CCS(8) | System Manager's Manual | CCS(8) |
NAME¶
ccs - Cluster Configuration System
SYNOPSIS¶
ccs [-h hostname | -f filename] [options]
DESCRIPTION¶
ccs allows an administrator to create, modify and view a cluster configuration file on a remote node through ricci or on a local file system. Using ccs an administrator can also start and stop the cluster services on one or all of the nodes in a configured cluster.
OPTIONS¶
Generally all command lines must specify either a hostname or filename to act upon.
MAIN OPTIONS¶
- --help
- Display this help and exit
- -h, --host host
- Cluster node to perform actions on
- --usealt
- If primary node name is unavailable, attempt to connect to ricci on the alt interface (only works from a cluster member)
- -f, --file file
- File to perform actions on
- -i, --ignore
- Ignore validation errors in cluster.conf file
- -p, --password
- Ricci user password for node running ricci
- --getconf
- Print current cluster.conf file
- --setconf
- Use the file specified by '-f' to send to the host specified with '-h'
- --checkconf
- If file is specified, verify that all the nodes in the file have the same cluster.conf as the file. If a host is specified then verify that all nodes in the host's cluster.conf file have the identical
- --getschema
- Print current cluster schema file (if using -h use schema from network, if using -f use local schema)
- --sync
- Sync config file to all nodes and optionally activating that configuration on all nodes
- --activate
- Activate config on node (use this option with --sync to activate config on all nodes)
- -b, --backup
- Backup cluster.conf file before changes in ~/.ccs/backup directory
- -d, --debug
- Display debugging information to help troubleshoot connection issues with ricci
- --exp tag [location] [options]
- Expert mode to add elements not currently defined in ccs
- --exprm location
- Expert mode to remove elements not currently defined in ccs
CLUSTER OPERATIONS¶
- --createcluster cluster
- Create a new cluster.conf (removing old one if it exists)
- --getversion
- Get the current cluster.conf version
- --setversion <n>
- Set the cluster.conf version
- --incversion
- Increment the cluster.conf version by 1
- --startall [--noenable|--nodisable]
- Start *AND* enable cluster services on reboot on all nodes (if --noenable or --nodisable is specified cluster services will only be started)
- --stopall [--noenable|--nodisable]
- Stop *AND* disable cluster services on reboot for all nodes (if --noenable or --nodisable is specified cluster services will only be stopped)
- --start [--noenable|--nodisable]
- Start *AND* enable cluster services on reboot for host specified with -h or localhost if no host is provided (if --noenable or --nodisable is specified cluster services will only be started)
- --stop [--noenable|--nodisable]
- Stop *AND* disable cluster services on reboot for host specified with -h or localhost if no host is provided (if --noenable or --nodisable is specified cluster services will only be stopped)
NODE OPERATIONS¶
- --lsnodes
- List all nodes in the cluster
- --addnode <node>
- Add node <node> to the cluster
- --rmnode <node>
- Remove a node from the cluster
- --nodeid <nodeid>
- Specify nodeid when adding a node
- --votes <votes>
- Specify number of votes when adding a node
- --addalt <node name> <alt name> [alt options]
- Add an altname to a node for RRP
- --rmalt <node name>
- Remove an altname from a node for RRP
FENCING OPERATIONS¶
- --lsfenceopts [fence type]
- List available fence devices. If a fence type is specified, then list options for the specified fence type
- --lsfencedev
- List all of the fence devices configured
- --lsfenceinst [<node>]
- List all of the fence methods and instances on the specified node or all nodes if no node is specified
- --addmethod <method> <node>
- Add a fence method to a specific node
- --rmmethod <method> <node>
- Remove a fence method from a specific node
- --addfencedev <device name> [fence device options]
- Add fence device. Fence devices and parameters can be found in online documentation in 'Fence Device Parameters'
- --rmfencedev <fence device name>
- Remove fence device
- --addfenceinst <fence device name> <node> <method> [options] [--nounfence]
- Add fence instance. Fence instance parameters can be found in online documentation in 'Fence Device Parameters'. Using --nounfence prevents ccs from automatically including an unfencing section for agents that require unfencing (ie. fence_scsi, fence_sanlock, etc.)
- --rmfenceinst <fence device name> <node> <method>
- Remove all instances of the fence device listed from the given method and node
- --addunfenceinst <fence device name> <node> [options]
- Add an unfence instance
- --rmunfenceinst <fence device name> <node>
- Remove all instances of the fence device listed from the unfence section of the node
FAILOVER DOMAIN OPERATIONS¶
- --lsfailoverdomain
- Lists all of the failover domains and failover domain nodes configured in the cluster
- --addfailoverdomain <name> [restricted] [ordered] [nofailback]
- Add failover domain
- --rmfailoverdomain <name>
- Remove failover domain
- --addfailoverdomainnode <failover domain> <node> [priority]
- Add node to given failover domain
- --rmfailoverdomainnode <failover domain> <node>
- Remove node from failover domain
SERVICE OPERATIONS¶
- --lsserviceopts [service type]
- List available services. If a service type is specified, then list options for the specified service type
- --lsresourceopts [service type]
- An alias to --lsserviceopts
- --lsservices
- List currently configured services and resources in the cluster
- --addresource <resource type> [resource options] ...
- Add global cluster resources to the cluster Resource types and variables can be found in the online documentation under 'HA Resource Parameters'
- --rmresource <resource type> [resource options]
- Remove specified resource with resource options
- --addservice <servicename> [service options] ...
- Add service to cluster
- --rmservice <servicename>
- Removes a service and all of its subservices
- --addaction <resource/name> <action_name> <action_option=val>
- Add an action to the specified resource.
- --rmaction <resource name> [<action_name> [action options]]
- Remove all actions from resource, or actions matching action name and options specified
- --addvm <virtual machine name> [vm options] ...
- Adds a virtual machine to the cluster
- --rmvm <virtual machine name>
- Removes named virtual machine from the cluster
- --addsubservice <servicename> <subservice> [service options] ...
- Add individual subservices, if adding child services, use ':' to separate
parent and child subservices and brackets to identify subservices of the
same type
Subservice types and variables can be found in the online documentation under 'HA Resource Parameters'
To add a nfsclient subservice as a child of the 2nd nfsclient subservice in the 'service_a' service use the following example:
ccs <...> --addsubservice service_a nfsclient[1]:nfsclient ref=/test - --rmsubservice <servicename> <subservice>
- Removes a specific subservice specified by the subservice, using ':' to
separate elements and brackets to identify between subservices of the same
type.
To remove the 1st nfsclient child subservice of the 2nd nfsclient subservice in the 'service_a' service, use the following example:
ccs <...> --rmsubservice service_a nfsclient[1]:nfsclient
QUORUM OPERATIONS¶
- --lsquorum
- List quorum options and heuristics
- --setquorumd [quorumd options] ...
- Add quorumd options
- --addheuristic [heuristic options] ...
- Add heuristics to quorumd
- --rmheuristic [heuristic options] ...
- Remove heuristic specified by heurstic options
MISC OPTIONS¶
- --lsmisc
- List all of the misc options
- --settotem [totem options]
- Set totem options
- --setuidgid uid=<uid> gid=<gid>
- Set uidgid options
- --rmuidgid uid=<uid> gid=<gid>
- Remove uidgid entry matching specified uid/gid
- --setdlm [dlm options]
- Set dlm options
- --setrm [resource manager options]
- Set resource manager options
- --setcman [cman options]
- Set cman options
- --setmulticast [multicast address] [multicast options]
- Sets the multicast address to use (or removes it if no multicast address is given)
- --setaltmulticast [alt multicast address] [alt multicast options]
- Sets the alt multicast address to use (or removes it if no alt multicast address is given)
- --setfencedaemon [fence daemon options]
- Set fence daemon options
- --setlogging [logging options]
- Set logging options
- --addlogging [logging daemon options]
- Add a logging daemon (see cluster.conf for options)
- --rmlogging [logging daemon options]
- Remove the logging daemon with the specified options
FILES¶
~/.ccs/cacert.config ~/.ccs/cacert.pem ~/.ccs/privkey.pem
EXAMPLES¶
Create and start a 3 node cluster with apc fencing:
ccs -h host1 --createcluster mycluster
ccs -h host1 --addnode host1
ccs -h host1 --addnode host2
ccs -h host1 --addnode host3
ccs -h host1 --addmethod primary host1
ccs -h host1 --addmethod primary host2
ccs -h host1 --addmethod primary host3
ccs -h host1 --addfencedev myfence agent=fence_apc ipaddr=192.168.0.200
login=apc passwd=apc
ccs -h host1 --addfenceinst myfence host1 primary port=1
ccs -h host1 --addfenceinst myfence host2 primary port=2
ccs -h host1 --addfenceinst myfence host3 primary port=3
ccs -h host1 --sync --activate
ccs -h host1 --startall
AUTHOR¶
Chris Feist <cfeist@redhat.com>
SEE ALSO¶
NOTES¶
Fence Device Parameters
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Cluster_Administration/ap-fence-device-param-CA.html
HA Resource Parameters
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Cluster_Administration/ap-ha-resource-params-CA.html