hard hang when using l2tpd/pppd with IPv6
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux (Ubuntu) |
Expired
|
Medium
|
Unassigned |
Bug Description
I'm not completely sure how to summarise this bug, but it's obviously potentially fatal and related to the kernel components of either xl2tpd, pppd or IPv6.
I have an IPv6 connection using xl2tpd to make a ppp connection and then route IPv6 over that... this works fine, but if I stress it using heavy traffic such as bittorrent, the entire system will lock up at the kernel level after a time, between 10 minutes and 1 hour approximately.
This is using Trusty with the 3.12 kernel as of 17/Nov 2013... I'm really not sure how to go about debugging this, the only clue I can find is this:-
Nov 16 23:32:17 trinity xl2tpd[2849]: check_control: Received out of order control packet on tunnel 35506 (got 4, expected 5)
Nov 16 23:32:17 trinity xl2tpd[2849]: handle_packet: bad control packet!
Nov 16 23:32:18 trinity xl2tpd[2849]: check_control: Received out of order control packet on tunnel 35506 (got 4, expected 5)
Nov 16 23:32:18 trinity xl2tpd[2849]: handle_packet: bad control packet!
Nov 16 23:32:19 trinity xl2tpd[2849]: check_control: Received out of order control packet on tunnel 35506 (got 4, expected 5)
Nov 16 23:32:19 trinity xl2tpd[2849]: handle_packet: bad control packet!
Nov 16 23:32:20 trinity xl2tpd[2849]: check_control: Received out of order control packet on tunnel 35506 (got 4, expected 5)
Nov 16 23:32:20 trinity xl2tpd[2849]: handle_packet: bad control packet!
Nov 16 23:32:21 trinity xl2tpd[2849]: control_finish: Connection closed to 85.119.84.18, port 1701 (Timeout), Local: 25286, Remote: 35506
Nov 16 23:32:21 trinity xl2tpd[2849]: Terminating pppd: sending TERM signal to pid 7386
The connection is then re-established..
Is this a new issue in Trusty? Did you see this in prior releases?