All MIBs › LSF-SNMP-MIB
Category: Domain: Server Hardware
Description: Platform LSF (Load Sharing Facility) SNMP MIB defining objects for monitoring cluster job scheduling, queues, and host utilization.
Start monitoring vendor-neutral, standards-based MIB, any SNMPv3-capable network device (SNMP engine identity/boot/clock status) with a free 30-day trial of IPNetwork Monitor. Import MIBs, browse SNMP OIDs, create custom SNMP monitors, configure alerts, and monitor any SNMP-enabled network device from a single console.
What Is LSF-SNMP-MIB?
LSF-SNMP-MIB is a legacy Platform Computing (enterprise OID 2766, Load Sharing Facility batch-scheduling cluster software) proprietary MIB mirroring the output of the lshosts/lsload commands. Its static table, keyed by IP, records hostname, type, model, CPU factor, CPU count, max memory/swap/temp-space, execution priority, is-server flag, host resources, and cluster-host count, while a dynamic table records host status, 15-second/1-minute/15-minute run-queue length, CPU utilization, paging rate, and I/O rate. As a performance/status MIB it lets an administrator watch per-host CPU/memory load and run-queue depth across an LSF compute cluster. It is deployed on Platform Computing LSF (Load Sharing Facility) compute-cluster hosts. Engineers can download the LSF-SNMP-MIB file directly to load it into their MIB browser.
IPNetwork Monitor allows you to monitor SNMP objects defined in LSF-SNMP-MIB. Use the built-in SNMP Monitoring Browser to explore available variables, view their current values and descriptions, and select the objects you want to monitor. You can then create SNMP monitors, configure thresholds, and receive alerts when monitored values change.
Supported Devices
- Platform Computing LSF (Load Sharing Facility) compute-cluster hosts
Monitoring Examples
An administrator checks lsfCPUUtilization and lsfOneMinuteRunQueue to watch load on a host within an LSF compute cluster.
What Can Be Monitored
- per-host CPU/memory load and run-queue depth
This MIB depends on
Related MIBs
Imported Objects
From RFC-1212
| OBJECT-TYPE |
From RFC-1215
| TRAP-TYPE |
From RFC1155-SMI
| IpAddress | |
| enterprises |
From RFC1213-MIB
| DisplayString |
OIDs
| OID symbolic | OID numeric | Type | Access | Description |
|---|---|---|---|---|
| lsbHostEntry | 1.3.6.1.4.1.2766.100.3.1.1 | not-accessible | One instance of batch host information. It is indexed by the host's IP address. | |
| IP lsbHostIp | 1.3.6.1.4.1.2766.100.3.1.1.1 | IpAddress | read-only | IP address of this host. |
| INT lsbHostMaximumJobs | 1.3.6.1.4.1.2766.100.3.1.1.5 | INTEGER | read-only | Maximum number of job slots that the host can provide in total. If the number of slots is unlimited, -1 will be returned. |
| STR lsbHostName | 1.3.6.1.4.1.2766.100.3.1.1.2 | DisplayString | read-only | Host name for this entry. |
| INT lsbHostNumberOfJobs | 1.3.6.1.4.1.2766.100.3.1.1.6 | INTEGER | read-only | Number of job slots used by jobs started on this host. |
| INT lsbHostReservedJobSlots | 1.3.6.1.4.1.2766.100.3.1.1.10 | INTEGER | read-only | Number of job slots used by pending jobs which have job slots reserved on the host. |
| INT lsbHostRunningJobs | 1.3.6.1.4.1.2766.100.3.1.1.7 | INTEGER | read-only | Number of job slots used by jobs running on this host. |
| INT lsbHostStatus | 1.3.6.1.4.1.2766.100.3.1.1.3 | INTEGER | read-write | Current status of this host. Only when a host is in ok status can batch jobs be dispatched to it. An SNMP management entity can close hosts by setting this variable to closed(4), and can reopen hosts by setting this variable to ok(1). |
| lsbHostTable | 1.3.6.1.4.1.2766.100.3.1 | not-accessible | A table which provides the same information as the bhosts command without any arguments. | |
| INT lsbHostUserJobLimit | 1.3.6.1.4.1.2766.100.3.1.1.4 | INTEGER | read-only | Maximum batch job slots that the host can provide per user. If the number of slots is unlimited, -1 will be returned. |
| INT lsbJobCpuUsage | 1.3.6.1.4.1.2766.100.3.3.1.11 | INTEGER | read-only | The cumulative total CPU time in seconds of all processes in a job. |
| lsbJobEntry | 1.3.6.1.4.1.2766.100.3.3.1 | not-accessible | An instance of a running job. The entries are indexed by jobID, since it uniquely identifies running jobs in the batch system. | |
| STR lsbJobExecutionHost | 1.3.6.1.4.1.2766.100.3.3.1.7 | DisplayString | read-only | The name of one or more hosts on which the job is executing. If the field is empty, the job has not been dispatched yet. If the execution host for a job has been removed from the configuration, this will be displayed as lost_and_found. |
| INT lsbJobId | 1.3.6.1.4.1.2766.100.3.3.1.1 | INTEGER | read-only | The number which uniquely identifies this job in the batch system. |
| INT lsbJobMemoryUsage | 1.3.6.1.4.1.2766.100.3.3.1.12 | INTEGER | read-only | The total resident memory usage in KBytes of all the processes in a job. |
| STR lsbJobName | 1.3.6.1.4.1.2766.100.3.3.1.2 | DisplayString | read-only | The job name assigned by the user, or the command string assigned by default. |
| STR lsbJobProcessGroupIds | 1.3.6.1.4.1.2766.100.3.3.1.9 | DisplayString | read-only | A space delimited list of process group ids which make up this job. |
| STR lsbJobProcessIds | 1.3.6.1.4.1.2766.100.3.3.1.10 | DisplayString | read-only | A space delimited list of process ids which make up this job. |
| STR lsbJobQueue | 1.3.6.1.4.1.2766.100.3.3.1.4 | DisplayString | read-only | The name of the queue to which the job belongs. |
| INT lsbJobStatus | 1.3.6.1.4.1.2766.100.3.3.1.5 | INTEGER | read-only | The current status of the job. |
| STR lsbJobSubmissionHost | 1.3.6.1.4.1.2766.100.3.3.1.6 | DisplayString | read-only | The name of the host from which the job was submitted. |
| STR lsbJobSubmitTime | 1.3.6.1.4.1.2766.100.3.3.1.8 | DisplayString | read-only | The submission time of the job. |
| lsbJobTable | 1.3.6.1.4.1.2766.100.3.3 | not-accessible | A table which provides the same information as the bjobs -r -u all command, and the resource usage of each running job. | |
| STR lsbJobUser | 1.3.6.1.4.1.2766.100.3.3.1.3 | DisplayString | read-only | The user who submitted the job. |
| INT lsbJobVirtualMemoryUsage | 1.3.6.1.4.1.2766.100.3.3.1.13 | INTEGER | read-only | The total virtual memory usage in KBytes of all the processes in a job. |
| lsbQueueEntry | 1.3.6.1.4.1.2766.100.3.2.1 | not-accessible | One instance of batch queue information. The index field is generated by the agent by sorting the list of queues, and numbering the queue names in ascending alphabetical order. | |
| INT lsbQueueHostJobLimit | 1.3.6.1.4.1.2766.100.3.2.1.9 | INTEGER | read-only | The maximum number of job slots which a host can process from this queue. This limit controls the number of jobs sent to each host, regardless of whether a host is a uniprocessor or a multiprocessor. If the number of slots is unlimited, -1 will be returned. |
| INT lsbQueueIndex | 1.3.6.1.4.1.2766.100.3.2.1.1 | INTEGER | read-only | A number generated by the agent to uniquely identify this queue. |
| INT lsbQueueIsActive | 1.3.6.1.4.1.2766.100.3.2.1.5 | INTEGER | read-write | A flag telling whether the queue is active (dispatching jobs to run), or whether it is inactive so that jobs are not being dispatched. The management entity can set this variable to active(1) or inactive(2) to change the dispatching status of this queue. |
| INT lsbQueueIsOpen | 1.3.6.1.4.1.2766.100.3.2.1.4 | INTEGER | read-write | A flag telling whether the queue is open to accept jobs submitted to it, or whether it is closed so that it will not accept jobs. The management entity can set this variable to open(1) or closed(2) to change the submission status of this queue. |
| INT lsbQueueMaximumJobs | 1.3.6.1.4.1.2766.100.3.2.1.6 | INTEGER | read-only | The maximum number of job slots which can be used by jobs from this queue. If the number of slots is unlimited, -1 will be returned. |
| STR lsbQueueName | 1.3.6.1.4.1.2766.100.3.2.1.2 | DisplayString | read-only | The name of this queue. |
| INT lsbQueueNumberOfJobs | 1.3.6.1.4.1.2766.100.3.2.1.10 | INTEGER | read-only | The total number of job slots which jobs in this queue are currently holding. |
| INT lsbQueuePendingJobs | 1.3.6.1.4.1.2766.100.3.2.1.11 | INTEGER | read-only | The number of pending job slots in this queue. |
| INT lsbQueuePriority | 1.3.6.1.4.1.2766.100.3.2.1.3 | INTEGER | read-only | The priority of the queue. It determines the order in which job queues are searched for jobs to dispatch. Queues with higher priorities are searched first. |
| INT lsbQueueProcessorJobLimit | 1.3.6.1.4.1.2766.100.3.2.1.8 | INTEGER | read-only | The maximum number of job slots which a processor can process from this queue. This limit controls the number of jobs sent to each host. The limit is configured per processor so that a multiprocessor host is automatically allowed to run more jobs. If the number of slots is unlimited, -1 will be returned. |
| INT lsbQueueRunningJobs | 1.3.6.1.4.1.2766.100.3.2.1.12 | INTEGER | read-only | The number of running job slots in this queue. |
| INT lsbQueueSystemSuspendedJobs | 1.3.6.1.4.1.2766.100.3.2.1.13 | INTEGER | read-only | The number of job slots suspended by the system in this queue. |
| lsbQueueTable | 1.3.6.1.4.1.2766.100.3.2 | not-accessible | A table which provides the same information as the bqueues command without any arguments. | |
| INT lsbQueueUserJobLimit | 1.3.6.1.4.1.2766.100.3.2.1.7 | INTEGER | read-only | The maximum number of job slots a user can use for his/her jobs in this queue. If the number of slots is unlimited, -1 will be returned. |
| INT lsbQueueUserSuspendedJobs | 1.3.6.1.4.1.2766.100.3.2.1.14 | INTEGER | read-only | The number of job slots suspended by the user in this queue. |
| INT lsbSystemSuspendedJobs | 1.3.6.1.4.1.2766.100.3.1.1.8 | INTEGER | read-only | Number of job slots used by jobs suspended by LSF on the host. |
| INT lsbUserSuspendedJobs | 1.3.6.1.4.1.2766.100.3.1.1.9 | INTEGER | read-only | Number of job slots used by jobs suspended by the job owner or an LSF administrator |
| lsfAgent | 1.3.6.1.4.1.2766.100 | |||
| lsfBatch | 1.3.6.1.4.1.2766.100.3 | |||
| lsfCluster | 1.3.6.1.4.1.2766.100.4 | |||
| STR lsfClusterName | 1.3.6.1.4.1.2766.100.4.1 | DisplayString | read-only | The name of the cluster for which this agent is returning information. |
| INT lsfCPUFactor | 1.3.6.1.4.1.2766.100.1.1.1.5 | INTEGER | read-only | A relative measure of this host's CPU speed. It is normally represented as a floating point number, and thus has been scaled by 100 for representation in the SNMP. |
| INT lsfCPUUtilization | 1.3.6.1.4.1.2766.100.1.2.1.6 | INTEGER | read-only | A percentage of how much CPU time is spent processing system and user processes. A high utilization means that a host's CPU is kept busy. |
| lsfDynamicEntry | 1.3.6.1.4.1.2766.100.1.2.1 | not-accessible | One instance (row) of host information. It is indexed by the host's IP address in order to provide a deterministic way of accessing a specific host's load information. | |
| IP lsfDynamicIpIndex | 1.3.6.1.4.1.2766.100.1.2.1.1 | IpAddress | read-only | IP address of this host as known by LSF. |
| lsfDynamicTable | 1.3.6.1.4.1.2766.100.1.2 | not-accessible | A table which corresponds to information displayed by the lsload command. | |
| lsfEvents | 1.3.6.1.4.1.2766.1000 | |||
| INT lsfExecutionPriority | 1.3.6.1.4.1.2766.100.1.1.1.10 | INTEGER | read-only | The execution priority, or nice level, of processes spawned by the RES on this host. |
| INT lsfFifteenMinuteRunQueue | 1.3.6.1.4.1.2766.100.1.2.1.5 | INTEGER | read-only | The 15 minute run queue length on this host. This value is a float which is scaled by 100 to format it as an integer. |
| INT lsfFifteenSecondRunQueue | 1.3.6.1.4.1.2766.100.1.2.1.3 | INTEGER | read-only | The 15 second run queue length on this host. This value is a float which is scaled by 100 to format it as an integer. |
| INT lsfFreeMemory | 1.3.6.1.4.1.2766.100.1.2.1.11 | INTEGER | read-only | The amount of free real memory on this host. The units are megabytes, scaled by 100. For example, 25 megabytes would be returned as 2500. |
| INT lsfFreeSwap | 1.3.6.1.4.1.2766.100.1.2.1.12 | INTEGER | read-only | The amount of free physical swap on this host. The units are megabytes, scaled by 100. For example, 25 megabytes would be returned as 2500. |
| INT lsfFreeTempSpace | 1.3.6.1.4.1.2766.100.1.2.1.13 | INTEGER | read-only | The amount of free space in the temporary filesystem. This is typically /tmp, but can be overridden in the LSF configuration files using the LSF_TMPDIR environment variable. The units are megabytes, scaled by 100. For example, 25 megabytes would be returned as 2500. |
| STR lsfHostName | 1.3.6.1.4.1.2766.100.1.1.1.2 | DisplayString | read-only | Host name for this host as known by LSF. |
| STR lsfHostResources | 1.3.6.1.4.1.2766.100.1.1.1.12 | DisplayString | read-only | A string containing the boolean resources applied to this host. Available boolean resources can be displayed using the lsinfo command. |
| lsfHosts | 1.3.6.1.4.1.2766.100.1 | |||
| INT lsfHostStatus | 1.3.6.1.4.1.2766.100.1.2.1.2 | INTEGER | read-write | The operating status of LSF on this host. The host can be locked from a management entity by setting this variable to lockU(4). If the host status is lockW(5), the status will become lockUW(6). The management entity can also unlock a host by attempting to set it to ok(1). |
| NTF lsfHostUnlicensed | 1.3.6.1.4.1.2766.1000.0.4 | This trap is sent when a host in the cluster becomes unlicensed. | ||
| INT lsfIdleTime | 1.3.6.1.4.1.2766.100.1.2.1.10 | INTEGER | read-only | The number of seconds for which all interactive login sessions have been idle on this host. |
| INT lsfIoRate | 1.3.6.1.4.1.2766.100.1.2.1.8 | INTEGER | read-only | A measure of number of disk operations on this host over the last lim sample interval. The value is a floating point number in LSF, and is thus scaled by 100 to represent it as an INTEGER. |
| INT lsfIsServer | 1.3.6.1.4.1.2766.100.1.1.1.11 | INTEGER | read-only | true if this host will accept and run remote processes, and false if it won't. |
| NTF lsfLimDown | 1.3.6.1.4.1.2766.1000.0.1 | This trap is sent when the lim on a host goes down. The variable points to the status of the host on which the lim went down. | ||
| INT lsfLoginSessions | 1.3.6.1.4.1.2766.100.1.2.1.9 | INTEGER | read-only | The number of login sessions on this host. |
| NTF lsfMasterElect | 1.3.6.1.4.1.2766.1000.0.5 | This trap is sent to announce which host has won the master election. | ||
| STR lsfMasterName | 1.3.6.1.4.1.2766.100.4.2 | DisplayString | read-only | The hostname of the host which is running as the LSF master server for this cluster. |
| NTF lsfMasterResign | 1.3.6.1.4.1.2766.1000.0.6 | This trap is sent to indicate that a host has resigned as the master host. | ||
| INT lsfMaxMemory | 1.3.6.1.4.1.2766.100.1.1.1.7 | INTEGER | read-only | The configured amount of RAM in this host, in megabytes. |
| INT lsfMaxSwap | 1.3.6.1.4.1.2766.100.1.1.1.8 | INTEGER | read-only | The configured amount of physical swap space on this host, in megabytes. |
| INT lsfMaxTempSpace | 1.3.6.1.4.1.2766.100.1.1.1.9 | INTEGER | read-only | The size of the temporary file space on this machine. Typically this corresponds to /tmp, but can be overridden in LSF configuration files by the LSF_TMPDIR environment variable. |
| NTF lsfMbdDown | 1.3.6.1.4.1.2766.1000.0.8 | This trap is sent if the mbatchd goes down in the cluster. It could be sent by mbatchd itself, or could be sent by some other agent which detects the condition. | ||
| NTF lsfMbdReconfig | 1.3.6.1.4.1.2766.1000.0.9 | This trap is sent by the mbatchd when it is going to reconfigure itself. | ||
| NTF lsfMbdUp | 1.3.6.1.4.1.2766.1000.0.7 | This trap is sent when an mbatchd comes up on a host, has done its initialization and replay, and is ready to start accepting and scheduling jobs. | ||
| STR lsfModel | 1.3.6.1.4.1.2766.100.1.1.1.4 | DisplayString | read-only | Model of this host. Available host models are defined in the lsf.shared file, and can be displayed using the lsinfo command. |
| INT lsfNumClusterHosts | 1.3.6.1.4.1.2766.100.1.3 | INTEGER | read-only | This is a count of how many hosts are configured into the cluster. This includes both servers and clients. It corresponds to the number of hosts returned in the ls_gethostinfo command, or the number of hosts listed by the lshosts command. |
| INT lsfNumCPU | 1.3.6.1.4.1.2766.100.1.1.1.6 | INTEGER | read-only | Number of CPUs in this host. |
| lsfNumericEntry | 1.3.6.1.4.1.2766.100.2.1.1 | not-accessible | One instance of resource information. It is indexed by two entries: host IpAddress and a resource index which is generated by the agent by sorting a list of resource names, and numbering the resource names in ascending alphabetical order. | |
| INT lsfNumericIndex | 1.3.6.1.4.1.2766.100.2.1.1.1 | INTEGER | read-only | A number generated by the agent to uniquely identify this resource name. |
| IP lsfNumericIP | 1.3.6.1.4.1.2766.100.2.1.1.2 | IpAddress | read-only | The IP address corresponding to the location of this resource. |
| STR lsfNumericLocation | 1.3.6.1.4.1.2766.100.2.1.1.3 | DisplayString | read-only | The host on which this resource is located. |
| STR lsfNumericName | 1.3.6.1.4.1.2766.100.2.1.1.4 | DisplayString | read-only | The name of this resource. |
| INT lsfNumericOrder | 1.3.6.1.4.1.2766.100.2.1.1.5 | INTEGER | read-only | Indicates the direction which denotes higher load. If a resource is ascending, the load is higher when the resource is higher (e.g. incoming packets). If the resource is descending, the load is higher when the resource is lower (e.g. space on a temporary filesystem). |
| lsfNumericTable | 1.3.6.1.4.1.2766.100.2.1 | not-accessible | A table which describes numeric type external indices and shared resources defined for a cluster. | |
| INT lsfNumericValue | 1.3.6.1.4.1.2766.100.2.1.1.6 | INTEGER | read-only | The current value of this resource. Note that floating point values will be scaled by 100 before being returned. |
| INT lsfOneMinuteRunQueue | 1.3.6.1.4.1.2766.100.1.2.1.4 | INTEGER | read-only | The 1 minute run queue length on this host. This value is a float which is scaled by 100 to format it as an integer. |
| INT lsfPagingRate | 1.3.6.1.4.1.2766.100.1.2.1.7 | INTEGER | read-only | A measure of page-in and page-out requests on this host over the lim sample interval. Since it is a floating point number in LSF, it is scaled by 100 to represent it as an INTEGER. |
| NTF lsfResDown | 1.3.6.1.4.1.2766.1000.0.2 | This trap is sent when the res on a host goes down. The variable points to the status of the host on which the res went down. | ||
| lsfResources | 1.3.6.1.4.1.2766.100.2 | |||
| NTF lsfSbdDown | 1.3.6.1.4.1.2766.1000.0.3 | This trap is sent when the sbatchd on a host goes down. The variable points to the status of the batch host on which the sbatchd went down. | ||
| lsfStaticEntry | 1.3.6.1.4.1.2766.100.1.1.1 | not-accessible | One instance (row) of host information. It is indexed by the host's IP address in order to provide a deterministic way of accessing a specific host's configuration information. | |
| IP lsfStaticIpIndex | 1.3.6.1.4.1.2766.100.1.1.1.1 | IpAddress | read-only | IP address of this host as known by LSF. |
| lsfStaticTable | 1.3.6.1.4.1.2766.100.1.1 | not-accessible | A table which corresponds to information displayed by the lshosts command. | |
| STR lsfType | 1.3.6.1.4.1.2766.100.1.1.1.3 | DisplayString | read-only | Type of this host. Available host types are defined in the lsf.shared file, and can be displayed using the lsinfo command. |
| NTF lsfWorkdirFull | 1.3.6.1.4.1.2766.1000.0.10 | This trap is sent if the mbatchd cannot write to the LSB_SHAREDIR because it is full. | ||
| platform | 1.3.6.1.4.1.2766 |
RFC description
Host, resource, batch job, and cluster management information for IBM LSF (Load Sharing Facility) systems.
Start monitoring Platform Computing LSF (Load Sharing Facility) compute-cluster host with a free 30-day trial of IPNetwork Monitor. Import MIBs, browse SNMP OIDs, create custom SNMP monitors, configure alerts, and monitor any SNMP-enabled network device from a single console.