• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 288
  • Last Modified:

2003 PDC & 2003 server (peer to peer) network issues

Hello experts,
I have just walked into a new network setup and there are issues I do not know how to address.
There is a DC server running server 2003 sp2 with 9 systems connected to domain.  No issues there.  This DC is handling DHCP and DNS.

There is another server running server 2003 sp2 not in domain, just workgroup.  there are 9 point of sales systems in the same workgroup.   Let's call this server the POS server.
The POS server is also running SQL 2005 Standard edition and doing replication of the POS databases through VPN tunnels to two other stores.
Replication is working fine.

All local systems are in the same subnet.  

POS server has static IP and is using PDC for primary DNS and 8.8.4.4 for secondary DNS.
POS workstations are using workgroup settings and static ip's.  Only IP address and subnet mask setup.  No Gateway or DNS as these systems do not have internet privileges.
Most POS systems have 10/100 nics, a few have GB nics.
The POS server has two GB nics, one is disabled.

Setup has all GB managed switches.  None of the switches show any collisions or dropped packets.
There are no errors in the Windows logs or the SQL logs.

They are having lag issues with the POS systems in the workgroup completing sales.
Invoices are recorded and then disappear, sales do not complete, workstations hang at the end of a sale.
It has been suggested to me that the issue could be that the POS server does not have enough bandwidth to handle SQL doing replication along with 9 workstations hitting SQL for sales and that the communication between the POS program and SQL is spotty which is creating all the strange issues.
One thing that was mentioned was that it would help to increase the bandwidth into the POS server.  I have been reading about binding the adapters but cannot find the proper way (if there is one) to use my secondary adapter to help give me more bandwidth on the POS Server side.  

My experience is with single DC networks and I do not know if this thing is even setup right.

Any help is sincerely appreciated and I hope I have given enough information to get this started.

Thank you in advance,
JG
0
4rg
Asked:
4rg
  • 5
  • 3
2 Solutions
 
NetfloCommented:
Hi,

I would suggest taking a look at the following areas as your root issue:

1. Ensure you have a full system state backup of your SQL box, double check if unsure.
2. Review Event Viewer logs, any clues on lag, memory consumption?
3. Perform a CHKDSK in read only mode, does it complain of any errors - this would cause the machine to appear to pause. CHKDSK /R on all drives.
4. Is your SQL box fully up to date? OS updates, SQL service pack and updates?
5. Are you able to change the replication schedule to calm the server down from working so hard, to eliminate that issue.
6. Have you considered that your disks or subsystem may not be setup correctly. Disks may be too slow, incorrect RAID config, AV may be running real time scanning of transactions?

Over to you for the replies and results.
0
 
4rgAuthor Commented:
In Response.

1. Ensure you have a full system state backup of your SQL box, double check if unsure.
Daily backups are run.  Yes, full system state.

2. Review Event Viewer logs, any clues on lag, memory consumption?
As mentioned in original question, no errors in Event logs (server or workstation) or SQL logs on server.

3. Perform a CHKDSK in read only mode, does it complain of any errors - this would cause the machine to appear to pause. CHKDSK /R on all drives.
Not sure if you mean the local workstation or the server but have done on workstation.  Also ran Memtest for 24 hours on local machine, no problems.

4. Is your SQL box fully up to date? OS updates, SQL service pack and updates?
All updates done as of April, then I turn all off for the season.  Will do again in September.

5. Are you able to change the replication schedule to calm the server down from working so hard, to eliminate that issue.
They have multiple locations selling gift cards and are concerned about an issue if replication not constant.  Can certainly turn off for an hour to test if need be.  That is on my list for the next test but wanted something more if that did not help.  Just recently loaded the application to local drive (rather than running off mapped drive).  Did not help at all.

6. Have you considered that your disks or subsystem may not be setup correctly. Disks may be too slow, incorrect RAID config, AV may be running real time scanning of transactions?
Running RAID 5 on three SATA disks with hot spare.  All looks well in app.  Had first disabled then uninstalled AV completely (temporarily) as test.  No help there.

No thoughts on binding nics together for additional bandwidth?  Can it be done?  I just keep finding things on NLB and CLB and I don't think that is what I want.
BTW, POS software and SQL on same machine.  Dual Xeon processors, 4gb ram. Intel server board.
That is recommended setup by POS software company.

Thanks for your thoughts.
0
 
AkinsdNetwork AdministratorCommented:
To bind the nics on the PC, you need to create an ether channel on the nics. Server 2008 or 2012 refers to it as NIC teaming. Check the adapter on your server and configure as necessary
0
Get quick recovery of individual SharePoint items

Free tool – Veeam Explorer for Microsoft SharePoint, enables fast, easy restores of SharePoint sites, documents, libraries and lists — all with no agents to manage and no additional licenses to buy.

 
4rgAuthor Commented:
Hello Experts,
I teamed the nics and setup the ports on the managed switch with LAG/LCAP.  
Everything seemed fine, the clients could connect and run the app.  Then I tested the credit card portion and it did not work.  I also noticed that LogMeIn that I use to remote connect in to the server said it was enabled but not online.  
So, two applications on the server that need to connect to the internet to work, did not.  
I could browse the internet just fine out of IE and Chrome.  
I had to change everything back so they could work today.

Does anyone have an idea why these two apps would not work?
Do you think I need to open a new question to address this issue?
Thanks again for the help.
0
 
NetfloCommented:
Port 443 being blocked outbound?

Did you manage to address some of the questions on my first post?
0
 
4rgAuthor Commented:
Netflo,
Would port 443 be blocked outbound just because I teamed nics?  Is not a problem before or after teaming?

Yes, I did answer all of your questions in my first response.  I copied your message and responded to each one.  See above.
Thanks for your time.
0
 
4rgAuthor Commented:
That was a weird post...  What I meant was it was not a problem using those two apps before I teamed the nics or after I (un)teamed the nics.
0
 
NetfloCommented:
Not necessarily, teaming shouldn't affect specific type of traffic.

In reply to your answers to my questions:

3. CHDSK /R needs to be done on the server. Bear in mind that this can take some time to run so you may want to plan for this to happen out of hours.
0
 
4rgAuthor Commented:
Have done additional testing on the database and application.  Am going down a different path for now.  thanks for your responses.
0

Featured Post

Vote for the Most Valuable Expert

It’s time to recognize experts that go above and beyond with helpful solutions and engagement on site. Choose from the top experts in the Hall of Fame or on the right rail of your favorite topic page. Look for the blue “Nominate” button on their profile to vote.

  • 5
  • 3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now