How to monitor HCI message for the Delta PSU?
How to monitor HCI message for the Delta PSU?
The answer to this questions are two solutions:
Monitor the event log messages
Send a HCI command into a HX Delta matrix . A new HCI manual will be released in 3-4 weeks’ time.
Monitor the event log messages The alarm messages are sent only by HX Delta Matrix only.
The delta matrix reports the psu status automatically via the event log and you can monitor the event logs for the psu status message.The current PSU alarm state is checked automatically (polled, by the CPU card) every 12 minutes. An event log message is only output if there is an active alarm.
A good string to look for in the event log is “PSU Alarms Active”. When alarm happens we send an unsolicited event log message raised.
The format of the event is
<Matrix type> PSU Alarms Active <Semicolon delimited array of active alarm names>
The Matrix type strings are below
char *RACK_TYPE_STRING[] =
{
"Unknown",
"E-Pico",
"Median",
"Omega",
"Delta"
};
The alarms below are HX Delta only.
The Alarm strings are
“EXT. PSU 1 FAIL"
“EXT. PSU 2 FAIL”
“INT. PSU 1 FAIL”
“INT. PSU 2 FAIL”
“FAN 1 FAIL”
“FAN 2 FAIL”
“CONFIG FAIL”
“EXT. ALARM”
“OVERTEMP”
Send a HCI command into a Delta matrix
Using a HCI client you can send a message into the Delta matrix and request the frame status
This is done in the Matrix Status message, the format is (payload shown doesn’t include standard HCI header)
REQUEST_FRAME_STATUS message
Msg Id (16 bits) 97
Flags (1 byte) 0x8
HCI 2 ID bytes (64 bits) 0xABBACEDE
Schema (8 bits) 1
REPLY REQUEST_FRAME_STATUS message
Msg Id (16 bits) 98
Flags (1 byte) 0x8
HCI 2 ID bytes (64 bits) 0xABBACEDE
Schema (8 bits) 1
PSU Alarm Bits (16 bits) (see below)
Rack Temp (16 bits)
The PSU alarm bits are as follows
Ext Psu 1 fail - 0x0001
Ext Psu 2 fail - 0x0002
Int Psu 1 fail - 0x0004
Int Psu 2 fail - 0x0008
Fan 1 fail - 0x0010
Fan 2 fail - 0x0020
Config fail - 0x0040 // Fired by CPU App e.g. when master slave switchover occurs
Ext. Alarm - 0x0080 // alarm GPO state i.e. activated when core alarms bits are high
Overtemp - 0x0100
Related articles
We are looking for your help! Please consider sharing your stories, update an old solution or help us with a new one. Follow this link to share!