Internet stops working after reboot

KhattabKhattab Member Posts: 97 ■■□□□□□□□□
Hi All...

I'm convinced this is a small issue, but it's driving me up the wall because i just cant seem to figure out what is causing it!

It's a windows 2003 server (running DNS, AD etc) and was working fine. Last week, the power cut out and ever since, when i reboot the server - the internet stops working.

- I am able to ping the network card of the server and the loopback address.
- I am able to ping the router/default gateway
- I am unable to ping any external addresses (via IP or Domain Names)

I can usually get it back up by fiddling with a multitude of settings (ranging from forwarders, to enabling and disabling wins) then internet starts working. But as soon as i restart the server, low and behold - internet stops working again!

Anyone have any ideas?

Thanks!

Comments

  • JDMurrayJDMurray MSIT InfoSec CISSP SSCP GSEC EnCE C|EH Cloud+ CySA+ CASP+ PenTest+ Security+ Surf City, USAAdmin Posts: 12,268 Admin
    It sounds as if there is a service or application your server requires for Internet connectivity that is not automatically starting on boot-up, and your "fiddling" is causing it to start or be started.
  • Johnny 5Johnny 5 Member Posts: 24 ■□□□□□□□□□
    As soon as the server boots up, I'd try the following:

    1) Test connectivity. As you said, ping the default gateway should be sufficient for this.

    2) Test routing configuration. Run "netstat -rn" and make sure the route for 0.0.0.0 points to the default gateway and has a metric lower than any backup routes (should be 20 by default).

    3) Do a traceroute by IP ("tracert -d"). 66.235.51.212 is one you can use.

    4) Test DNS. Run "ifconfig /all" to see what DNS servers will be used, then run "nslookup" then try www.google.com, etc.
  • KhattabKhattab Member Posts: 97 ■■□□□□□□□□
    1) Test connectivity. As you said, ping the default gateway should be sufficient for this.
    This ping's fine with no probs.
    2) Test routing configuration. Run "netstat -rn" and make sure the route for 0.0.0.0 points to the default gateway and has a metric lower than any backup routes (should be 20 by default).
    I checked this. It is set to the default, both interfaces for the server point to 0.0.0.0 and the metric is 20.
    3) Do a traceroute by IP ("tracert -d"). 66.235.51.212 is one you can use.
    Did the tracert, but it didnt get anywhere! EVERY hop timed out - and there was no hop to our internal router (the server is a DNS server with forwarders set). It doesnt seem to hop to any servers at all..... which i'm guessing is the problem - but where do i go from there?
    4) Test DNS. Run "ifconfig /all" to see what DNS servers will be used, then run "nslookup" then try www.google.com, etc.
    The server is the DNS Server. When i do the nslookup it - DNS Request timed out!

    Any ideas?
  • blargoeblargoe Self-Described Huguenot NC, USAMember Posts: 4,174 ■■■■■■■■■□
    Can you post the output of netstat -rn
    IT guy since 12/00

    Recent: 11/2019 - RHCSA (RHEL 7); 2/2019 - Updated VCP to 6.5 (just a few days before VMware discontinued the re-cert policy...)
    Working on: RHCE/Ansible
    Future: Probably continued Red Hat Immersion, Possibly VCAP Design, or maybe a completely different path. Depends on job demands...
  • sprkymrksprkymrk Member Posts: 4,884 ■■■□□□□□□□
    You might be barking up the wrong tree. Check your router/gateway. It may have been damaged by the power outage.
    All things are possible, only believe.
  • KhattabKhattab Member Posts: 97 ■■□□□□□□□□
    Check your router/gateway. It may have been damaged by the power outage.
    I dont think this is the case.... the gateway is our Netgear modem/router and it seems to be working fine.

    I can access the external interface via the web without any problems, and all other pc's (if i change their DNS Server to our ISP's DNS) are able to access the internet fine.

    I'll post the results of the netstat -rn later tonight and see if that tells us anything...
  • sprkymrksprkymrk Member Posts: 4,884 ■■■□□□□□□□
    Khattab wrote:
    Check your router/gateway. It may have been damaged by the power outage.
    I dont think this is the case.... the gateway is our Netgear modem/router and it seems to be working fine.

    I can access the external interface via the web without any problems, and all other pc's (if i change their DNS Server to our ISP's DNS) are able to access the internet fine.

    I'll post the results of the netstat -rn later tonight and see if that tells us anything...

    Interesting. How about running netdiag /v /l /fix and see what the netdiag.log file looks like.
    All things are possible, only believe.
  • KGhaleonKGhaleon Member Posts: 1,346 ■■■■□□□□□□
    So putting the gateway aside, can you ping other computers on your network? (assuming that there are other machines on your subnet)

    KG
    Present goals: MCAS, MCSA, 70-680
  • KhattabKhattab Member Posts: 97 ■■□□□□□□□□
    To answer KGhaleon - yeah, the Server is able to ping all other PC's on the network....

    I think i've found the problem though (i could have sworn it wasnt like this when i checked it a couple of days ago, but) seems that the route for 0.0.0.0 (netstat -rn) doesnt point to the default gateway... it actually points to the interface of the second NIC (which i am using for RRAS).

    192.168.0.34 - This is the address of the DC/AD/DNS Server.
    192.168.1.34 - This is the address of the interface im using for the RRAS
    192.168.0.249 - This is the address of the Router/Default Gateway.

    ***
    EDIT - the output of the netstat -rn has changed. My next post has the most updated output.
    ***
    You can find the output of the netstat -rn here: http://www.mwa.org.au/netstat.txt


    The issue as i see it, is that for the first entry for 0.0.0.0 - the Interface should be set to the IP Address of the Default Gateway 192.168.0.249.

    The other thing i notice, the final entry - Default Gateway should not be set to 0.34, it should also be set to 192.168.0.249. Is this correct?

    The question is - how do i fix this? Is there a way to flush the settings and restore them to what they should be?

    Thanks in advance guys!
  • blargoeblargoe Self-Described Huguenot NC, USAMember Posts: 4,174 ■■■■■■■■■□
    In the very first line in the routing table from the netstat printout, it looks like it is trying to send traffic destined for the default gateway of 192.168.0.34 through the 192.168.1.34 interface.
    IT guy since 12/00

    Recent: 11/2019 - RHCSA (RHEL 7); 2/2019 - Updated VCP to 6.5 (just a few days before VMware discontinued the re-cert policy...)
    Working on: RHCE/Ansible
    Future: Probably continued Red Hat Immersion, Possibly VCAP Design, or maybe a completely different path. Depends on job demands...
  • KhattabKhattab Member Posts: 97 ■■□□□□□□□□
    I just ran another netstat -rn on the server, and the output has changed.

    http://www.mwa.org.au/netstat.JPG

    Not sure where exactly the problem is - but i am thinking that the issue could be the "Default Gateway" on the last line of output... it is set to 192.168.0.34.... but that is the address of the NIC. Should that be set to the IP address of the router?
  • blargoeblargoe Self-Described Huguenot NC, USAMember Posts: 4,174 ■■■■■■■■■□
    That output shows two "default" routes (the 0.0.0.0 routes). The second one in the list I imagine is the correct one. The correct default route should read 0.0.0.0 0.0.0.0 <ip address of gateway> <ip address of NIC>

    How did you end up with two? Who knows... I noticed you have an "USB to Ethernet adapter" in addition to your built-in broadcom ethernet. If you have a default gateway defined on the card that doesn't connect to the Internet, take it off. If you're not using one of the ethernet adapters, disable it. If you need to keep both, I would uninstall the USB one and see if you can get the routing stuff worked out and try to add it back and see what happens.
    IT guy since 12/00

    Recent: 11/2019 - RHCSA (RHEL 7); 2/2019 - Updated VCP to 6.5 (just a few days before VMware discontinued the re-cert policy...)
    Working on: RHCE/Ansible
    Future: Probably continued Red Hat Immersion, Possibly VCAP Design, or maybe a completely different path. Depends on job demands...
  • KhattabKhattab Member Posts: 97 ■■□□□□□□□□
    It's all working now.

    As you said, the problem was that there were two default routes.

    All was working fine previously, but seems someone set the default gateway on the 2nd NIC to be the IP address of the DNS Server - and that was what caused the problems.

    As soon as i changed it, the 2nd default route disappeared, and all is working fine!

    You guys are legends!!
  • blargoeblargoe Self-Described Huguenot NC, USAMember Posts: 4,174 ■■■■■■■■■□
    Glad it's working now :)
    IT guy since 12/00

    Recent: 11/2019 - RHCSA (RHEL 7); 2/2019 - Updated VCP to 6.5 (just a few days before VMware discontinued the re-cert policy...)
    Working on: RHCE/Ansible
    Future: Probably continued Red Hat Immersion, Possibly VCAP Design, or maybe a completely different path. Depends on job demands...
Sign In or Register to comment.