configurable OID for each device or device type which returns an interface name table. This way I can use any variety of router/switch and by digging out the appropriate MIB table have an interface description which is more useful to me than just the interface type or name.
>Monitoring a Catalyst 3000 InterMapper shows an incredible error rate >on the uplink ports. There is nothing wrong with the link-=20 Depending on the switch (Cisco's switch line is almost totally erratic) the values placed into the MIB are garbage. The 3200 switches are effectively overpriced garbage; I'll elaborate privately for those who want to know how badly broken they are. >I have ~100 of these devices - they all show the same symptom. I've >checked "Don't report interface errors" to get around it, but I've lost >management capability in the process. >Monitoring a Catalyst 2916M-XL, InterMapper won't give me any traffic >info on the interfaces. I know there is traffic to the downstream >devices yet I get zero's for the interface information. You have old firmware on the switch. This is a known bug in the IOS you are using; the MIB values only get updated when you do a 'show interface'. Upgrade the switch to 11.2(8)SA4 and this problem will go away. >Is this a problem with the MIB structure that Cisco chose to implement? >All other Cisco devices, 7500's, 5000's, 5500's, 5300's, 5200's, etc, >work fine. -- David Milton University of Manitoba, Academic Computing & Networking email: [EMAIL PROTECTED] voice: (204) 474-6986 fax: 474-7515 http://home.cc.umanitoba.ca/~milton/ Public key available on request. ---------------------------------------------------------------------- Subject: Re: a few sugestions: disable "saving", password acces... From: David Milton <[EMAIL PROTECTED]> Date: Tue, 20 Apr 1999 12:03:48 -0500 >Subject: a few sugestions: disable "saving", password acces, DNS > (Datmouth network suite" >From: [EMAIL PROTECTED] >Date: Mon, 19 Apr 1999 16:02:21 +0200 > > >- I would like a preference to disable the warning (save or cancel) > when something is changed. In our situation I use a dedicated Mac > to survey witch round midnight perform shutdown and restart > automatically and the warning causes hang-ups. And the Mac is > sometimes used as diagnostic-tool when you for instance hide a > chart you must save so a hang-up risk. Alternatively an auto save or don't auto save on quit would work. As long as InterMapper always quits without user intervention when it receives a quit message it probably doesn't matter. > >- Splitting the syslog in devicelog and http-log. > >- A password when HTTP-acces > >- A password / IP-mask for each map (you don't want that your users see > everything) I think the web server function itself should be 'pulled' from InterMapper and built as a seperate product to let me use QuidProQuo, Web* or apache under OS X if I want to. All of the security is available in full blown web servers; why should the InterMapper development team re-invent it? I'd rather see Apple Events or another method of obtaining the requested web page on demand. These pages can then be served to the user by a seperate server process. This does turn the web server into a proxy which sends a query to InterMapper but allows for all of the above using 'standard' web style authentication or restrictions. It also seperates the http access log entries since the web server is a seperate process. Finally, it would allow a CGI in the middle to permit me to customize some of the information returned in the InterMapper generated web page; perhaps having some general information about the Map inserted into the page or links to other local resources. -- David Milton University of Manitoba, Academic Computing & Networking email: [EMAIL PROTECTED] voice: (204) 474-6986 fax: 474-7515 http://home.cc.umanitoba.ca/~milton/ Public key available on request. ---------------------------------------------------------------------- Subject: Re: a few sugestions: disable "saving", password acces... From: Chris Pepper <[EMAIL PROTECTED]> Date: Tue, 20 Apr 1999 13:38:28 -0400 At 12:03 PM -0500 04/20/99, David Milton wrote: >I think the web server function itself should be 'pulled' from InterMapper >and built as a seperate product to let me use QuidProQuo, Web* or apache >under OS X if I want to. All of the security is available in full blown web >servers; why should the InterMapper development team re-invent it? I'd >rather see Apple Events or another method of obtaining the requested web >page on demand. These pages can then be served to the user by a seperate >server process. This does turn the web server into a proxy which sends a >query to InterMapper but allows for all of the above using 'standard' web >style authentication or restrictions. It also seperates the http access log >entries since the web server is a seperate process. Finally, it would allow >a CGI in the middle to permit me to customize some of the information >returned in the InterMapper generated web page; perhaps having some general >information about the Map inserted into the page or links to other local >resources. I disagree. I don't want to run IM on a production webserver, and don't want to run a full webserver on my IM machine. My impression is that web serving component is relatively simple, compared to all the querying and map generation. I do understand that additional flexibility would be helpful in some cases, but I'd hate to see the current (excellent) webserver functionality go away. Splitting HTTP (and telnet?) server logging into separate files seems like a good idea, though.... Chris Pepper -- Chris Pepper | National Audubon Society: Web & List Manager 212 979 3092 | <http://www.audubon.org/staff/pepper/> ---------------------------------------------------------------------- End of InterMapper-Talk Digest ____________________________________________________________________ Note: To unsubscribe from this mailing list, please send email to: <mailto:[EMAIL PROTECTED]> Thanks!
