Multipath asynchronous events (SNMP traps) (When the Agent version of the server node is 1.x)
Failure event | Level | Event display | Troubleshooting |
|---|---|---|---|
One path blocked | Error | MPType access path(MPAccessPath) fault | Refer to the manual for the multipath driver and take required actions. Generally, the connection to storage devices and the cable connections must be checked. |
All paths blocked | Error | MPType access path(MPAccessPath) fault and no more path | |
A multipath asynchronous event (SNMP trap) was reported to Manager; however, Manager does not manage the corresponding multipath information. | Error | Multi path(Controler Controler Number) fault | Use the resource management view to search the server node related to the asynchronous event (SNMP trap). |
"MPType" of the event display is displayed, "MPLB" or "MPHD".
"MPAccessPath" of the event display is displayed, "c1t1" etc.
"Controller Number" of the event display is displayed, "No.2" etc.
Asynchronous event for monitored keywords
The system message file (/var/adm/messages) is monitored with monitoring keywords. When a message that includes monitoring keywords is output, an asynchronous event is sent to Manager according to the level corresponding to the monitoring keywords. In this case, the detected message containing the monitoring keywords is sent to Manager as is. If asynchronous events corresponding to the same message are detected several times within the message monitoring time (Polling Time in the Correlation.ini parameter), these events are reported to Manager only once.
Multipath asynchronous events (When the Agent version of the server node is 2.0 and 13.0 or later)
".*" in the monitoring keywords is a regular expression (which indicates that any characters appears 0 times and more).
Monitoring keyword | Level | Event display | Troubleshooting |
|---|---|---|---|
.*NOTICE: mphd.*I/O path switchover succeed.* | Warning | One-line message including monitoring keywords | Refer to the manual for the multipath driver and take required actions. Generally, the connection to storage devices and the cable connections must be checked. |
.*NOTICE: mplb.*I/O path failed, and standby.* | |||
.*NOTICE: mplb.*I/O path failed, and remaining online.* | |||
.*NOTICE: mplb.*disk controller connection is wrong.* | |||
.*NOTICE: mplb.*device connection is wrong.* | |||
.*WARNING: mphd.*I/O path for switch.* | |||
.*WARNING: mplb.*I/O path failed, no more.* | |||
.*WARNING: mplb.*connection of all paths is wrong.* |
EMC PowerPath asynchronous event (When the Agent version of the server node is 2.0 and 13.0 or later)
".*" in the monitoring keywords is a regular expression (which indicates that any characters appears 0 times and more).
Monitoring keyword | Level | Event display | Troubleshooting |
|---|---|---|---|
.*All paths to.*are dead.* | Warning | One-line message including monitoring keywords | Refer to the manual for the corresponding middleware product. Generally, the connection to storage devices and the cable connections must be checked. |
.*Path.*to.*is dead.* |
HITACHI JP1/HiCommand Dynamic Link Manager asynchronous event (When the Agent version of the server node is 2.0 and 13.0 or later)
".*" in the monitoring keywords is a regular expression (which indicates that any characters appears 0 times and more).
Monitoring keyword | Level | Event display | Troubleshooting |
|---|---|---|---|
.*All paths to.*are dead.* | Warning | One-line message including monitoring keywords | Refer to the manual for the corresponding middleware product. Generally, the connection to storage devices and the cable connections must be checked. |
.*KAPL08.*-E.* |
Safe/PRIMECLUSTER GDS, GFS series asynchronous event
".*" in the monitoring keywords is a regular expression (which indicates that any characters appears 0 times and more).
Monitoring keyword | Level | Event display | Troubleshooting |
|---|---|---|---|
NOTICE: sfx | Information | One-line message including monitoring keywords | Refer to the manual for the corresponding middleware product. |
WARNING: sfx | Warning | One-line message including monitoring keywords | Refer to the manual for the corresponding middleware product. Generally, the connection to storage devices and the cable connections must be checked. |
PANIC: sfx | Error | One-line message including monitoring keywords |
Sun MPxIO asynchronous event (Server node Agent version 13.2 or later)
".*" in the monitoring keywords is a regular expression (which indicates that any characters appears 0 times and more).
Monitoring keyword | Level | Event display | Troubleshooting |
|---|---|---|---|
.*/scsi_vhci/ssd.*multipath.*status:.*degraded.* | Warning | One-line message including monitoring keywords | Refer to the manual for the corresponding middleware product.Generally, the connection to storage devices and the cable connections must be checked. |
.*/scsi_vhci/ssd.*multipath.*status:.*failed.* |
Event reporting a changed host name
Manager uses the host name to identify each of the server nodes. However, in cases where server nodes cannot be identified correctly after their host name has changed, the following event is output:
Event | Level | Event display | Troubleshooting |
|---|---|---|---|
The host name of a server node has changed | Warning | hostname of server node (IPaddress) has been changed from Server-A to Server-B | The hostname for the server having the IP address indicated in the event display has been changed to Server-B. Reset the hostname to Server-A. To do so, edit /etc/inet/hosts file, which contains the definitions of the hostnames. |
The "IPaddress" indicated in the event display is the IP address of the corresponding server node.
"Server-A" is the name set to which the corresponding server node was originally set.
"Server-B" is the name to which the corresponding server node is currently set.
