|
Why, yes, there is a reason. Another thing with discovery, is there any reason why nv groups two remote site servers and itself in the same IP group? It's like my nv is on its own management LAN and the two windows servers are on remote sites and nv still puts them into the same subnet even after i delete them. What is the network address of the subnet (management LAN, as you say) that they are on? Is it correct? Netview draws subnets based on IP addresses and subnet masks. It gets the mask from the device if it supports SNMP, and it gets the mask from a subsequent special ICMP request if it is not SNMP supported. If it cannot get it either way, it makes some assumptions based on the address. And very often it will put non-snmp things in the subnet with the SNMP-capable device whose arp-cache it found those devices in. The stuff on your map is supposed to be connected. At the top level of the map, networks are connected by routers or route-switches. If you don't have SNMP access to those routers, then they appear in lower level submaps and are not present in the top level to do the connecting. You want to try to clear these up as much as possible. If you have access to the router that connects you to those two remote PCs, then netview will discover them in the arp cache of that router, and you have a better chance of getting them in the right subnet. If, on the other hand, you do have SNMP access to those PCs, check the their object properties to see what masks are on their addresses. Compare that to the actual addresses and masks returned by the MIB application for Addresses. I forget where that is on the Windows version. Maybe Netview is just telling you that they are poorly configured. Or maybe they have multiple addresses, and one of them really is on the same subnet as the Netview. Things do show up in all subnets on which they have addresses. I've seen cases where an SNMP-enabled device has a bad mask, and is in a disconnected subnet, and a bunch of other non-SNMP stuff from its arp cache is added to the same subnet. I delete the bad-masked device that led them astray, so to speak, and rediscover the non-snmp stuff, and it comes up on correct subnet. Cordially, Leslie A. Clark IT Services Specialist, Network Mgmt Information Technology Services Americas IBM Global Services (248) 552-4968 Voicemail, Fax, Pager
Hi leslie, Thanks for the advice! I would check up on the server! Think this could be the issue, as the old machines are running SNMP service where as the new ones aren't allowed for security reasons. I think what I'm asking is stupid, but is there a best practice on discovering/adding devices? What i was told is to ping or poll the IP of the device and "hope" nv adds the object. But often it doesnt work, I've even did a search beforehand to ensure there's no objects with the same IP. Another thing with discovery, is there any reason why nv groups two remote site servers and itself in the same IP group? It's like my nv is on its own management LAN and the two windows servers are on remote sites and nv still puts them into the same subnet even after i delete them. best regards, sixx From: nv-l-bounces@lists.ca.ibm.com [mailto:nv-l-bounces@lists.ca.ibm.com] On Behalf Of Leslie Clark Sent: Wednesday, January 10, 2007 2:44 AM To: Tivoli NetView Discussions Subject: RE: [NV-L] Objects turning red after awhile even when up? There are other ways for nodes to be polled via SNMP besides the $ in the seedfile. 1) The SNMP sysObjectID of the devices is in the /usr/OV/conf/oid_to_type file with a flag of P, meaning poll all devices of this type via SNMP. This is useful, for instance, for certain kinds of servers that don't let you ping all of their internal addresses. For this to be your problem, the devices would have to have been SNMP-supported before, but no longer are. 2) If netmon discovers an unnumbered serial interface on the device, and also finds the other end of the connection in the map, then the nodes are automatically converted to polling via SNMP, since the status of the unnumbered interfaces it creates cannot be checked with ping. 3) If the interfaces are included in the seedfile with the + sign, indicating that they are ISDN connections and down should be good and up should be bad, the devices are also changed to be polled via SNMP. Note that when you are swapping out devices or servers, and reusing the same addresses, it is recommended, as a Best Practice, to delete that old object from Netview and rediscover it from scratch. This prevents erroneous assumptions. Cordially, Leslie A. Clark IT Services Specialist, Network Mgmt Information Technology Services Americas IBM Global Services (248) 552-4968 Voicemail, Fax, Pager
Hi guys, Thanks for the replies. This would sound strange, but my PC/DHCP segments are configured to "@" in my netmon.seed, so i guess they should be discovered and polled by using ping. But how come only certain PCs/DHCP IPs would be polled by snmp? I've checked my ovsnmp.conf and there's no PC/DHCP address listed in here. Is there another location whereby nv store's an object's polling method? best regards, sixx From: nv-l-bounces@lists.ca.ibm.com [mailto:nv-l-bounces@lists.ca.ibm.com] On Behalf Of James Shanks Sent: Monday, January 08, 2007 10:22 PM To: Tivoli NetView Discussions Subject: RE: [NV-L] Objects turning red after awhile even when up? Francois is correct. DHCP entries are indicated by "@",
not by "$". You are forcing netmon to use SNMP polling with that.
Here's an excerpt from the Window netmon.seed file
I am using an Unix version of Netview so this "DHCP" feature is not available for me. On my netview, the "$" means that I force the status polling of an (or range of) object to snmp. Removing the "$" turns back the status polling to ping. (Unless specific cases like unnumbered interfaces...). You should double-check the documentation for your seed file as I am surprised that $ has a different meaning on windows. I am aware of this DHCP monitoring feature on windows, but don't remember how is it used. Salutations, / Regards, Francois Le Hir Network Projects & Consulting Services IBM Global Technology Services Phone: (514) 964 2145  <sixxl@cse.com.sg>
Hi guys, Thanks for the prompt reply! The pc segment is already configured using "$" to signify DHCP clients. Strange thing is this status issue only occurs to new PCs added to the segment as they are replacing the old existing NT PCs to XP machines. I believe the new PCs took the old IP addresses as most of them is on DHCP and some on static. btw how do i change the object's polling from SNMP to ping? Edit the object in ovsnmp.conf file? Or should i delete the object and let nv rediscover it again? best regards, sixx From: nv-l-bounces@lists.ca.ibm.com [mailto:nv-l-bounces@lists.ca.ibm.com] On Behalf Of Leslie Clark Sent: Friday, January 05, 2007 10:20 PM To: Tivoli NetView Discussions Subject: Re: [NV-L] Objects turning red after awhile even when up? Good answer, Francois! Cordially, Leslie A. Clark IT Services Specialist, Network Mgmt Information Technology Services Americas IBM Global Services (248) 552-4968 Voicemail, Fax, Pager
You issue is probably because netmon is trying to monitor the status of this object using snmp instead of ping. If it doesn't answer to snmp, it will still show down, even if it respond to ping. The way this is configured is usually with the "$" in the netmon seed file. You can test if this is the problem with the quicktest command: it will do a status polling using the same method as netmon. Salutations, / Regards, Francois Le Hir Network Projects & Consulting Services IBM Global Technology Services  <sixxl@cse.com.sg>
Hi there, I'm a new netview administrator and facing an issue. We've got a netview is a default installation with a configured netmon.seed, netview discovers all objects and all device status are reflected properly when down. On the user PC objects sometimes it would turn red signifying down status. Verfication of the physical PC is up and netview doing test ping would turn the object to green. After leaving the object for awhile it would turn red again, even though the PC is up. Please advice on what steps should I take to resolve this issue. The netview is 7.1.4 FP4 running on a windows 2003 server and console is using remote desktop to monitor the netview. Best regards, Sixx _______________________________________________ NV-L mailing list NV-L@lists.ca.ibm.com Unsubscribe:NV-L-leave@lists.ca.ibm.com http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only)  _______________________________________________NV-L mailing list NV-L@lists.ca.ibm.com Unsubscribe:NV-L-leave@lists.ca.ibm.com http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only) _______________________________________________ NV-L mailing list NV-L@lists.ca.ibm.com Unsubscribe:NV-L-leave@lists.ca.ibm.com http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only)  _______________________________________________NV-L mailing list NV-L@lists.ca.ibm.com Unsubscribe:NV-L-leave@lists.ca.ibm.com http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only) _______________________________________________ NV-L mailing list NV-L@lists.ca.ibm.com Unsubscribe:NV-L-leave@lists.ca.ibm.com http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only) _______________________________________________ NV-L mailing list NV-L@lists.ca.ibm.com Unsubscribe:NV-L-leave@lists.ca.ibm.com http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only) _______________________________________________ NV-L mailing list NV-L@lists.ca.ibm.com Unsubscribe:NV-L-leave@lists.ca.ibm.com http://lists.ca.ibm.com/mailman/listinfo/nv-l (Browser access limited to internal IBM'ers only) |
Francois
Le Hir <flehir@ca.ibm.com>
