Hi,
I think this is a cisco problem, as I've been tackling it from the MS/NT side and have had no luck. I have 5 2924XL switches in my network and there is one server always gives me problems. It started around the time when we installed the switches and that is why I think they might be related. The server is my PDC running NT4sp6a, as well as being a file server running MS Visual Source Safe 5.0. It's serving out some printer shares as well.
What happens, besides one of my engineers comming into my office complaining about no access to VSS, is that the server seems to lose all basic network connectivity. It doesn't show up in Network Neighborhood, none of it's shares/printers are accessable, no one can ping it and it can't ping any other node. The physical link lights are on, both on the switch that it is hooked up to and the NIC. The system event log will show this error:
EventID 8003
Source: Rdr
The master browser has received a server announcement from the computer BOOLIAN that believes that it is the master browser for the domain on transport NetBT_El90x1. The master browser is stopping or an election is being forced.
This happens every 12mins, once the conectivity is lost. Which I believe is the NetBIOS blast trying to update browser master list. Thru disscussions with coleagues and on IRC I am under the impression that this is but a symptom of the connectivity being lost. What makes it even more frustrating is the this problem seems to be *totally* random. I can go for days with out so mush as a hiccup, and then one day it can go out on me like 3-4 times that day. The only way I think I've been able to purposely cause it to happen is to transfer huge amounts of data to and from it (1-2GB at least). This is a production server, so I don't have much time to do that.
I've found 3 MS Knowledge base articles that come close to describing my problem, but none of them seemed to help.
This one brought up the duplex issue that I was unaware of in general, but the card is a brand new (I replaced it first thing along with the CAT5 cable) 3com 905C 10/100 card. None the less I've set all the different possible combonations on the card and likewise on the port it's in. (it's currently running at 10/half into a NETGEAR switch that is cascaded to a different port on a different Catalyst Switch!) Crazy ain't it!! ;-)
This white paper suggested a fix to d/l from MS, in regards to a problem with NT not properly closing NetBT sessions. I applied that to no avail either. :-(
THIS is the one that directly talks about interaction with Cisco hardware. The only thing is that I'm only running on one subnet, and no VLAN's are setup. It also references Cisco's "IP Helper" which I am not using.
As you can see when it comes NT I'm in my element, but the lower level network/cisco hardware is a little bit of murky ground for me. Any help from ANY angle would be much appreciated, as you can tell I'm very frustrated with this on going problem.
I think this is a cisco problem, as I've been tackling it from the MS/NT side and have had no luck. I have 5 2924XL switches in my network and there is one server always gives me problems. It started around the time when we installed the switches and that is why I think they might be related. The server is my PDC running NT4sp6a, as well as being a file server running MS Visual Source Safe 5.0. It's serving out some printer shares as well.
What happens, besides one of my engineers comming into my office complaining about no access to VSS, is that the server seems to lose all basic network connectivity. It doesn't show up in Network Neighborhood, none of it's shares/printers are accessable, no one can ping it and it can't ping any other node. The physical link lights are on, both on the switch that it is hooked up to and the NIC. The system event log will show this error:
EventID 8003
Source: Rdr
The master browser has received a server announcement from the computer BOOLIAN that believes that it is the master browser for the domain on transport NetBT_El90x1. The master browser is stopping or an election is being forced.
This happens every 12mins, once the conectivity is lost. Which I believe is the NetBIOS blast trying to update browser master list. Thru disscussions with coleagues and on IRC I am under the impression that this is but a symptom of the connectivity being lost. What makes it even more frustrating is the this problem seems to be *totally* random. I can go for days with out so mush as a hiccup, and then one day it can go out on me like 3-4 times that day. The only way I think I've been able to purposely cause it to happen is to transfer huge amounts of data to and from it (1-2GB at least). This is a production server, so I don't have much time to do that.
I've found 3 MS Knowledge base articles that come close to describing my problem, but none of them seemed to help.
This one brought up the duplex issue that I was unaware of in general, but the card is a brand new (I replaced it first thing along with the CAT5 cable) 3com 905C 10/100 card. None the less I've set all the different possible combonations on the card and likewise on the port it's in. (it's currently running at 10/half into a NETGEAR switch that is cascaded to a different port on a different Catalyst Switch!) Crazy ain't it!! ;-)
This white paper suggested a fix to d/l from MS, in regards to a problem with NT not properly closing NetBT sessions. I applied that to no avail either. :-(
THIS is the one that directly talks about interaction with Cisco hardware. The only thing is that I'm only running on one subnet, and no VLAN's are setup. It also references Cisco's "IP Helper" which I am not using.
As you can see when it comes NT I'm in my element, but the lower level network/cisco hardware is a little bit of murky ground for me. Any help from ANY angle would be much appreciated, as you can tell I'm very frustrated with this on going problem.
