WebesNotification is not send after each event

[Witas]

Hi All,

Once we did not receive any SNMP trap from OpenVMS when power supply failed on MSA1000 and now performing the tests (power off/on one of power supplies). We can see event in errlog.sys:

Event:       153
Description: VMS Device Error Event  at Tue 25 Oct 2011 16:58:12 GMT+02:00 from XXXX31
. . .
TLV_DSR_String         HP rx2660  (1.59GHz/12.0MB)
TLV_DDR_String         HP MSA VOLUME
. . .
Display_Message_41      409                    STORAGE BOX POWER SUPPLY FAILED

As result of successful test we receive the trap which contains details similar to Problem Report:

Generic: 6; Specific: 164002; Enterprise: .1.3.6.1.4.1.232.164;
Variables:
[2] .1.3.6.1.4.1.232.164.2.2.0 (Integer): 4
[3] .1.3.6.1.4.1.232.164.2.3.0 (OctetString): https://XXXX31.domain.com:7906/webesNotification?id=XXXX31.domain.com@1377777777723
[4] .1.3.6.1.4.1.232.164.2.4.0 (OctetString): MSA1500 or MSA1000 : Device Port/Target/LUN 0/0/0 : LCD message= 409:   Active-Active Firmware
[6] .1.3.6.1.4.1.232.164.2.6.0 (OctetString): 01/12/12 14:20:33 +01
[7] .1.3.6.1.4.1.232.164.2.7.0 (OctetString): XXXX31.domain.com
[8] .1.3.6.1.4.1.232.164.2.8.0 (IpAddress): xxxx31
...
[12] .1.3.6.1.4.1.232.164.2.14.0 (OctetString): Modular Smart Array Controller DISK ENCLOSURE POWER SUPPLY fault. 
  DISK ENCLOSURE POWER SUPPLY FAILED
  One if the power supplies has had a major error,
  or there was a input power failure.

  The repair action is as follows:
  - Ensure there was...
[13] .1.3.6.1.4.1.232.164.2.15.0 (OctetString): Probability:                 High
  Description:                 Input Power Failure
  Description:                 Power/Cooling Assembly
  Spare FRU Part Number:    MSA1000 & MSA500 212398-001

After each test we have event in errlog.sys but not always we receive this trap. We need trap after each event, I am not sure if it is configurable behavior.

Versions:    WEBES V5.1; OpenVMS V8.3-1H1

Please help, thank you in advance,

Vitalis.

[Hoff]

If you have followed the WEBES documentation for establishing SNMP notifications and if you have the current WEBES software and patch level installed and if it's failed, then call HP Support.

[Witas]

HP says : "WEBES V5.1 on Open VMS is no longer supported".

In WEBES documentation I can see commands to enable/disable SNMP notifications, but cannot find anything about configurable behavior, about "customer's notification preferences" (to send notification immidiately or once per some defined period, etc )

"...the analysis engine collects the error data and creates a problem
report containing a description of the problem and any corrective actions
required. Once the problem report is created, it is distributed in
accordance with the customer's notification preferences."

BR, Vitalis.

[Witas]

.. or maybe that I put question wrong way, maybe I need to know when, under which circumstances  Problem Report is being generated

[Hoff]

:If you have followed the WEBES documentation for establishing SNMP notifications...

:

:****and if you have the current WEBES software and patch level installed*****

:

:...and if it's failed, then call HP Support.

If it's not supported, then ask what's current; what HP recommends for addressing this. 

Ask HP what they want you to run for this requirement, in other words. 

Most of the SNMP stuff resided in the System Management Homepage (SMH) stuff, and most of the WEBES stuff (on OpenVMS) moved into a client probe that was then connected into some Windows-based server running WEBES software.   The replacement product (probe) might well be the Event Log Monitor Collector (ELMC) stuff, based on the last time I went digging here.

Another option is implementing something like Nagios and NRPE, and manage this stuff yourself.

[labadie_1]

Or Zabbix, as there is an OpenVMS agent

http://www.openvms.org/stories.php?story=09/09/07/7523546

[Witas]

Thank you for answers!

We have also DCL script which is checking errlog.sys periodically and sends traps. But our script does not provide so nice formatted info as webes notification and we are using "request /to=oper1" command to send a trap, not snmp_trapsnd directly, that's why cannot be free in choosing OID for trap to make it Critical on OpenView.

Thank you.