[mosh-devel] Mosh connections didn't come back after ca. 18000 sec over 2x NAT

[Quentin Smith]

Hi Axel,

Just to check the obvious first - to your knowledge, the server resolved 
to the same IP address before and after you restarted the AP? (That is, 
the server didn't appear to move for any reason?)

What version of mosh are you using? Mosh 1.2.3 adds a new behavior where 
it will try opening a new connection on a new port if it hasn't heard from 
the server in a while (I think 10 seconds?). This is to work around some 
braindead NAT devices that have behavior similar to what you're 
describing.

If you capture a tcpdump on both the client and server for ~30 seconds or 
so, that should show conclusively if your network is eating the packets, 
or the client or server is confused.

--Quentin

On Sun, 30 Dec 2012, Axel Beckert wrote:

> Hi,
>
> today (after close to one year of occassional mosh usage) I had the
> first case where mosh connections didn't come back and I had to kill
> and restart them. After which they worked fine again.
>
> I was online for over 7 days over an so called mobile AP (which does
> NAT). My mobile carrier (E-Plus in Germany) does NAT again, too. I had
> 2 or 3 network interruptions (AP reboot) during this time and the mosh
> connections came back without problems after those reboots.
>
> Last night I seemed to have a network interuption on maybe the
> carrier's side. My firefox hung as it would hung with no DNS, my
> autossh sessions reported "host not found" the whole night and all my
> mosh sessions said "Last reply <n> seconds ago" with <n> being around
> 18000.
>
> After rebooting the AP again, my autossh sessions came back one after
> another, one firefox window became usable again (but the other one was
> still stuck), but none of my mosh sessions came back. I could easily
> start new mosh sessions and in nearly all cases it reported the
> existing sessions as detached.
>
> Any idea what could have cause such a bad lockup in a mosh connection?
> IIRC as of now, mosh does DNS lookups only once at start, so it
> couldn't be a cached bad DNS reply or such.
>
> I still kept one of the non-responding mosh sessions open (now at
> 21521 seconds), so I can possibly debug that session.
>
> 		Kind regards, Axel
> -- 
> /~\  Plain Text Ribbon Campaign                   | Axel Beckert
> \ /  Say No to HTML in E-Mail and News            | abe at deuxchevaux.org  (Mail)
> X   See http://www.asciiribbon.org/              | abe at noone.org (Mail+Jabber)
> / \  I love long mails: http://email.is-not-s.ms/ | http://noone.org/abe/ (Web)
> _______________________________________________
> mosh-devel mailing list
> mosh-devel at mit.edu
> http://mailman.mit.edu/mailman/listinfo/mosh-devel
>