Let me refer responders to this thread on Linux Mint forums:
I've seen this four times in the past two weeks or so. The loss of connection does eventually clear after an unknown time (longest has been 4 days).
In a nut shell, almost all URL-s work but key ones you'd need to do (actual) stuff like google.com, linkedin.com, youtube.com or yahoo.com, etc. fail. While other sites you'd think are related work. It is unpredictable. Tonight is a good example. URL:
The issue affects everything I've tested (for the 'missing' url)...
I don't know if it is specifically the DNS.
The fortunate experience swapping to and back from a mobile broadband modem was not a 'workaround' just a lucky-break. Results from such a swap don't always work.
I think the DNS issue(s) are just a symptom. The message for
Gives back is...
curl: (7) Couldn't connect to server
Which I'm guessing means it can "see" the server but won't connect. As mentioned I can send email from other things using the same access point e.g. Android phone so wondering what diagnostics there are for this kind of thing in Linux Mint so I can get a handle on what part of the stack is getting stuck?
I have some new information. It may help others, or give the brain's trust a tip on how to resolve this problem. This evening I was unable to access GitHub, stackexchange.com, netbeans, and Wikipedia, to name a few. I am here now, so what happened?
When I pressed F5 here, I received this banner message at the top of the screen:
woops I thought, it has happened again. I really wanted to check for any responses, because as you can imagine this error is retarding significant progress -- So I have a little credit on a USB modem. I thought I'd try to use that.
- Swapped-out the WI-FI access point
- Connected to the USB/mobile network
Result: No change, can't get wikipedia and the annoying script message thing was showing on this page.
- Disconnect from the USB/Mobile broadband network
- Reconnect he Wi-Fi access point
One of my Dig tests gave a different result After connecting (back) to the Wi-Fi access point. So (now) ...
- The annoying banner message on StackExchange is gone. I can see wikipedia and github again.
It appears that something in the stack doesn't get reset/refreshed until the 'Wired' (access point) connection is reconnected. But not just reconnected; being usurped so that it has to reconnect everything (again).
The other odd thing; is that the mobile broadband did NOT refresh/reset the (effective) stuff on the data comm. stack. Why not? How?
More importantly what may I do to force the reset/refresh on the Wired connection and the Wi-Fi connection without renewing a defunct mobile broadband when the current merger residual credits drip away??!!
Otherwise things remain the same as they were for the last update
At the moment all I can say is that google search (working) and gmail (not working) demonstrate the following responses from dig.
$ dig google.com
; <<>> DiG 9.9.5-3ubuntu0.5-Ubuntu <<>> google.com ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 19398 ;; flags: qr rd ra; QUERY: 1, ANSWER: 15, AUTHORITY: 0, ADDITIONAL: 0 ;; QUESTION SECTION: ;google.com. IN A ;; ANSWER SECTION: google.com. 114 IN A 18.104.22.168 google.com. 114 IN A 22.214.171.124 google.com. 114 IN A 126.96.36.199 google.com. 114 IN A 188.8.131.52 google.com. 114 IN A 184.108.40.206 google.com. 114 IN A 220.127.116.11 google.com. 114 IN A 18.104.22.168 google.com. 114 IN A 22.214.171.124 google.com. 114 IN A 126.96.36.199 google.com. 114 IN A 188.8.131.52 google.com. 114 IN A 184.108.40.206 google.com. 114 IN A 220.127.116.11 google.com. 114 IN A 18.104.22.168 google.com. 114 IN A 22.214.171.124 google.com. 114 IN A 126.96.36.199 ;; Query time: 108 msec ;; SERVER: 127.0.1.1#53(127.0.1.1) ;; WHEN: Tue Nov 03 23:14:22 AEDT 2015 ;; MSG SIZE rcvd: 268
$ dig mail.google.com
; <<>> DiG 9.9.5-3ubuntu0.5-Ubuntu <<>> mail.google.com ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 40641 ;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 0 ;; QUESTION SECTION: ;mail.google.com. IN A ;; AUTHORITY SECTION: . 49606 IN NS localhost. ;; Query time: 106 msec ;; SERVER: 127.0.1.1#53(127.0.1.1) ;; WHEN: Tue Nov 03 23:15:02 AEDT 2015 ;; MSG SIZE rcvd: 55
In contrast the dig output on a working box I have access to shows a different story as far as nameservers go.
On a working machine:
$ dig gmail.com (added in update)
; <<>> DiG 9.6-ESV-R11 <<>> gmail.com ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 22330 ;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 0, ADDITIONAL: 0 ;; QUESTION SECTION: ;gmail.com. IN A ;; ANSWER SECTION: gmail.com. 192 IN A 188.8.131.52 ;; Query time: 15 msec ;; SERVER: 192.168.172.1#53(192.168.172.1) ;; WHEN: Wed Nov 04 14:02:50 EST 2015 ;; MSG SIZE rcvd: 43
In the first non-working example, the Linux set-up is using localhost. To me, that sounds like I need it to do a refresh or something. Comments?
Nitty gritty detail, is:
- Linux Mint v17.2 - Rafaela (Cinnamon 64-bit)
- dnsmasq version 2.68
I tested the network restart suggested on the forum, alas it didn't assist me accessing gmail. Back to the drawing boars/stackexchange :-)
Apart from the question on the Mint forum I haven't found anything to indicate 'yea' or 'nay' for this problem. The first post on this issue was way back in 2011 and it looks like it was skipped as a 'non-good' question so this phenomena has been about for a while. It would be good to find an effective response. Hopefully one of us has a fix already.
Suggestions welcome and happy to look-up further information.