Recommended Posts

I have a small network domain at work running Windows Server 2008 and 10 and 8 on the workstations. We had a windstorm the other day and lost power 3 times in 5 minutes. Everything is back up now and working except for 2 of the workstations. They appear to log in to the domain but give an error when trying to reconnect to shared drives. They also can't locate my FileMaker database. Additionally, when I try try to navigate to the server in Explorer I get an error.

 

What could my issue be and how could a power outage cause this and only with 2 workstations?

 

@BudMan to the rescue?

Link to comment
https://www.neowin.net/forum/topic/1324874-network-issue-after-power-outage/
Share on other sites

5 minutes ago, patseguin said:

I have a small network domain at work running Windows Server 2008 and 10 and 8 on the workstations. We had a windstorm the other day and lost power 3 times in 5 minutes. Everything is back up now and working except for 2 of the workstations. They appear to log in to the domain but give an error when trying to reconnect to shared drives. They also can't locate my FileMaker database. Additionally, when I try try to navigate to the server in Explorer I get an error.

 

What could my issue be and how could a power outage cause this and only with 2 workstations?

 

@BudMan to the rescue?

I will let @BudMan address the network side. 

 

I will put money that they are not actually logging into the Domain.  I will say they are using cache domain settings.  This could be why they would appear to login even thought they do not have network access.   (Have you tried any other network access other than the Database/Drives?

 

 

You need to track down the point where the connection is failing. Here are some first steps/questions:

1) How are the workstations networked? Wired or wirelessly?

2) If wired, are you getting send/receive lights on the network adapter on both workstation and switch?

3) Have you checked the power/connectivity of the network switch or AP that the devices are connected to?

4) Are you able to ping the server from the workstation?

5) Can you ping the workstation from the server?

6) If wired, are you able to test the cable from a nearby workstation that works in one of the one's that isn't working?

7) Do you have a spare wired/wireless adapter that you can use in the workstation to test the on-board/current NIC?

 

Hopefully they will be able to tell us a little more of where the issue may be coming from. You may also find some system logs which may help identify the issue.

18 minutes ago, Clirion said:

I will put money that they are not actually logging into the Domain.  I will say they are using cache domain settings.

That was my first thought as well.

 

I assume that you've checked they can connect to the Internet and so on, right? Have you done a ping test from the machines? What about the server, if you ping the workstation what does it come back with? Do you have a test workstation that you can try adding to the domain? I want to stress "adding" rather than using a machine that has already been on the network.

Dns. Amateurs who setup a domain can't get basic principals of dns functionality. 

 

internet domain name servers do not know anything about your internal domain.  

 

Rule 1: if you have issues with your shares, change all dns servers on your dhcp, or if you have statically assigned addresses on your computers, change dns to only point to your internal dns servers.  The dns server will redirect to internet dns servers when needed. 

 

Rule 2: if you are having issues with connecting or authenticating you did not follow rule 1, see rule 1. 

 

Where did he mention anything about his DNS settings?

 

If they were working prior and the other workstations are still working, DNS is probably setup correctly. Try removing them an re-adding them to the domain? Make sure you have the local admin un/password first.

Your point? Ad domain + randomly can't auth to a network share = misconfigured dns 99.9999% of the time.

 

Been around AD for a very long time...have seen this more times than you would believe. It always comes down to this specific issue. Whether it be on the D.C.'s themselves or the clients or both..it always goes back to that.

 

54 minutes ago, sc302 said:

Your point? Ad domain + randomly can't auth to a network share = misconfigured dns 99.9999% of the time.

 

Been around AD for a very long time...have seen this more times than you would believe. It always comes down to this specific issue. Whether it be on the D.C.'s themselves or the clients or both..it always goes back to that.

 

So a lightning strike changed his LAN settings from DHCP to static with a bad DNS.... doubt it.

 

Something is corrupt here, probably the SID token to the domain controller.

55 minutes ago, sc302 said:

Your point? Ad domain + randomly can't auth to a network share = misconfigured dns 99.9999% of the time.

 

Been around AD for a very long time...have seen this more times than you would believe. It always comes down to this specific issue. Whether it be on the D.C.'s themselves or the clients or both..it always goes back to that.

 

No misconfigured from the beginning.  It doesn't happen all the time if your secondary dns server is an internet dns server.  It can happen any time. Lightening is a coincidence, if it were really lightening related nothing would work or the network card would be fried. 

2 hours ago, xendrome said:

So a lightning strike changed his LAN settings from DHCP to static with a bad DNS.... doubt it.

 

Something is corrupt here, probably the SID token to the domain controller.

Yeah that was kind of my thinking. Don't see how a lightning strike could change network settings. I'm going in tomorrow and will try some suggestions here, starting with  pinging.

13 minutes ago, patseguin said:

Yeah that was kind of my thinking. Don't see how a lightning strike could change network settings. I'm going in tomorrow and will try some suggestions here, starting with  pinging.

You can run NLTEST from the client to test the domain connectivity - https://social.technet.microsoft.com/wiki/contents/articles/16067.nltest-to-test-the-trust-relationship-between-a-workstation-and-domain.aspx

3 hours ago, xendrome said:

They lost power multi-times in a short period of time, the workstations likely didn't have APCs and died...

Irrelevant.  Lost power causes complete failure, Either at the component level or system level. If they can communicate, either via ping or by browsing the internet, the components aren't damaged. 

6 minutes ago, sc302 said:

Irrelevant.  Lost power causes complete failure, Either at the component level or system level. If they can communicate, either via ping or by browsing the internet, the components aren't damaged. 

Ok then. Somehow you are going on about hardware components being damaged, and I am talking about a corrupt SID to the domain controller due to some form of corruption from a hard power off of workstations.

2 hours ago, xendrome said:

I am talking about a corrupt SID to the domain controller

Then he wouldn't log in, etc.  He couldn't auth - would be a clear error.

 

Im with sc302 here - most likely something stupid and for whatever reason its manifesting itself now.  Just that they had power outages that they are pointing to that being the cause...

 

Huge missing piece of the puzzle is what exactly is the error??

 

"but give an error when trying to reconnect to shared drives"

 

This is like telling the mechanic - yeah car won't start..

On ‎3‎/‎11‎/‎2017 at 9:25 PM, BudMan said:

Then he wouldn't log in, etc.  He couldn't auth - would be a clear error.

 

Im with sc302 here - most likely something stupid and for whatever reason its manifesting itself now.  Just that they had power outages that they are pointing to that being the cause...

 

Huge missing piece of the puzzle is what exactly is the error??

 

"but give an error when trying to reconnect to shared drives"

 

This is like telling the mechanic - yeah car won't start..

They wouldn't connect to the Filemaker databases on the server anymore. Also, when I tried top open network drives, I would get "local device already in use".

 

However, I came in Monday and everything was working fine again. Weird, I didn't do anything it just started working.

This makes me point to a misconfiguration in the network properties/dhcp server with dns. Glad it is working for you, expect it to fail again at some random point.

  Maybe it will be another lightening storm, maybe it will be because someone was stomping their feet, maybe because someone hit the lottery, maybe because someone picked their nose and flung it at the server....all of which will be absolutely coincidental to a misconfiguration that causes random connection issues.

 

 

17 minutes ago, sc302 said:

This makes me point to a misconfiguration in the network properties/dhcp server with dns. Glad it is working for you, expect it to fail again at some random point.

  Maybe it will be another lightening storm, maybe it will be because someone was stomping their feet, maybe because someone hit the lottery, maybe because someone picked their nose and flung it at the server....all of which will be absolutely coincidental to a misconfiguration that causes random connection issues.

 

 

Haha, good way of putting it. I did a clean install of Server 2008 a few months ago with a little help from Budman. I did everything myself though and I have no training in Server OS's. I just used the built in wizard to set up the domain. If I recall, I used dcpromo.exe. Is there anything I should do that might point to the misconfiguration?

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.