This section describes cluster management-related requests.
All database examples provided in this guide are specific to MySQL. If you use a different database, use syntax that is appropriate for your respective database. |
createServerComment
Creates a new server comment. Server comments are displayed on the Home page of the Cluster Manager machine.
deleteLicense
deleteMessage
Deletes a specific message, including all dependent records. Messages are listed in the Cluster Manager interface Messages tab and generally are notifications about issues with agents or the Cluster Manager.
deleteMessages
Deletes a set of messages, including all dependent records.
Optional arguments
Field | Description |
---|---|
filter |
A SQL query used to limit the result set. For a list of possible SQL values, see the getMessages command. |
exportData
Exports Cluster Manager data to a file. This is a full database dump, which might take substantial time for a large database.
NOTE:
-
Manual migration using the
exportData
andimportData
commands is recommended only to replicate data between Cluster Manager instances running the same version of the Cluster Manager. For example, you could use it to make a backup dump and then restore it to a new instance of the Cluster Manager of the same version. -
For a very large database, you should work with a database administrator to use the native database export/import facilities rather than using
cmtool
.
Required arguments
Field | Description |
---|---|
fileName |
Target file name or path. If you use a file name, the destination is the current working directory of the Java process ( |
getLicense
getMessage
Retrieves a particular message.
Required arguments
Field | Description |
---|---|
messageId |
The numeric value that uniquely identifies each message. |
Result tags
See the getMessages command for descriptions.
agentIdagentNamebuildIdbuildNamecreateTimemessageIdseveritytext
getMessages
Retrieves a list of messages
Optional arguments
Field | Description | ||
---|---|---|---|
filter |
A SQL query used to limit the result set. See the possible values below.
|
||
maxResults |
The maximum number of elements to return from a query. |
||
firstResult |
The starting index for the query result set.
|
||
order |
A SQL order by clause. Used to specify ordering for the query result set. |
||
profile |
Can be details or info. This is the level of detail to return from a query; details gets all information and info gets a reduced information set. |
Result tags and SQL query names
Field | Description |
---|---|
agentId |
A unique, internal number assigned to each agent by the Cluster Manager; this number can change. SQL query name for |
agentName |
A name defined by the host where the agent resides (numbers and/or letters). SQL query name for |
buildId |
A unique number assigned by the Cluster Manager for each build. SQL query name for |
buildName |
The build name that is the expanded build class tag. SQL query name for |
createTime |
The time when the item was created. SQL query name for |
messageId |
The numeric value that uniquely identifies each message. SQL query name for |
severity |
The severity level of the event: Info, Warning, or Error. For |
text |
The text of the item. SQL query name for |
getResourceStats
Retrieves resource usage statistics.
Optional arguments
Field | Description | ||
---|---|---|---|
filter |
A SQL query used to limit the result set. See the possible values below.
|
||
maxResults |
The maximum number of elements to return from a query. |
||
firstResult |
The starting index for the query result set.
|
||
order |
A SQL order by clause. Used to specify ordering for the query result set. |
||
profile |
Can be details or info. This is the level of detail to return from a query; details gets all information and info gets a reduced information set. |
Result tags and SQL query names
Field | Description |
---|---|
agentClusterShortage |
How many additional agents could have been used by the builds over the specified time period. This value is filled in only for cluster statistics—it is not available for individual resource statistics. SQL query name for |
agentDemand |
The average number of agents all builds could have used if those agents were available. For example, if two builds use two different resources and each build could use 15 agents, the cluster load shows an Agent Demand of 30 agents, and each resource shows 15. SQL query name for |
agentLicenseShortage |
The difference between the maximum request for agents by all builds and the number of agents the license allows. SQL query name for |
agentsAvailable |
The average number of enabled and active agents in the cluster over the specified time period. This value is available only for cluster statistics—it is not available for individual resource statistics. SQL query name for |
agentsInUse |
The total number of agents assigned to builds. SQL query name for |
availableResults |
This is a count of ’max’ or ’first’ results if |
buildsDuration |
The average amount of time the current builds have been running. SQL query name for |
buildsRunning |
Average number of simultaneous builds running during a specific time period. SQL query name for |
createTime |
The time when the item was created. SQL query name for |
duration |
The number of milli-seconds the build has been running. SQL query name for |
resourceName |
This name is used on the eMake parameter: |
resourceStatId |
The resource ID number that uniquely identifies every resource. SQL query name for |
getServer
Retrieves server configuration.
Result tags
Field | Description |
---|---|
agentAllocationPolicy |
Defined as either exclusive or shared. |
agentLockTimerSec |
When jobs run beyond this number of seconds, the agent should be locked. |
badAgents |
The number of enabled agents with a bad status. |
disabledAgents |
The number of disabled agents. |
emailInterval |
The number of minutes between email notifications. |
emailItemLimit |
Maximum number of messages per email notification. |
goodAgents |
The number of enabled agents with a good status. |
logDaysToKeep |
The number of days to keep message log entries. |
lsfAvailable |
No longer used. False (0) is always returned. |
mailFrom |
The value to use in the From header element. |
mailPrefix |
The string used to prefix subject lines. |
maxAgents |
The maximum number of agents to request for this build. |
maxClockSkew |
The maximum clock skew (in seconds) allowed between the eMake client and agents in the cluster. |
minAgents |
The minimum number of agents required for this build to run. |
preemptionPolicy |
The allocation preemption policy. |
priority |
The build priority level. When assigning resources, an optional priority boost value can be selected to give a build class preference over other builds of the same priority level. Higher boost values correspond to greater preference. |
resourceStatInterval |
In minutes, the interval to collect stats on resource usage. |
resourceStatKeep |
The number of minutes of resource usage statistics to keep. |
runningBuilds |
The number of incomplete builds in the system. |
getServerComments
Retrieves a list of related server comments.
getVersion
importData
Imports Cluster Manager data from a file. This command imports a full database dump, which might take substantial time for a large database.
NOTE:
-
You must manually delete any old or unused agents from the agents list.
-
You must update the license file after importing it, if it has expired.
-
Manual migration using the
exportData
andimportData
commands is recommended only to replicate data between Cluster Manager instances running the same version of the Cluster Manager. For example, you could use it to make a backup dump and then restore it to a new instance of the Cluster Manager of the same version. -
For a very large database, you should work with a database administrator to use the native database export/import facilities rather than using
cmtool
.
logMessage
Creates a custom message on the Cluster Manager Messages page.
Optional arguments
If --buildId and --agentName are on the same line, the message is applied to the build and the agent name.
|
Field | Description |
---|---|
severity |
Can be Debug, Info, Warning, or Error. You can also use 0, 1, 2, or 3. |
buildId |
The message applies to this specified build only. |
agentName |
The message applies to this specified agent name only. |
modifyServer
Modifies the server configuration.
Optional arguments
Field | Description |
---|---|
priority |
The default priority value is 120 (normal). 220 is high and 20 is low. Priority value can be adjusted up or down by 1-10 to “boost” the priority to give certain build classes preference over other builds of the same priority level. Higher boost values correspond to greater preference. |
emailInterval |
The number of minutes between email notifications. |
emailItemLimit |
The maximum number of messages per email notification. |
agentAllocationPolicy |
Can be exclusive or shared. Exclusive means all agents on a specific machine are assigned to the same build. Shared means all agents on the same machine can be assigned to different builds. This policy requires that eMake client and agent machines have synchronized clocks. |
wideDeepAllocationPolicy |
Can be deep or wide. Deep means the agent allocation algorithm favors assigning more agents on the same host to a build. Wide means the algorithm favors assigning more agents from different hosts. If wide, be sure |
preemptionPolicy |
The allocation preemption policy. |
maxClockSkew |
The maximum clock skew (in seconds) allowed between the eMake client and agents in the cluster. |
maxAgents |
The maximum number of agents to request for this build. |
minAgents |
The minimum number of agents required for this build to run. |
mailFrom |
The value to use in the From header element. |
mailPrefix |
The string used to prefix subject lines. |
logDaysToKeep |
The number of days to keep message log entries. |
resourceStatInterval |
In minute units, this is the interval to collect statistics on resource usage. |
resourceStatKeep |
The number of days of Resource usage statistics to keep. |
agentLockTimerSec |
The number of seconds an agent is locked from being taken by another build. This interval provides a buffer so the current build can perform any housekeeping tasks before being taken over by a subsequent build. |
reportUsage |
Flag to control whether the cluster manager reports build performance metrics to CloudBees. If true, metrics are reported. |
buildAccessLimited |
Flag to control whether access to build information is limited to the owner of the build and the administrator only. If true, build information is limited. |
serverIpAddress |
The IP address of the Cluster Manager server used for agent connection during cloud bursting. |
testAgents
Instructs the Cluster Manager to contact each active agent and update its status.
Optional arguments
Field | Description |
---|---|
agentId |
A unique, internal number that can change; assigned by the Cluster Manager. |
agentName |
The name defined by the host where the agent resides (numbers and/or letters). |
filter |
A SQL query used to limit the result set. For a list of possible SQL values, see the getAgents command. |