Odd netowrking issues on 2 Dell PowerEdge servers

Question

Hi, we have recently installed two Dell PowerEdge servers both set-up with Windows Server 2008 R2. I used the Dell Systems Management disk to install the OS which pre-loads certain drivers, etc.
Since the install I have been trying to track down and fix some very strange behaviour with both servers, in particular with MS Exchange 2007 and IIS.
The issues on server 1 (file server) are no access to Windows Updates (0x80072EE2), certain users on other VPN sites having problems accessing certain areas of our Intranet that require authentication, odd problems using IE on the server - for example
I can go to Google, carry out a search but when I click on a result it just sits there waiting for Google. There are no issues with users on the same site, and there are no issues accessing the server shares from any site.
The issues on server 2 (Exchange server) are even more odd and have me rapidly pulling out my hair. With the site we can send and receive between this server and an old Exchange 2003 server with no problems, Public Folder replication from the old server
to the new is occurring with no problems and we can send messages with large attachments both to external users and users on other sites. However when we start looking at incoming connections this is where it gets odd. Incoming external messages over 4Mb are
delayed and eventually bounced - our provider says that they are getting timeouts. Incoming messages with attachments from two other internal Exchange 2007 sites are getting delayed and from one sites are not getting through at all. Messages without attachments
are not getting through at all. In the Exchange logs I am seeing Connection Timeout and Connection Dropped messages.
I am very sure that the problem is not with message size as all parameter relating to message size are set to 100Mb. I know it is not a firewall issue as redirecting incoming external messages to the old Exchange server is fine.
To my mind this looks like there is something on these servers which is messing with http/https/smtp when communicating outside the local subnet. I have checked for viruses and mal-ware and for testing purposes disabled AV and firewall but this makes no
difference.
I have two other PowerEdge server on another site with similar configurations which have been working fine for 18 months. The only big difference I can see between the servers is that these two are the only one where I used the Dell Systems Management disk
to install the OS. I have Googled 80072EE2 and it does appear that other have seen the same problem and their way to resolve it was to wipe and reinstall directly from the Microsoft disk and no use the Dell Systems Management Disk.
However, this is not an option here as these servers are both in a live environment (I know, should have made sure everything was running properly before going live, but these had to be put in place rapidly to replace failing servers).
From what I can see of the servers it does appear that somewhere along the line Dell have used files which are causing some networking issues on these two servers - can anyone advise me if there is a reasonably straight-forward way to remove the network
components form the servers and then reinstate from the Microsoft disk?
I have tried the fixed relating to Windows Update from the suggestion offered searching for 80072EE2.
I have tried installing the latest Dell drivers for the network adaptor, and also the latest Broadcom ones.
I have tried switching off various Offload settings on the NIC as suggested by other although that was more relating to Hyper-V issues.
Anything else I should be looking at, or can anyone tell me what else to look for on the servers?
Many thanks, Eddie

Answer 1

Hi Eddie,
Based on your description, we have many different problems in two servers as below,
In the file server.
-> The issues on server 1 (file server) are no access to Windows Updates (0x80072EE2).
-> odd problems using IE on the server - for example I can go to Google, carry out a search but when I click on a result it just sits there waiting for Google.
I noticed that you mentioned that you have tried the fixed relating to Windows Update. Have we solved this issue? And another issue, IE could access Google but can’t access
the links in search result. It seems that this issues are related to name resolution. To verify if the name resolution works fine. Please ping the DNS server at first. If we can ping DNS server, we can also ping
www.google.com and other domain names to verify the function of name resolution. Please try to use ipconfig /flushdns and ipconfig /registerdns command in server 1. And the DNS port(53) was opened. After this, if the problem
still exist, please involve the administrator of DNS server to help verify if the domain name resolution has some problems.
-> certain users on other VPN sites having problems accessing certain areas of our Intranet that require authentication
I’m a little unclear about this part. If the VPN connection has set up successful, it is possible that the user need be authenticate when the user access some resource of the
Intranet. So what is the certain areas and what prompt we received.
In the Exchange server.
->Incoming external messages over 4Mb are delayed and eventually bounced - our provider says that they are getting timeouts. Incoming messages with
attachments from two other internal Exchange 2007 sites are getting delayed and from one sites are not getting through at all. Messages without attachments are not getting through at all. In the Exchange logs I am seeing Connection Timeout and Connection Dropped
messages.
-> I am very sure that the problem is not with message size as all parameter relating to message size are set to 100Mb.
We can send messages but we can’t receive messages, it was delayed or it can’t get through. And I notice that you mentioned that the message size is
not the problem. To verify if the network have problems about the size messages. We can also download a file whose size is over 5Mb from the Internet. If we can download the file, maybe it’s not the network problems.
-> I know it is not a firewall issue as redirecting incoming external messages to the old Exchange server is fine.
Maybe we could also turn off the firewall temporarily to see if the firewall block some connection.
-> To my mind this looks like there is something on these servers which is messing with http/https/smtp when communicating outside the local subnet.
If we think the http/https/smtp are related to this issues, maybe we could check if all ports about this service was opened.
And due to server 2 is an Exchange server, there might be some wrong with the Exchange services. We have the related Exchange forum with professional
to deal with this issue. If there are still some problems with Exchange server, maybe we can try to create a new thread about Exchange.
Best Regards,
Tina