Any suggestions on troubleshooting dropped mapped network drives?

[osmhquser]

We are having some issues between our (2) Windows 2003 Storage servers and Windows 7 client machines. The issue that is occuring is that the users indicate they temporarily loose network connection to their mapped drives on these servers. This may happen several times a day. The regional manager wants to say it's a network issue but I believe it's something b/t the server and workstation.

I like to know if there are some additional parameters, debugging or configuration that might be recommended to perform on our (4) Cisco 3750G-48port POE switches (stacked) that would help determine that it's not a network issue. I do not see any errors or collisions on the ports that the servers are connected nor am I seeing much of anything wrong with the user ports. There are only 3 vlans at this office, vlan 30=users, vlan 20=servers, vlan 40=mgmt.

Thx in advance for any help given.

[Peter Paluch]

Hello,

During the time of the lost connection to network drives, are the affected stations capable of any other IP communication, i.e. pinging their default gateway, pinging the Windows server, etc.?

What STP version are you running? If you are running RSTP, are the ports connecting end devices configured as edge ports? Failure to do so could cause similar issues.

Best regards,

Peter

[osmhquser]

Users are able to still ping default gateway and other IPs. Not sure if they loose connectivity to server yet, I've been doing a constant ping to the server from a workstation to see if any drops occurs.

Below is the spanning-tree info. If there is additional detail needed, please let me know what the command is.
spanning-tree mode pvst
spanning-tree portfast bpduguard default
spanning-tree portfast bpdufilter default
spanning-tree extend system-id
spanning-tree uplinkfast

This is what our standard workstation switch port is like:

switchport access vlan 30
switchport mode access
switchport nonegotiate
snmp trap mac-notification added
snmp trap mac-notification removed
no cdp enable
spanning-tree portfast

[Peter Paluch]

Hello,

Thank you for your response. The configuration seems to be correct - you are running PVST.

Can you describe in more detail how the outage looks like? Is it possible that the Windows clients close the TCP connection to the server after a longer period of inactivity, and the reconnection needs some time? We may very well be hunting an application issue here which is quite hard to debug.

Best regards,

Peter

[matyas_prokop]

Hi,

we've been having some issues with Windows 7 more particulary with network card drivers so I would definately check this. Also we've had similar issues because of dot1x bug in some IOS versions but I guess from your configuration this is not your case.

Sent from Cisco Technical Support iPad App

[shocke]

We have experienced this before in our network. Check out this Microsoft blog http://social.technet.microsoft.com/Forums/en-US/w7itpronetworking/thread/4224b7f3-131f-4489-8397-5330bdb4369a/

Setting the autodisconnect really helped us out. For us it was a winodws client side issue.

Eli

Sent from Cisco Technical Support iPhone App

[matyas_prokop]

Eli's link reminds me that we've had issues not only with drivers but more particulary with network card power saving. Your issues seems to me more on the side of client than Cisco.

[osmhquser]

I want to thank everyone for their feedback as this has been helpful. I will look into the additional areas that everyone has mentioned. Thx and happy holidays.