

  # service smb status
  

• How to start or stop it:

  # service smb start
  # service smb stop
  

dhcpd

• What it does: The daemon for the DHCP server. Assigns the IPs that the user selected in Abiquo to each virtual machine.
• When it is required: On our Remote Services servers. It can also be installed separately and requires a trunk port connection to reach the hypervisors.

• How to check its status:

  # service dhcpd status
  

• How to start or stop it:

  # service dhcpd start
  # service dhcpd stop
  

dhrelay

• What it does: It is the main component on DHCP Relay servers, see Abiquo DHCP documentation for more information.
• When it is required: On DHCP Relay servers. See the documentation above.

• How to check its status:

  # service dhrelay status
  

• How to start or stop it:

  # service dhcpd start
  # service dhcpd stop
  

abiquo-aim

• What it does: It is the link that communicates between Libvirt on KVM hypervisors and the Remote Services, in particular Redis.
• When it is required: On KVM hypervisors

• How to check its status:

  # ps aux | grep abiquo-aim
  

• If abiquo-aim is running, you will see a line like this:

  root      2816  0.0  0.4  95528  2104 ?        Ssl  13:26   0:00 /usr/sbin/abiquo-aim --config-file /etc/abiquo-aim.ini -d
  

• How to start or stop it:

  # service abiquo-aim start
  # service abiquo-aim stop
  

abiquo-emmett

• What it does: It is part of the Watchtower monitoring module that interacts with the VSM and Abiquo API and the KairosDB to manage metrics and alarms.
• When it is required: On the Abiquo Monitoring server

• How to check its status:

  # ps aux | grep abiquo-emmett
  

• If abiquo-emmett is running, you will see a line like this:

  XXX TODO

• How to start or stop it:
  NOTES ABOUT START / STOP ISSUES 

  # service abiquo-emmett start
  # service abiquo-emmett stop

abiquo-delorean

• What it does: It is part of the Watchtower monitoring module. It evaluates monitoring alarms and alerts
• When it is required: On the Abiquo Monitoring server

• How to check its status:

  # ps aux | grep abiquo-delorean
  

• If abiquo-delorean is running, you will see a line like this:

  XXX TODO

• How to start or stop it:
  NOTES ABOUT START / STOP ISSUES 

  # service abiquo-delorean start
  # service abiquo-delorean stop

kairosdb

• What it does: It is part of the Abiquo Monitoring system. It is a fast scalable time series database. Abiquo uses it to store metrics with the Cassandra NoSQL database.
• Reference: http://kairosdb.github.io/docs/build/html/GettingStarted.html
• When it is required: It is required for monitoring. By default it is installed on the Abiquo Monitoring server. In a larger installation it should be installed on a separate server.

• How to check its status:

  # ps aux | grep kairosdb

  If kairosdb is running, you will see a line like this:

• How to start or stop it:
  NOTES ABOUT START / STOP ISSUES 

  # ./kairosdb.sh start
  # ./kairosdb.sh stop

cassandra

• What it does: It is part of the Abiquo Monitoring system. To quote its wiki: "Cassandra is a highly scalable, eventually consistent, distributed, structured key-value store". Abiquo uses it to store metrics. 
• Reference: https://wiki.apache.org/cassandra/GettingStarted
• When it is required: It is required for monitoring. By default it is installed on the Abiquo Monitoring server. In a larger installation it should be installed on a separate server or even a cluster.

• How to check its status:

  # ps aux | grep cassandra

  If Cassandra is running, you will see a line like this:

  XXX TODO

• How to start or stop it:
  NOTES ABOUT START / STOP ISSUES 

  # bin/cassandra
  # pkill -f CassandraDaemon