Solved Desktop Frequently disconnected

tuxador

tuxador

Hi I have Freebsd R13 installed in my workstation, which also serves for local storage via SMB, it's connected to my router with Ethernet.
Since my last freebsd-install upgrade (to p5 revision) I got very frequent and rendom disconnection from the internet , I can access the local printer , samba works as expected, but I struggle to get a stable internet connection.
My IP is static 192.168.100.33.
My resolve.conf includes 192.168.100.1 and 0.0.0.0
IPv6 is disabled.
 
Geezer said:
I can understand 1.1.1.1 and 8.8.8.8, but I don't think 0.0.0.0 will resolve anything.
Click to expand...
I tried with 1.1.1.1 and 8.8.8.8 without success.
Note that this issus occured since the last update.
Another noticeable thing is that my coworker has to devices connected to my wifi , making the number of connected devices at once to 10.
 
tuxador said:
I got very frequent and rendom disconnection from the internet , I can access the local printer , samba works as expected, but I struggle to get a stable internet connection.
Click to expand...
So, access to your local network is good, but not anything behind your router. Then maybe you should look for the fault there.

NB. DNS issues have nothing to do with disconnects. There is no DNS activity when a TCP session is up. DNS is only used to resolve the IP address, once it has the IP address DNS is not involved any more. Thus, DNS issues will never cause disconnects.
 
crappy routers may bomb for various reasons
including but not limited to shitty power adaptors, nat table overflow...
 
The problem still persists.
The weird fact is I can ping my local printer:
ping 192.168.100.20
But pinging my router gives me:
Ping: sendto: Host is down.
 
Not using local_unbound at all, I also have no /etc/resolvconf.conf file. This appears to be normal for an out-of-the-box 13.0-RELEASE install.

If I simply delete EVERYTHING from /etc/resolv.conf, and then run service netif restart, it will obtain my ISP's two DNS servers' IP addresses, presumably using DHCP in conjunction with the resolvconf command, and THEN my /etc/resolv.conf will be auto-magically edited to contain something like:
Code: 
# Generated by resolvconf
nameserver 71.10.216.1
nameserver 71.10.216.2

...or...

Code: 
# Generated by resolvconf
nameserver 71.10.216.1
nameserver 71.10.216.2
nameserver 192.168.1.1

... depending on whether or not my router is configured to function also as a nameserver.

In other words, I don't have to put anything in /etc/resolv.conf. Instead, resolvconf(8) will be automatically invoked to configure that file for me, every time networking is restarted.

I suspect that your static IP assignment might be related to the problem, especially if you have recently added other devices to your network.

I suggest changing your /etc/rc.conf to let DHCP configure your networking for you, at least temporarily. Then, besides seeing what IP addresses will have been assigned for you and your nameservers by DHCP, you can also try to ping 192.168.100.33 and see if some other device has already been assigned with the static IP you are trying to assign for yourself. You can always reestablish your static IP address assignment afterwards if you wish.

You can refer to resolvconf(8) for more information.
 
Vull said:
Not using local_unbound at all, I also have no /etc/resolvconf.conf file. This appears to be normal for an out-of-the-box 13.0-RELEASE install.

If I simply delete EVERYTHING from /etc/resolv.conf, and then run service netif restart, it will obtain my ISP's two DNS servers' IP addresses, presumably using DHCP in conjunction with the resolvconf command, and THEN my /etc/resolv.conf will be auto-magically edited to contain something like:
Code: 
# Generated by resolvconf
nameserver 71.10.216.1
nameserver 71.10.216.2

...or...

Code: 
# Generated by resolvconf
nameserver 71.10.216.1
nameserver 71.10.216.2
nameserver 192.168.1.1

... depending on whether or not my router is configured to function also as a nameserver.

In other words, I don't have to put anything in /etc/resolv.conf. Instead, resolvconf(8) will be automatically invoked to configure that file for me, every time networking is restarted.

I suspect that your static IP assignment might be related to the problem, especially if you have recently added other devices to your network.

I suggest changing your /etc/rc.conf to let DHCP configure your networking for you, at least temporarily. Then, besides seeing what IP addresses will have been assigned for you and your nameservers by DHCP, you can also try to ping 192.168.100.33 and see if some other device has already been assigned with the static IP you are trying to assign for yourself. You can always reestablish your static IP address assignment afterwards if you wish.

You can refer to resolvconf(8) for more information.
Click to expand...
Unfortunately this didn't work.
Now I can't even ping to my local printer.
 
ifconfig -a to make sure you have an IP address (do you have it configured for DHCP or static?)
netstat -rn to show the routing table to make sure you have a route to your router and the printer you are trying to ping
 
tuxador said:
Yeah but the problem occurs only in my FreeBSD machine.
2 Archlinux laptops + one windows 10 desktop + 4 android phones work ok
Click to expand...
These all connected by wifi, right? And your FreeBSD machine -- by ethernet, right?
Rebooting the wifi router helps sometimes. Another good question is, how are those coworkers using internet? Could it be they're just eating all the traffic so nothing gets left for your FreeBSD machine?
 
service netif restart is not entirely reliable. If you can, avoid this.

tuxador said:
… pinging my router gives me:
Ping: sendto: Host is down.
Click to expand...

For now, maybe ignore that.

The absence of a response to ping, from a router, does not signify an Internet connection problem. Here, for example:

Code: 
% cat /etc/resolv.conf
# Generated by resolvconf
search lan
nameserver 192.168.1.1

% ping -4 192.168.1.1
PING 192.168.1.1 (192.168.1.1): 56 data bytes

– no response. In an adjacent tab:

Code: 
% ping -4 freshports.org
PING freshports.org (54.227.255.74): 56 data bytes
64 bytes from 54.227.255.74: icmp_seq=0 ttl=45 time=81.847 ms
64 bytes from 54.227.255.74: icmp_seq=1 ttl=45 time=82.346 ms
^C
--- freshports.org ping statistics ---
2 packets transmitted, 2 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 81.847/82.097/82.346/0.249 ms
%

– there's a connection to the Internet ☑ with 192.168.1.1 as my router.

After ending the first ping:

Code: 
^C
--- 192.168.1.1 ping statistics ---
581 packets transmitted, 0 packets received, 100.0% packet loss
%

tuxador said:
… My IP is static 192.168.100.33. ?…
Click to expand...

Was the router configured to require static addressing?

tuxador said:
… disconnection from the internet …
Click to expand...

When this happens, what exactly are the Internet symptoms? (Ignoring local devices.)
 
grahamperrin said:
service netif restart is not entirely reliable. If you can, avoid this.



For now, maybe ignore that.

The absence of a response to ping, from a router, does not signify an Internet connection problem. Here, for example:

Code: 
% cat /etc/resolv.conf
# Generated by resolvconf
search lan
nameserver 192.168.1.1

% ping -4 192.168.1.1
PING 192.168.1.1 (192.168.1.1): 56 data bytes

– no response. In an adjacent tab:

Code: 
% ping -4 freshports.org
PING freshports.org (54.227.255.74): 56 data bytes
64 bytes from 54.227.255.74: icmp_seq=0 ttl=45 time=81.847 ms
64 bytes from 54.227.255.74: icmp_seq=1 ttl=45 time=82.346 ms
^C
--- freshports.org ping statistics ---
2 packets transmitted, 2 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 81.847/82.097/82.346/0.249 ms
%

– there's a connection to the Internet ☑ with 192.168.1.1 as my router.

After ending the first ping:

Code: 
^C
--- 192.168.1.1 ping statistics ---
581 packets transmitted, 0 packets received, 100.0% packet loss
%



Was the router configured to require static addressing?



When this happens, what exactly are the Internet symptoms? (Ignoring local devices.)
Click to expand...
- thank you for the ping hint, I didn't know about that.
- yes my router is configured to accept static IP, I have a local printer, an echograph and two CCTV wired with static IP, but sometimes CCTV cameras are also disconnected.
- on the other part the problem never affects my wireless devices.
 
tuxador said:
- thank you for the ping hint, I didn't know about that.
- yes my router is configured to accept static IP, I have a local printer, an echograph and two CCTV wired with static IP, but sometimes CCTV cameras are also disconnected.
- on the other part the problem never affects my wireless devices.
Click to expand...
I noticed that, too. So maybe you pay more attention to your router configuration? Remember: this may NOT be related to your FreeBSD update at all. It's not impossible that the problems in your router started about the time you updated your FreeBSD machine.
 
You must log in or register to reply here.
Back
Top