This chapter contains a detailed list of all RMS warnings that appear in the switchlog.
Check displayed component names of messages and then see the table below to determine references. The component names are explained in numerical order of messages.
Component | Reference |
|---|---|
ADC | |
ADM | |
BAS | |
BM | |
CTL | |
CUP | |
DET | |
SCR | |
SWT | |
SYS | |
UAP | |
US | |
WLT | |
WRP |
Content:
Information message.
Corrective action:
No action is required.
Content:
A file to be sent to a remote node cannot be opened.
Corrective action:
Check the error text <errortext> or other WARNING/ERROR messages.
Content:
A file to be sent to a remote node cannot be read.
Corrective action:
Message (ADC, 23) is also output. Check the error test of (ADC, 23) <errortext> or other WARNING/ERROR messages.
Content:
The hvshut command was timed out.
When the hvshut command is executed with either -l/-s/-a option, some resources that are included in cluster applications may fail to stop.
Corrective action:
To prevent the timeout of the hvshut command, depending on your environment, change RELIANT_SHUT_MIN_WAIT, which is the global environment variable of RMS, to a larger value.
See
For details on RELIANT_SHUT_MIN_WAIT, see "RELIANT_SHUT_MIN_WAIT" in "Global environment variables" of the following manual below:
For PRIMECLUSTER 4.3A30 or later: "PRIMECLUSTER Reliant Monitor Services (RMS) with Wizard Tools Configuration and Administration Guide."
See "PRIMECLUSTER Reliant Monitor Services (RMS) with Wizard Tools Configuration and Administration Guide" for how to refer to and change the RMS environment variables.
Take either of following actions depending on with which option the hvshut command has been executed.
With -l option
Shut down the OS on the node on which the command has been executed, or stop the node forcibly.
With -s option
Shut down the OS on the target node of the command, or stop the node forcibly.
With -a option
Shut down the OS on all the nodes except a node on which RMS has ended normally, or stop the node forcibly.
With -L option
When the BM (base monitor) process does not stop on the node on which the command has been executed, execute the hvshut -f command to stop RMS forcibly. No action is required when the BM process stops.
With -A option
When the BM process does not stop on some nodes, execute the hvshut -f command on these nodes to stop RMS forcibly. No action is required when the BM process stops on all the nodes.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
A switch request cannot be performed for a userApplication in the Deact State
Corrective action:
Activate the userApplication and issue the switch request again.
Content:
The hvswitch command has been executed for a currently shutdown node.
Corrective action:
Start the target node to execute the switch request again or select other node.
Content:
RMS is shut down even though a resource is not offline.
Corrective action:
Shut down the OS or stop the node forcibly.
Content:
Information message.
Corrective action:
No action is required.
Content:
The target node of the switch request is responding to an earlier shutdown request. The switch request is cancelled.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
The hvshut command was timed out.
When the hvshut command is executed with either -l/-s/-a option, some resources that are included in cluster applications may fail to stop.
Corrective action:
To prevent the timeout of the hvshut command, depending on your environment, change RELIANT_SHUT_MIN_WAIT, which is the global environment variable of RMS, to a larger value.
See
For details on RELIANT_SHUT_MIN_WAIT, see "RELIANT_SHUT_MIN_WAIT" in "Global environment variables" of the following manual below:
For PRIMECLUSTER 4.3A30 or later: "PRIMECLUSTER Reliant Monitor Services (RMS) with Wizard Tools Configuration and Administration Guide."
See "PRIMECLUSTER Reliant Monitor Services (RMS) with Wizard Tools Configuration and Administration Guide" for how to refer to and change the RMS environment variables.
Take either of following actions depending on with which option the hvshut command has been executed.
With -l option
Shut down the OS on the node on which the command has been executed, or stop the node forcibly.
With -s option
Shut down the OS on the target node of the command, or stop the node forcibly.
With -a option
Shut down the OS on all the nodes except a node on which RMS has ended normally, or stop the node forcibly.
With -L option
When the BM (base monitor) process does not stop on the node on which the command has been executed, execute the hvshut -f command to stop RMS forcibly. No action is required when the BM process stops.
With -A option
When the BM process does not stop on some nodes, execute the hvshut -f command on these nodes to stop RMS forcibly. No action is required when the BM process stops on all the nodes.
Content:
Information message.
Corrective action:
No action is required.
Content:
A userApplication userApplication failed to switch to the Offline state while RMS is being shut down. In this case, the switch request is cancelled even if the ShutDown option is specified for the AutoSwitchOver attribute.
Corrective action:
Check if RMS was already shut down and the switch request is cancelled. After that, switch the userApplication userApplication manually. Check logs for why the userApplication failed to switch to the Offline state.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
The target node of the switch request is responding to an earlier shutdown request. The switch request is cancelled.
Corrective action:
No action is required.
Content:
The offline processing for the object <object> failed, and the object is still partially online, so the switch request is cancelled.
Corrective action:
Check the log files to see why the offline processing of the object <object> failed.
Content:
The object <object> has no rName attribute. This attribute is required for a generic RMS detector; however, it may not exist in a custom detector.
Corrective action:
No action is required if the corresponding custom detector is properly designed. However, if the generic detector is soon used or will be used with this object, specify the rName attribute.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
When RMS encounters some problems in transmitting the message message to some other host in the cluster, it prints this message. This could be due to the fact that the RMS on the other host is down or there might be a network problem.
Corrective action:
Make sure that the RMS is running on the other hosts in the cluster and no network issues exist.
When fjsnap command, pclsnap command, or hvdump command is executed while some nodes that configure a cluster are stopped, this message may be printed. In this case, no action is required.
Content:
The hvutil -f/-c command is ignored because the userApplication <userapplication> is controlled by a scalable application.
Corrective action:
Use the command for a scalable application.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
Information message.
Corrective action:
No action is required.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
The socket() call failed to allocate a port for rmshb.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
The listen() system call failed to call rmshb port.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
Make sure remotehost is running and that communication between the two hosts is possible. Check if communication with the remote node is possible by using a standard method such as ping. After that, restart RMS on the local node.
Corrective action:
The communication between two hosts should be completely established. After that, restart the RMS monitor.
Content:
RMS was unable to set the close-on-exec flag using fcntl.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
The hvutil -d command is ignored because the userApplication <userapplication> is controlled by a scalable application.
Corrective action:
Use the command for a scalable application.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
When the environment variable HV_MLOCKALL is set to 1, the base monitor process and a memory allocated by the base monitor process are fixed. This message is output to indicate that the base monitor failed to lock the memory. In this case, RMS uses the unlocked memory to keep running.
Corrective action:
See the error text to find the cause. Check if the memory is sufficient.
Content:
Information message.
Corrective action:
No action is required.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
In a priority list, priority of the destination node of a userApplication is configured. When the priority list is renewed by starting or stopping of RMS, the priority list is synchronized to be consistent between the nodes However, when the priority list between the nodes is not consistent temporary, this message is printed.
Corrective action:
No action is required.
Content:
A switch request was cancelled because processing of a current online host contract is not yet settled.
Corrective action:
If the userApplication didn't go online, invoke a manual switch request.
Content:
The offline processing for the userApplication failed, and the userApplication is still partially online, so the switch request iscancelled.
Corrective action:
Check the log files to see why the offline processing failed.
Content:
A switch request to a SysNode has been executed. However, the userApplication userApplication cannot be switched to Online on the SysNode.
Corrective action:
Execute the switch request again after the userApplication userApplication can be switched to Online on the SysNode.
Content:
A switch request has been executed. However, No SysNode that includes the userApplication userApplication, which can be switched to Online state, exists.
Corrective action:
Execute the switch request again when a SysNode that includes the userApplication userApplication, which can be switched to Online state, exists.
Content:
A timeout occurred during the contract processing.
Corrective action:
If the userApplication didn't eventually go online, make sure that the userApplication is not online on any of the other nodes, and then invoke a manual switch request.
Content:
Information message.
Corrective action:
No action is required.
Content:
The offline processing for the userApplication failed and the userApplication is still partially online, so the switch request is cancelled.
Corrective action:
Check the log files to see why the offline processing failed.
Content:
Although a local Inconsistent state existed, the current online host request with the forced switch option ('hvswitch -f') has been accepted. The local inconsistency has been overridden.
Corrective action:
No action is required.
Content:
The current online host request is denied due to a local Inconsistent state.
Corrective action:
Clear the Inconsistent state first.
Content:
The application is currently online on the local host but is inconsistent on another host. The application is switched to another host with the forced switch option to override the inconsistency.
Corrective action:
No action is required.
Content:
The AutoStartUp processing is cancelled due to the Inconsistent state.
Corrective action:
Clear the Inconsistent state.
Content:
The failover processing is cancelled due to the Inconsistent state.
Corrective action:
Clear the Inconsistent state.
Content:
The switch request is cancelled due to the Inconsistent state.
Corrective action:
Clear the Inconsistent state.
Content:
The switch request is cancelled due to the Inconsistent state.
Corrective action:
Clear the Inconsistent state.
Content:
Although a state is inconsistent, a switch request with the forced switch option ('hvswitch -f') is accepted and the local inconsistency has been overridden.
Corrective action:
No action is required.
Content:
The userApplication is currently in an Inconsistent state on the local host. The application cannot be switched until the inconsistency is resolved, so the switch request is cancelled.
Corrective action:
Clear the Inconsistent state.
Content:
A LastOnlineHost conflict is detected and the local host is the LastOnlineHost, so the application will be brought online on the local host.
Corrective action:
No action is required.
Content:
A LastOnlineHost conflict is detected and the local host is not the LastOnlineHost, so the application will be brought online on the other host.
Corrective action:
No action is required.
Content:
A LastOnlineHost conflict is detected, and RMS cannot determine the LastOnlineHost, so the application will not go online anywhere.
Corrective action:
Invoke a switch request specifying the target host.
Content:
A LastOnlineHost conflict is detected, and the timestamps of conflicting LastOnlineHost entries do not allow a safe decision because their difference is lower than HV_LOH_INTERVAL. Therefore, the application will not go online anywhere.
Corrective action:
Invalidate the LastOnlineHost entry with 'hvutil -i <userapplication>', and then invoke a switch request specifying the target host.
Content:
A maintenance mode request, i.e., 'hvutil -m on/off' is denied because the userApplication is busy or is in the Faulted state.
Corrective action:
Clear the Faulted state and retry the maintenance mode request.
Content:
A maintenance mode request, i.e., 'hvutil -m on/off', is denied because the userApplication is busy, is in the Faulted state, or is not ready to leave maintenance mode.
Corrective action:
See the remote switchlog for details.
Content:
A maintenance mode request ('hvutil -m on/off') is denied because the resources are not in an appropriate state for safely returning to normal operation.
Corrective action:
Fix the states of the listed resources.
Content:
A maintenance mode request ('hvutil -m on/off') is denied because the initialization of the state of the userApplication is not yet complete.
Corrective action:
Wait for the initialization of the state of the userApplication and retry the maintenance mode request.
Content:
A LastOnlineHost conflict is detected, and the local host is the LastOnlineHost, so the application will be brought online on the local host.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
The WarningScript of the resource <resource> has ended abnormally with the status <status>.
Corrective action:
Investigate if the WarningScript that is set for the resource <resource> has problems.
Content:
StateChangeScript exited with exit code n.
Corrective action:
Check that there are no problems with the controller that notified exit code n and is set by StateChangeScript.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
This message indicates that AutoStartup=1, PartialCluster=0, and not all the nodes were started within the HV_AUTOSTART_WAIT time.
Corrective action:
No action is required.
Content:
The AutoStartUp is cancelled due to the Faulted state.
Corrective action:
Clear the Faulted state.
Content:
The AutoStartUp is cancelled due to the Faulted state.
Corrective action:
Clear the Faulted state.
Content:
The AutoStartUp is cancelled because the userApplication is in the Deact state.
Corrective action:
Activate the userApplication and start the application manually.
Content:
The AutoStartUp is cancelled because the PartialCluster attribute is set to 0 and not all necessary cluster hosts are online.
Corrective action:
Start RMS on all necessary cluster hosts and then start the application manually if necessary.
Content:
The switch request is cancelled because no node to which a userApplication can be switched is available.
Corrective action:
Execute the switch request again after the node to which userApplication can be switched. The destination node should include userApplications that are in Offline or Standby state.
Content:
The switch request is cancelled because <object> is either busy or locked.
Corrective action:
Wait until <object> is in a switchable state and then issue the request again.
Content:
The switch request is cancelled because not all necessary cluster hosts for the application are online.
Corrective action:
If the application should be brought online anyway, use the forced switch option ('hvswitch -f').
Note
A forced application switch overrides all safety checks and could therefore result in data corruption or other inconsistencies.In PRIMECLUSTER 4.3A10 or later (Solaris)or PRIMECLUSTER 4.3A30 or later (Linux), RMS may kill the node on which RMS is not running before starting the application to reduce the risk of data corruption when the Forced switch request of an application is issued.
Content:
The switch request is cancelled because the application has been deactivated.
Corrective action:
Activate the application and then issue the request again.
Content:
The target host was either not found or not ready to go online, so the switch request is cancelled.
Corrective action:
Wait for the target host to go online or start the target host.
Content:
The switch request is cancelled because the application or the local host is in a transitional state.
Corrective action:
Wait until both the application and the local host are online and then issue the request again.
Content:
For a priority or 'last online host' switch, if the target host of the switch is the node where the application is faulted, then the switch request is denied and the switch request is forwarded to another host in the cluster.
Content:
The switch request is cancelled because the local node has Faulted or OfflineFaulted descendants and no other node is ready to go online.
Corrective action:
Clear the Faulted/OfflineFaulted state.
Content:
Information message.
Corrective action:
No action is required.
Content:
The Deact request cannot be processed because the target application is in busy or locked state.
Corrective action:
Wait until the target application status is changed, and then execute the Deact request again.
Content:
The switch request cannot be processed because the target application is in the Deact state.
Corrective action:
Activate the target application.
Content:
Information message.
Corrective action:
No action is required.
Content:
The object was Online on the remote node onlinehost, but it's currently inoperable. This could occur due to a previous shutdown of onlinehost via 'hvshut -f', or it could be a timing issue. The switch request is cancelled to protect data.
Corrective action:
If the application should be brought online anyway, use the forced switch option ('hvswitch -f').
Note
A forced application switch overrides all safety checks and could therefore result in data corruption or other inconsistencies.
In PRIMECLUSTER 4.3A10 or later (Solaris) or PRIMECLUSTER 4.3A30 or later (Linux), RMS may kill the node on which RMS is not running before starting the application to reduce the risk of data corruption when the Forced switch request of an application is issued. If the previous shutdown of onlinehost was not via 'hvshut -f', this could be a timing issue, so wait a moment and try it again.
Content:
The object was Online on the remote node onlinehost, but it's currently inoperable. However, the switch request is processed because the forced switch option ('hvswitch -f') is used.
Corrective action:
No action is required.
Content:
The application is currently in an Inconsistent state on the local host. The application cannot be switched until the inconsistency is resolved, so the switch request is cancelled.
Corrective action:
You can either clear the inconsistency first, or you can override this restriction by using the forced switch option ('hvswitch -f').
Note
A forced application switch overrides all safety checks and could therefore result in data corruption or other inconsistencies.
In PRIMECLUSTER 4.3A10 or later (Solaris) or PRIMECLUSTER 4.3A30 or later (Linux), RMS may kill the node on which RMS is not running before starting the application to reduce the risk of data corruption when the Forced switch request of an application is issued.
Content:
The application is currently in an Inconsistent state on the local host. The application cannot be switched until the inconsistency is resolved, so the switch request is cancelled.
Corrective action:
You can either clear the inconsistency first, or you can override this restriction by using the forced switch option ('hvswitch -f').
Note
A forced application switch overrides all safety checks and could therefore result in data corruption or other inconsistencies.
In PRIMECLUSTER 4.3A10 or later (Solaris) or PRIMECLUSTER 4.3A30 or later (Linux), RMS may kill the node on which RMS is not running before starting the application to reduce the risk of data corruption when the Forced switch request of an application is issued.
Content:
The userApplication is not ready to go online on the local host, so RMS forwards the switch request to the next host in its priority list.
Corrective action:
No action is required.
Content:
The userApplication is not ready to go online on the local host so the direct switch request is cancelled.
Corrective action:
Make sure that the userApplication is in Offline or Standby state on the local host.
Content:
The sysnode is in the Wait state, so the switch request is cancelled.
Corrective action:
Wait for the node to get out of the Wait state and try the switch request again.
Content:
Information message.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
During a policy switch, if an exclusive application switches to a node, then all applications in the Standby state must go offline because they have a lower priority. This message simply warns the user that the application is in the Standby state and will be going offline due to the above reason.
Corrective action:
No action is required.
Content:
The application doesn't start up automatically because the environment variable HV_AUTOSTARTUP is set to 0, and this overrides each application's AutoStartUp attribute.
Corrective action:
To allow application startup according to each application's AutoStartUp attribute, set the environment variable HV_AUTOSTARTUP to 1.
Content:
The maintenance mode request from the controlling userApplication is denied because the state is either Faulted or Deact or the application is busy or locked.
Corrective action:
Clear the Faulted or Deact state and try it again.
Content:
The SysNode of RMS is not consistent with the format <`uname -n`>RMS. On Oracle Solaris zone environments, in the configuration in which PRIMECLUSTER is used, when the host name in the non-global zone includes capital letters, this message may be printed when RMS in the non-global zone is started.
Corrective action:
No action is required.
Content:
The SysNode name of RMS is not consistent with <CFname>RMS.
Corrective action:
Change the SysNode name to <CFname>RMS.
Content:
As heartbeat between RMS's was lost and no responses were returned for <time > sec or more, forcible stop is executed.
Corrective action:
The following causes are possible. Take the necessary action according to the cause.
Cluster interconnect cannot communicate because of hardware failure. Remove the hardware failure cause by replacing the LAN card or cable.
High CPU load has been continued for long time to the degree RMS cannot process heartbeat. Remove the Review the process that the host on the SysNode <SysNode > has high load.
The clock was set back rapidly with NTP. Slowly adjust the clock with NTP.
Content:
Information message.
Corrective action:
No action is required.
Content:
This message gives a reason for skipping a particular action. This message is output when a process is executed by Cluster Admin or other function while some other process is executed. For example, it is generated to show that the Offline processing that has been requested during the PreCheck processing prior to the transition of the userApplication to Standby is ignored because the userApplication is currently executing the Standby processing.
Corrective action:
No action is required.
Content:
The HaltFlag attribute will be ignored if there are no more available hosts.
Corrective action:
Make sure that there is a sufficient number of available cluster hosts.
Content:
Information message.
Corrective action:
No action is required.
Content:
The userApplication is not ready to go online on the local node because it is busy or in the Faulted state.
Corrective action:
Clear the Faulted state.
Content:
Information message.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
During contract processing, the invalid token is received.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
The userApplication didn't start up automatically because not all necessary cluster hosts are online.
Corrective action:
Enable all necessary cluster hosts to be Online..
Content:
The userApplication is not ready to go online on local host so find another host.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
Information message.
Corrective action:
No action is required.
Content:
Further processing for < appli > will be stopped because of the double fault.
Corrective action:
Check the other messages in the switchlog to determine the reason for the double fault. Clear the double fault.
Content:
A PreCheckScript failed during a directed switch request, i.e., the target host of the request was explicitly specified. In this case the switch request is cancelled, so it is not forwarded to the next host in the priority list.
Corrective action:
Invoke a new switch request specifying the next host as target host. If you want RMS to forward the request automatically, you should invoke a priority switch (hvswitch without a specified target host).
Content:
A PreCheckScript failed during a priority switch request. In this case the switch request is forwarded to the next host in the priority list.
Corrective action:
No action is required.
Content:
Execution of the PreCheckScript has failed and standby processing will be stopped.
Corrective action:
Check to see why the PreCheckScript has failed and correct the script if necessary.
Content:
A PreCheckScript failed and the AutoSwitchOver attribute did not include the ResourceFailure option. In this case RMS will not take automatic action in the event of a script failure. The switch request is cancelled, and it is not forwarded to the next host in the priority list.
Corrective action:
Invoke a new switch request specifying the next host as the target. If you want RMS to forward the request automatically, turn on the ResourceFailure option of the AutoSwitchOver attribute.
Content:
A Clear request ('hvutil -c') was issued for an application <userapplication> in maintenance mode. It failed to clear the state of the graph and resulted in a Faulted state of the application.
Corrective action:
Check the switchlog for the origin of the failure. Fix the failure condition and re-run 'hvutil -c'. Do not leave maintenance mode until the fault condition has been cleared.
Content:
RMS internal error.
Corrective action:
Record this message, collect investigation information, and contact field engineers. For details on collecting the investigation information, see "PRIMECLUSTER Installation and Administration Guide."
Content:
A resource script failed to end normally.
Corrective action:
Check if the timeout has occurred in the script.
Content:
RMS exchanges messages between processes and hosts to maintain inter-host communication. If the delivery of a message has failed then this error is printed. This can occur if one or more hosts in the cluster are not active or if there is a problem with the network.
Corrective action:
(i) Check the other hosts in the cluster. If any are not alive, check the switchlog for information regarding why RMS has died on those hosts. Perform the following steps in order:
'hvdisp -a'
In the output of step 1., check if the state of any of the resources whose type is SysNode is offline. If so, that means that RMS is not running on that node.
Check the switchlogs of all the nodes that are offline to determine the reason why RMS on that node is not active.
(ii) If the other hosts that are part of the cluster are alive then that means there is some problem with the network.
When fjsnap command, pclsnap command, or hvdump command is executed while some nodes that configure a cluster are stopped, this message may be printed. In this case, no action is required.
Content:
RMS was unable to operation for certain seconds (n sec).
Corrective action:
This message may be generated by the temporary high load on CPU. The RMS will return to its normal operation when the load is reduced. You can usually ignore this message if the high CPU load lasts for only a short time.
Content:
The IP address of the interconnects <interconnect> and the existing interconnects <existinginterconnect> are the same.
Corrective action:
Check if different IP address is specified for each interconnects.
Content:
The echo service of UDP may not be valid on the local node.
Corrective action:
Check if the echo service is valid and activated.
