[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference azur::mcc

Title:	DECmcc user notes file. Does not replace IPMT.
Notice:	Use IPMT for problems. Newsletter location in note 6187
Moderator:	TAEC::BEROUD

Created:	Mon Aug 21 1989
Last Modified:	Wed Jun 04 1997
Last Successful Update:	Fri Jun 06 1997
Number of topics:	6497
Total number of notes:	27359

6321.0. "Another SNMP alarm question...." by SNOTTY::BARRY (Ploppy Sir, Son of Ploppy) Tue Jun 06 1995 17:50

Hi All,

I know this has probably already been discussed, but I have not been able
to weed out a definative step by step list, from the numerous replies, of
how to fix the problem.

I have a customer machine running :

VMS V5.5-2
UCX V2.0A
MCC V1.3.0

We are using SNMP to monitor devices in London and four remote sites (Tokyo,
New York, Hong Kong and Paris). When I interrogate devices from within
command line MCC, only the devices in the London domain have an ipReachAbility
flag of "up"......all the others have a flag of "down". Pinging these devices
tells me that they are all up but due to MCC's perception.....all my alarms
that look at the ipReachability flag (for a <> "up" state) keep firing.

I know it has been suggested that the Ip Poller be used but due to customer
issues this cannot be the case. 

Any help please (the customer is getting pretty annoyed with seeing these
mis-leading alarms).............

Would it help to upgrade to V1.4 ??

Regards

Barry

T.R	Title	User	Personal Name	Date	Lines
6321.1		AZUR::DURIF		`Mon Jun 12 1995 11:52`	18
	Hi Barry, First it is possible for you to check if it is a timeout problem ? Do you have a mean to test the response time of the ping command ? May be MCC does not wait enough time to get the response. Modification of "MCC 0 TCPIP_AM UDP Timeout" attribute may be of no use as MCC V1.3.0 is suspect to keep always the default value. If you can identify that the response time is the cause of the problem it can help installing V1.4 because UDP timeout is handled properly. Do not go in the process of upgrading before this point is checked. Regards, Benoit
6321.2	Help........PLEASE !!	SNOTTY::BARRY	Ploppy Sir, Son of Ploppy	`Fri Jun 23 1995 12:14`	5
	Can I PLEASE have some feedback on this.....even if its a "dont know".... This is a customer problem and I am getting a lot of pressure !! Barry
6321.3	Try ping	WELLIN::MCCALLUM		`Mon Jun 26 1995 11:59`	12
	Hi Barry, What does UCX think of the devices ? try ucx ping and see if it can see them, if it cant then you have an ip problem. If a UCX ping works then mcc ( show snmp fred ip_reach) should work, if not look at the timeout and retry counter on the snmp am. Dave
6321.4		AZUR::DURIF		`Mon Jun 26 1995 12:15`	12
	Hi Barry, .2 Can I PLEASE have some feedback on this.....even if its a "dont know".... .1 First it is possible for you to check if it is a timeout problem ? .1 Do you have a mean to test the response time of the ping command ? Did you make a try ? Did you trace the retries made by MCC. I need some feedback even if it is a "dont know" too. You may compare the MCC timeout versus the ping timeout. Benoit