How can I restore internet connection to my server?

S

scharfshutze009

2[H]4U
Joined
May 22, 2010
Messages
2,079
I lost internet connection to my server for some reason and I don't know if it has anything to do with AppArmor or Masquerading or just the DNS client service or DHCP client service. I marquerading seems to be fine because the final command involved with it in Ubuntu seems to be working, but my server can't resolve DNS to the ISP and can't ping the Windows Test Desktop. However, it can ping the Windows Test Server and the router. Can anyone help I'm using Ubuntu 16.4.2 LTS on it because Ubuntu Server 16.4.2 LTS had media issues and said unable to mount CD about half way through starting the installation. I haven't tried Debian yet because I haven't found or printed the documentation for it yet and I haven't found the server version. Also, Debian is 8 CD isn't of just one DVD or DL DVD.
 
You don't need 8 cd's for debian...

I don't know anything about your setup, but most ISP provided DNS services blow balls. Trying setting it to 8.8.8.8; 8.8.4.4 (googles DNS) and stop using DHCP. Statically assign an IP to the active server connection that is on the subnet of your network, but outside of (preferably) the DHCP pool.

This is all of course based on the assumption your switch/router is setup correctly...
 
Check /etc/resolv.conf - if it's empty you need to fill it by hand to restore function (i.e. add line Nameserver 8.8.8.8 to it. It should be generated automatically so if its empty you have a problem somewhere else.

Also note that if you fill resolv.conf by hand it will be empty again after the next reboot.
 
B00nie said:
Check /etc/resolv.conf - if it's empty you need to fill it by hand to restore function (i.e. add line Nameserver 8.8.8.8 to it. It should be generated automatically so if its empty you have a problem somewhere else.

Also note that if you fill resolv.conf by hand it will be empty again after the next reboot.
Click to expand...

Thanks B00nie that might help and I tried it by adding my ISP DNS server IP, but it made it worse because then it couldn't resolve dns at all after restoring my routers previous configs with the mirror and startup backups. Maybe I did something wrong though, but I'm affraid to disclose my ISP dns IP address on here, so is there a way you can still help me. I'm using a Cisco RV042G small business router.
 
KarsusTG said:
You don't need 8 cd's for debian...

I don't know anything about your setup, but most ISP provided DNS services blow balls. Trying setting it to 8.8.8.8; 8.8.4.4 (googles DNS) and stop using DHCP. Statically assign an IP to the active server connection that is on the subnet of your network, but outside of (preferably) the DHCP pool.

This is all of course based on the assumption your switch/router is setup correctly...
Click to expand...

Hmm that might be a good Idea. Thanks.
 
KarsusTG said:
You don't need 8 cd's for debian...

I don't know anything about your setup, but most ISP provided DNS services blow balls. Trying setting it to 8.8.8.8; 8.8.4.4 (googles DNS) and stop using DHCP. Statically assign an IP to the active server connection that is on the subnet of your network, but outside of (preferably) the DHCP pool.

This is all of course based on the assumption your switch/router is setup correctly...
Click to expand...

I tried this and it still won't let me browse. Are you sure it's not my /etc/network/inferfaces file, which is configured as follows:

#intefaces(5) file used by ifup(8) and ifdown(8)
auto eth0
auto eth1
auto lo
iface lo inet loopback
pre-up /sbin/ethtool -s eth0 speed 1000 duplex full
pre-up /sbin/ethtool -s eth1 speed 1000 duplex full
 
scharfshutze009 said:
I tried this and it still won't let me browse. Are you sure it's not my /etc/network/inferfaces file, which is configured as follows:

#intefaces(5) file used by ifup(8) and ifdown(8)
auto eth0
auto eth1
auto lo
iface lo inet loopback
pre-up /sbin/ethtool -s eth0 speed 1000 duplex full
pre-up /sbin/ethtool -s eth1 speed 1000 duplex full
Click to expand...

Your interfaces file does not define any interfaces for eth0 and eth1... You should have something to the order of:
iface eth0 inet dhcp
there if your server is using dhcp.
 
B00nie said:
Your interfaces file does not define any interfaces for eth0 and eth1... You should have something to the order of:
iface eth0 inet dhcp
there if your server is using dhcp.
Click to expand...

Thanks B00nie I will try this. I had no idea either because I didn't see this in the documentation. Also it's hard to trust internet sources because the answers could be spam or outdated syntax even if I should try first before I make that assumption because in this case it probably wouldn't do much harm.
 
B00nie said:
Your interfaces file does not define any interfaces for eth0 and eth1... You should have something to the order of:
iface eth0 inet dhcp
there if your server is using dhcp.
Click to expand...

I tried this, but it still didn't help and everytime I restart the network with the service-network restart command I gain to many ipv6 global address, which the only the only this that got rid of the unnecessary IPv6 addresses was rebooting the computer. I checked my firewall and HTTP port 80, 8080, and HTTPS 443 are allowed so I don't know what the problem is anymore unless dhcp isn't allowed.
 
B00nie said:
Your interfaces file does not define any interfaces for eth0 and eth1... You should have something to the order of:
iface eth0 inet dhcp
there if your server is using dhcp.
Click to expand...

I tried adding almost every if not every HTTP, DHCP, and DNS port to the allowed list for the software firewall, but it didn't help.

Here's the contents of my /etc/network/interfaces file:


# interfaces(5) file used by ifup(8) and ifdown(8)
auto eth0
auto eth1
auto lo
iface lo inet loopback
iface eth0 inet dhcp
#iface eth1 inet dhcp
pre-up /sbin/ethtool -s eth0 speed 1000 duplex full
pre-up /sbin/ethtool -s eth1 speed 1000 duplex full

Could it be the additions I made to the /etc/ufw/before.rules file as follows:


# Forward traffic from eth0
-A POSTROUTING -s 10.18.90.0/24 -o eth0 -j MASQUERADE

-A POSTROUTING -s 10.18.90.0/24 -o eth1 -j MASQUERADE

-A POSTROUTING -s 10.18.90.0/24 -o eth2 -j MASQUERADE
 
scharfshutze009 said:
I tried adding almost every if not every HTTP, DHCP, and DNS port to the allowed list for the software firewall, but it didn't help.

Here's the contents of my /etc/network/interfaces file:


# interfaces(5) file used by ifup(8) and ifdown(8)
auto eth0
auto eth1
auto lo
iface lo inet loopback
iface eth0 inet dhcp
#iface eth1 inet dhcp
pre-up /sbin/ethtool -s eth0 speed 1000 duplex full
pre-up /sbin/ethtool -s eth1 speed 1000 duplex full

Could it be the additions I made to the /etc/ufw/before.rules file as follows:


# Forward traffic from eth0
-A POSTROUTING -s 10.18.90.0/24 -o eth0 -j MASQUERADE

-A POSTROUTING -s 10.18.90.0/24 -o eth1 -j MASQUERADE

-A POSTROUTING -s 10.18.90.0/24 -o eth2 -j MASQUERADE
Click to expand...

Disable your firewall completely to check if it's the problem. When troubleshooting you need to eliminate unnecessary variables first.

Hint: Your computer will not get infected instantly even if you disable the firewall. At least if you have used a decent username and password to your public interfaces.
 
There is no reason to put the ethtool lines into your configuration.
Gigabit ethernet has auto-negotiation built in. It's not 1999 anymore.
 
mwarps said:
There is no reason to put the ethtool lines into your configuration.
Gigabit ethernet has auto-negotiation built in. It's not 1999 anymore.
Click to expand...

Thanks, but no matter what I do even if I comment these lines out. I can't restore internet connection. I had it resolving DNS again for a while, but IT still couldn't connect to a website or install with repositories or wget. Therefore, I don't know what I did wrong or how I can fix it.

I will say this though. I tried using the following command:

sudo ifup eth0

and it said something like:

interface name eth0 doesn't exist or not known.
 
Last edited:
1. is this box on an internal private lan?
2. you said you are using dhcp, dhcp from your home router or from the ISP? why are you using dhcp on a "server" anyways?
3. what is the output of "ifconfig -a"
4. what is the output of "netstat -rn"
5. you said you can ping on your local network/router. can you ping 8.8.8.8?
 
scharfshutze009 said:
Thanks, but no matter what I do even if I comment these lines out. I can't restore internet connection. I had it resolving DNS again for a while, but IT still couldn't connect to a website or install with repositories or wget. Therefore, I don't know what I did wrong or how I can fix it.

I will say this though. I tried using the following command:

sudo ifup eth0

and it said something like:

interface name eth0 doesn't exist or not known.
Click to expand...

You have to check what your interfaces are called, they might not be eth0 / eth1 or you have the cable plugged to eth1 not eth0. (Next question: which cable? LOL)
 
tbg said:
1. is this box on an internal private lan?
2. you said you are using dhcp, dhcp from your home router or from the ISP? why are you using dhcp on a "server" anyways?
3. what is the output of "ifconfig -a"
4. what is the output of "netstat -rn"
5. you said you can ping on your local network/router. can you ping 8.8.8.8?
Click to expand...

1. Yes the server is on a private lan connected to the slowest business class cable internet I could afford.
2. I not using the server as a dhcp server. I just meant I'm using DHCP client.
3. I'm going to have trouble showing you the output of the ifconfig -a because the machine doesn't have internet connection, but I could redirect the output to a text file and copy the file to USB Flash like so: ifconfig -a > /home/username/ifconfig.out
4. Same as answer to question 3 except I would redirect it to a text file as follows: netstat -m > /home/username/netstat.out
5. Yes I can ping my router, but I'll have to check and see if I can ping 8.8.8.8.
 
B00nie said:
You have to check what your interfaces are called, they might not be eth0 / eth1 or you have the cable plugged to eth1 not eth0. (Next question: which cable? LOL)
Click to expand...

I know that, but ifconfig confuses me because it calls eth0 as enp6s0 and eth1 as enp7s0. What do you mean which cable and why do you need to know or does that matter?
 
scharfshutze009 said:
I know that, but ifconfig confuses me because it calls eth0 as enp6s0 and eth1 as enp7s0. What do you mean which cable and why do you need to know or does that matter?
Click to expand...

If your interfaces are called enp6s0 you have to use that line in the 'iface enp6s0 inet dhcp' obviously or it won't work. And as obviously you should know which ethernet port your cable is plugged into unless you have both cables plugged in at all times. DHCP can't work if the cable is not connected :D
 
tbg said:
1. is this box on an internal private lan?
2. you said you are using dhcp, dhcp from your home router or from the ISP? why are you using dhcp on a "server" anyways?
3. what is the output of "ifconfig -a"
4. what is the output of "netstat -rn"
5. you said you can ping on your local network/router. can you ping 8.8.8.8?
Click to expand...

Here are the answers to questions 3, 4, and 5:

3.

enp6s0 Link encap:Ethernet HWaddr 74:d4:35:64:41:a0
inet addr:10.18.90.110 Bcast:10.18.90.255 Mask:255.255.255.0
inet6 addr: fc00::759b:f01f:a934:4486/64 Scope:Global
inet6 addr: fe80::98a0:83db:44e4:dd7f/64 Scope:Link
inet6 addr: fc00::127/128 Scope:Global
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:21 errors:0 dropped:0 overruns:0 frame:0
TX packets:138 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:2555 (2.5 KB) TX bytes:23318 (23.3 KB)
Interrupt:18 Memory:dfb00000-dfb20000

enp7s0 Link encap:Ethernet HWaddr 74:d4:35:64:41:a1
UP BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
Interrupt:19 Memory:dfa00000-dfa20000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets:832 errors:0 dropped:0 overruns:0 frame:0
TX packets:832 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1
RX bytes:60519 (60.5 KB) TX bytes:60519 (60.5 KB)

As you can see the enp6s0 connection the server uses has an IPv4 and IPv6 address. While the enp7s0 doesn't because it doesn't have anything connected to it or any devices connected to it that are powered on.


4.

Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
10.18.90.0 0.0.0.0 255.255.255.0 U 0 0 0 enp6s0
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 enp6s0

As you can see here the interface enp6s0 has an IPv4 network address, but the second or dns has an APIPA (Automatic Private IP Address) as an IPv4 address for some reason.

5. When I ping google or 8.8.8.8 I get the following response:

connect: Network is unreachable
 
Last edited:
When I do a dig on my ISP get something similar to the following:



; <<>> DiG 9.10.3-P4-Ubuntu <<>> www.ispname.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: REFUSED, id: 36741
;; flags: qr rd ra ad; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;www.ispname.com. IN A

;; Query time: 0 msec
;; SERVER: 127.0.1.1#53(127.0.1.1)
;; WHEN: Thu Apr 20 23:31:19 EDT 2017
;; MSG SIZE rcvd: 29
 
I'm wondering if I'm violating the Ubuntu License by using Desktop on a Server though, but I'm series when I say the Ubuntu Server Installer said unable to mount CD partially throught the installer and I don't know what to do about it.
 
Your local resolver is probably not configured to allow recursion. It is misconfigured.

post /etc/bind/named.conf.options




scharfshutze009 said:
When I do a dig on my ISP get something similar to the following:



; <<>> DiG 9.10.3-P4-Ubuntu <<>> www.ispname.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: REFUSED, id: 36741
;; flags: qr rd ra ad; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;www.ispname.com. IN A

;; Query time: 0 msec
;; SERVER: 127.0.1.1#53(127.0.1.1)
;; WHEN: Thu Apr 20 23:31:19 EDT 2017
;; MSG SIZE rcvd: 29
Click to expand...
 
scharfshutze009 said:
Here are the answers to questions 3, 4, and 5:


4.

Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
10.18.90.0 0.0.0.0 255.255.255.0 U 0 0 0 enp6s0
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 enp6s0

As you can see here the interface enp6s0 has an IPv4 network address, but the second or dns has an APIPA (Automatic Private IP Address) as an IPv4 address for some reason.

5. When I ping google or 8.8.8.8 I get the following response:

connect: Network is unreachable
Click to expand...

you do not have a default route, you would have something like

$ netstat -rn
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
0.0.0.0 192.168.1.1 0.0.0.0 UG 0 0 0 enp0s25
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 enp0s25

it does not know where to send anything on the 10.18.90.0 net. If it was just a DNS issue you would be able to ping 8.8.8.8.
 
This is not an issue of you violating the Ubuntu License, this is you not knowing what you are doing.
why do you have eth0 and eth1 in your /etc/network/interfaces when those are not even the names of your interfaces?

do something like this...

1 backup /etc/network/interfaces

2 edit /etc/network/interfaces delete everything and put the stuff bellow.
############################
# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
auto enp6s0
iface enp6s0 inet static
address 10.18.90.110
netmask 255.255.255.0
gateway 10.18.90.???
dns-nameservers 8.8.8.8
dns-nameservers 8.8.4.4
############################
3 reboot

I don't know the last octet of your router, you have to fill that in
 
tbg said:
you do not have a default route, you would have something like

$ netstat -rn
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
0.0.0.0 192.168.1.1 0.0.0.0 UG 0 0 0 enp0s25
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 enp0s25

it does not know where to send anything on the 10.18.90.0 net. If it was just a DNS issue you would be able to ping 8.8.8.8.
Click to expand...

Are you sure it's not because it's masquerading the IP. Also I'm just using a Cisco small business router too, so are you sure I need the default route.
 
tbg said:
This is not an issue of you violating the Ubuntu License, this is you not knowing what you are doing.
why do you have eth0 and eth1 in your /etc/network/interfaces when those are not even the names of your interfaces?

do something like this...

1 backup /etc/network/interfaces

2 edit /etc/network/interfaces delete everything and put the stuff bellow.
############################
# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
auto enp6s0
iface enp6s0 inet static
address 10.18.90.110
netmask 255.255.255.0
gateway 10.18.90.???
dns-nameservers 8.8.8.8
dns-nameservers 8.8.4.4
############################
3 reboot

I don't know the last octet of your router, you have to fill that in
Click to expand...

Sorry, but I only edited this file after I could no longer connect to the internet, so I don't have a backup of the original configuration if it became corrupt. I'll try this though.
 
scharfshutze009 said:
Are you sure it's not because it's masquerading the IP. Also I'm just using a Cisco small business router too, so are you sure I need the default route.
Click to expand...
It needs to know where to send traffic to get it off the local subnet.
 
tbg said:
This is not an issue of you violating the Ubuntu License, this is you not knowing what you are doing.
why do you have eth0 and eth1 in your /etc/network/interfaces when those are not even the names of your interfaces?

do something like this...

1 backup /etc/network/interfaces

2 edit /etc/network/interfaces delete everything and put the stuff bellow.
############################
# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
auto enp6s0
iface enp6s0 inet static
address 10.18.90.110
netmask 255.255.255.0
gateway 10.18.90.???
dns-nameservers 8.8.8.8
dns-nameservers 8.8.4.4
############################
3 reboot

I don't know the last octet of your router, you have to fill that in
Click to expand...

This worked, but after installing a few programs and attempting to install Nagios it said connection timed out when I did a dig on the IP
 
Also, I like that you can't even figure out how to configure your network, but your going to go setup Nagios.
 
tbg said:
Also, I like that you can't even figure out how to configure your network, but your going to go setup Nagios.
Click to expand...

Well I imagine in the beginning we all learned by biting off a bit more then we could chew. Your doing it right, if months later you have an ahhh haaa moment. The kind where you realise you wasted hours messing with something that you now understand.

But ya if you manually set the DNS to googles why would you be worried about the DNS from the IP. Check the file again did it change your DNS.

I didn't read through the entire thread so this may sound silly I'm not sure how you are setup, but what is the DNS set to on your router. Not sure if you have DHCP turned on or not but its possible it passed your machine a new DNS if got a new one from your ISP. If your ISP doesn't provide a static IP you may want to leave DHCP on, cisco router should have an option to set a static DNS, which you could change to googles dns.
 
I would not do DHCP on anything I planned to use as a server, even static, if I can avoid it at least.
I also probably would not leave the google dns, but I did not know what dns he had available when I gave him the basic config to get him up and running.

When the network is not working you have to start with the basic trouble shooting steps.
1) Can you ping known pingable IPs on the local network? The router on your local lan is a good test here.
If no, then something is probably mis-configured for the interface. check things like
- "ifconfig -a" or "ip addr list"
- look at /etc/network/interfaces in ubuntu or /etc/sysconfig/network-scripts/ifcfg-(interface) in redhat based os.
If yes, go to next test..

2) Can you ping an IP on the internet,(or other network you are trying to reach) something like 8.8.8.8 which is easy to remember.
If no, something is probably wrong with the routing. The box needs to know how to get traffic off the local lan to get to IPs on other networks.
- "netstat -rn" or "ip route list"
- tracerote <ip>
If yes. go to next test.

3) can you ping the FQDN
If no, start looking in to the DNS stuff
- look at /etc/resolve.conf , unfortunately in ubuntu this probably points to localhost 127.0.0.1.
- "nmcli device show"
- look at /etc/network/interfaces
If yes, go to the next step

4) You can reach the box but no response for some service?
Check firewall rules on both boxes.
Check "netstat -an | grep <port>" on server

That is kind of a very simplified set of steps to looking at networking issues. Its a lot easier to try to fix things if you know where the problem is then if you just flail about making changes to config files.
You need to be able to talk on the local lan before you would get traffic routed. you need to get traffic routing correctly before dns is going to mater to the internet. The box needs to be reachable in general before you worry about why is X service not working.
 
ChadD said:
Well I imagine in the beginning we all learned by biting off a bit more then we could chew. Your doing it right, if months later you have an ahhh haaa moment. The kind where you realise you wasted hours messing with something that you now understand.

But ya if you manually set the DNS to googles why would you be worried about the DNS from the IP. Check the file again did it change your DNS.

I didn't read through the entire thread so this may sound silly I'm not sure how you are setup, but what is the DNS set to on your router. Not sure if you have DHCP turned on or not but its possible it passed your machine a new DNS if got a new one from your ISP. If your ISP doesn't provide a static IP you may want to leave DHCP on, cisco router should have an option to set a static DNS, which you could change to googles dns.
Click to expand...

I have googles dns set as the alternate dns, so that is why I did a dig on the ISP.

I'm not telling you the dns on the router because it's my ISP's and anyone who views this thread can see it, which technically is the reason I'm not supposed to ask for help on a forum.

I had DCHP turned in the Networks GUI config, so maybe thats why it stopped working. However, now the settings in networks are set to static, but I guess I need to change the router to static too.

I'm just now realizing that the 8.8.8.8 8.8.4.4 you're all referring to is mentioned in the documentation I printed out double-sided.

I will set the router dns to static and as googles, but right now it's set to my ISP's.
 
tbg said:
I would not do DHCP on anything I planned to use as a server, even static, if I can avoid it at least.
I also probably would not leave the google dns, but I did not know what dns he had available when I gave him the basic config to get him up and running.

When the network is not working you have to start with the basic trouble shooting steps.
1) Can you ping known pingable IPs on the local network? The router on your local lan is a good test here.
If no, then something is probably mis-configured for the interface. check things like
- "ifconfig -a" or "ip addr list"
- look at /etc/network/interfaces in ubuntu or /etc/sysconfig/network-scripts/ifcfg-(interface) in redhat based os.
If yes, go to next test..

2) Can you ping an IP on the internet,(or other network you are trying to reach) something like 8.8.8.8 which is easy to remember.
If no, something is probably wrong with the routing. The box needs to know how to get traffic off the local lan to get to IPs on other networks.
- "netstat -rn" or "ip route list"
- tracerote <ip>
If yes. go to next test.

3) can you ping the FQDN
If no, start looking in to the DNS stuff
- look at /etc/resolve.conf , unfortunately in ubuntu this probably points to localhost 127.0.0.1.
- "nmcli device show"
- look at /etc/network/interfaces
If yes, go to the next step

4) You can reach the box but no response for some service?
Check firewall rules on both boxes.
Check "netstat -an | grep <port>" on server

That is kind of a very simplified set of steps to looking at networking issues. Its a lot easier to try to fix things if you know where the problem is then if you just flail about making changes to config files.
You need to be able to talk on the local lan before you would get traffic routed. you need to get traffic routing correctly before dns is going to mater to the internet. The box needs to be reachable in general before you worry about why is X service not working.
Click to expand...

1. Yes can ping every computer on the network attached to the router and I checked all my setting on both computers, such as firewalls, ifconfig or ip addr list, dig or nslookup. However, I had Networks in the GUI configured as DHCP and now its static, but the documentation didn't say how to do ipv6, so I searched the internet.

2. Yes I can ping my ISP and google and 8.8.8.8 or 8.8.4.4 mentioned in the documentation.

3. If you mean ping FQDN (Fully Qualified Domain Name) as in www.google.com then no. Also, the /etc/resolv.conf file is empty and if I backed it up I think I overwrote it. I'm not showing the output of the nmcli on here, because anyone on the internet can see it.

4. I check the firewall rules by using the sudo ufw status command and everything's fine. As for the Windows Machine it's fine too, but I did have more trouble understanding it.
 
ChadD said:
Well I imagine in the beginning we all learned by biting off a bit more then we could chew. Your doing it right, if months later you have an ahhh haaa moment. The kind where you realise you wasted hours messing with something that you now understand.

But ya if you manually set the DNS to googles why would you be worried about the DNS from the IP. Check the file again did it change your DNS.

I didn't read through the entire thread so this may sound silly I'm not sure how you are setup, but what is the DNS set to on your router. Not sure if you have DHCP turned on or not but its possible it passed your machine a new DNS if got a new one from your ISP. If your ISP doesn't provide a static IP you may want to leave DHCP on, cisco router should have an option to set a static DNS, which you could change to googles dns.
Click to expand...

I tried changing the dns in the interfaces file to 8.8.8.8 and 8.8.4.4, but it didn't work. Then I temporarilly added the following to /etc/resolv.conf:

nameserver 8.8.8.8
nameserver 8.8.4.4

and it temporarily fixed the problem for now.
 
you know changes to the interface file do not happen until the service is restarted right?
 
tbg said:
I would not do DHCP on anything I planned to use as a server, even static, if I can avoid it at least.
I also probably would not leave the google dns, but I did not know what dns he had available when I gave him the basic config to get him up and running.

When the network is not working you have to start with the basic trouble shooting steps.
1) Can you ping known pingable IPs on the local network? The router on your local lan is a good test here.
If no, then something is probably mis-configured for the interface. check things like
- "ifconfig -a" or "ip addr list"
- look at /etc/network/interfaces in ubuntu or /etc/sysconfig/network-scripts/ifcfg-(interface) in redhat based os.
If yes, go to next test..

2) Can you ping an IP on the internet,(or other network you are trying to reach) something like 8.8.8.8 which is easy to remember.
If no, something is probably wrong with the routing. The box needs to know how to get traffic off the local lan to get to IPs on other networks.
- "netstat -rn" or "ip route list"
- tracerote <ip>
If yes. go to next test.

3) can you ping the FQDN
If no, start looking in to the DNS stuff
- look at /etc/resolve.conf , unfortunately in ubuntu this probably points to localhost 127.0.0.1.
- "nmcli device show"
- look at /etc/network/interfaces
If yes, go to the next step

4) You can reach the box but no response for some service?
Check firewall rules on both boxes.
Check "netstat -an | grep <port>" on server

That is kind of a very simplified set of steps to looking at networking issues. Its a lot easier to try to fix things if you know where the problem is then if you just flail about making changes to config files.
You need to be able to talk on the local lan before you would get traffic routed. you need to get traffic routing correctly before dns is going to mater to the internet. The box needs to be reachable in general before you worry about why is X service not working.
Click to expand...

Something happened, since my last responses and since you originally replied with this response because I can only do number 3 to the server and enp6s0 and enp7s0 don't show up in the output of the ifconfig command by itself. What don't I know or what am I not thinking of? Does it matter if Networks the GUI interface is configured manually or with DCHP because shouldn't it be insync with the /etc/network/interfaces file and if not then what because I don't know of any mention in the documentation or any other solution? i had everything working perfectly until I shutdown my server last weekend and turned it back on a few days ago or attempted to change anything.
 
You must log in or register to reply here.
Back
Top