To: | nv-l@lists.us.ibm.com |
---|---|
Subject: | Re: [nv-l] SAN switch traps |
From: | James Shanks <jshanks@us.ibm.com> |
Date: | Thu, 29 Apr 2004 10:03:24 -0400 |
Delivery-date: | Thu, 29 Apr 2004 15:12:19 +0100 |
Envelope-to: | nv-l-archive@lists.skills-1st.co.uk |
In-reply-to: | <001e01c42ded$29a429a0$410101c1@domain.bisnet.ro> |
Reply-to: | nv-l@lists.us.ibm.com |
Sender: | owner-nv-l@lists.us.ibm.com |
Lucian, You are mistaken about mib2trap. It does work on SNMP V2 MIBs with Notifications statements. I know, I fixed that myself several years ago. Your confusion comes from assuming that because the MIB uses SNMP V2 constructs, that the agent cannot send the trap in SNMP V1 format. Most can and do. Almost all agents which support SNMP V2 also support SNMP V1, but often only an SNMP V2 MIB is provided. So mib2trap reads both. The deal with this Brocade MIB is that they are actually expecting you to run mib2trap on it, to get the formatting correct. I can see that from what you have provided. Brocade has provided all sorts of formatting hints in the comments in the MIB. The variables found in the - -#ARGUMENTS statements map into those %d placeholders in the - -#SUMMARY statements. Actually, since the ARGUMENTS start with zero and the trapd variables start with 1, you have to add 1 to each argument to get the mapping correct. But mib2trap will do that for you. If you run mib2trap, and get an addtrap bat file, when you run the bat, it should overlay the definitions you already added with good ones. James Shanks Level 3 Support for Tivoli NetView for UNIX and Windows Tivoli Software / IBM Software Group
Thanks james mib2trap doesn't worked on it (because I have notification-type for trap definition - so I suppose definitions are trap V2 but the switch is sending trap V1 as far as the server can receive them). So I think I added them manually. Anyway I tried $* and it prints all the variables. I know I should edit the traps and put there $1, $2 and so on but I'm not sure about the order. maybe I should send you one trap definition: swFCPortScn NOTIFICATION-TYPE OBJECTS { swFCPortOpStatus, swFCPortIndex, swFCPortName, swSsn, swGroupName, swGroupType, swGroupMemPos } STATUS current DESCRIPTION "A swFCPortScn(3) is generated whenever an FC_Port changes its operational state. For instance, the FC_Port goes from on-line to offline. The VarBind in the Trap Data Unit shall contain the corresponding instance of the FC_Port's operational status, index, swFCPortName and swSsn. swFCPortName, swSsn are optional." --#TYPE "A Fibre Channel Port changed its operational state." --#SUMMARY "Port Index %d changed state to %d Port Name: %s and SSN is #%s, GroupName %s,GroupType %s,GroupMemPosition %d" --#ARGUMENTS { 1, 0, 2, 3, 4, 5, 6 } --#SEVERITY INFORMATIONAL --#TIMEINDEX 1 --#STATE OPERATIONAL ::= { swTrapsV2 3 } thanks again Lucian ----- Original Message ----- From: James Shanks To: nv-l@lists.us.ibm.com Sent: Thursday, April 29, 2004 4:05 PM Subject: Re: [nv-l] SAN switch traps Well, then it is just a matter of changing your trapd.conf. The first question I might ask is, how did you mess it up? Where did you get the trap definitions which are bad? That might tell you how to fix it. Do you have a MIB which defines the trapd or not? Did you run \usr\ov\bin\mib2trap on it? If not, this is not rocket science in any case. Pick a garbled message and go look at it in trapd.conf. Those % signs are meaningless to trapd; in NetView he only recognizes $ If you change your trap message to be just $*, trapd will print all the variables to the trapd.log and then you can look at them, and decide how you want the message to read the next time and change it. James Shanks Level 3 Support for Tivoli NetView for UNIX and Windows Tivoli Software / IBM Software Group
for J Shanks: sorry for this late response yes, all 3 points below are correctly, and yes that's what I'm trying to do: modify trap definitions with trap settings in order to have correctly formated events thanks! Lucian ----- Original Message ----- From: James Shanks To: nv-l@lists.us.ibm.com Sent: Thursday, April 08, 2004 4:32 PM Subject: Re: [nv-l] SAN switch traps OK, so you are saying that you get the traps but they are garbled; the format is not what you expected. That's what I was trying to clarify. So that is good. It means (1) that the traps are being sent by Brocade, (2) they are in SNMP V1 format so NetView for Windows can receive them, and (3) they are being received, and formatted in some fashion, even if not correctly, by what you have in trapd.conf. So at this point you should be able to edit the definition for these traps in trapd.conf with Trap Settings and change it. Is that what you are having trouble with? Or did you want to review the process by which you imported the trap definitions into trapd.conf in the first place to figure out where it went wrong? James Shanks Level 3 Support for Tivoli NetView for UNIX and Windows Tivoli Software / IBM Software Group
in the mean time I realized that those traps are not correctly because in their definition I have %d and not $d (for example) so they are only some kind of models (?) as far as I understand.... referring to your question: I don' have "no fmt found" traps. I'm sure lots of such traps are sent by brocade switches: my events is somtehing like: trapname: swFCPortScn severity : Warning source: Agent category: Error Enterprise: sw Trap: 3 time:...... Node: 10.16.131.26 - and this is the switch address Description .... what I sent before I'm not sure what you mean with "device or software log ...to find out". I don't think traps are sent in V2 format as far as I can see them in my event console (even if uncorrecly formated) and I see them all in trapd.log. BUT in my mib file these are notification-type (this is snmp V2 name for trap-type ?) and also I think I had some problem to import them with mib2trap. in nv.log not traps are discarded! I have ethereal installed to capture traffic especially on port 162 but I don't see why to start capture at this moment as far as I'm very sure these traps are from Brocade switch ... thanks a lot James.. as allways your response are very good! Lucian ps I hope I made my self clear... :)
hi there list ! I want to send traps from a Brocade SAN switch to a windows nv 713 and I receive allways messages like that: - PORT INDEX %d changed to %d Port Name: %s and SSN is #%s, GroupName %s, GroupType %s, GroupMemPosition %d I'm sure I imported correctly the traps from the switch mib file so... what could be the reason? It really doesn't help now. thanks Lucian Sincerely Yours / al Dvs. Lucian Vanghele, Senior System Engineer, Business Information Systems- BIS srl tf./fax:+4021 255 4579 This message is strictly confidential, between the emitter's and the addressee's companies. It's mandatory to have the approval of the initiator, to send it entirely or partially, to a third company. Acest mesaj este strict confidential , pentru companiile emitentului si receptorului. Pentru a putea fi transmis, intreg sau partial, unei terte companii, este obligatoriu a avea aprobarea initiatorului. |
<Prev in Thread] | Current Thread | [Next in Thread> |
---|---|---|
|
Previous by Date: | Re: [nv-l] SAN switch traps, Lucian Vanghele |
---|---|
Next by Date: | Re: [nv-l] SAN switch traps, Lucian Vanghele |
Previous by Thread: | Re: [nv-l] SAN switch traps, Lucian Vanghele |
Next by Thread: | Re: [nv-l] SAN switch traps, Lucian Vanghele |
Indexes: | [Date] [Thread] [Top] [All Lists] |
Archive operated by Skills 1st Ltd
See also: The NetView Web