Long timeouts for stale PPPoE connections

[#1]

Do people reckon we're safe to consider this fixed then (or at least significantly better compared to before)?

[#2]

I'll try it again when I get home later

[#3]

Just over 2 minutes when I just tried it again. Would still like it to be less 

[#4]

Well two minutes was what we were hoping for so I think we can assume the work to have been successful
I'll question the possibility of making the timeout lower but I can't guarantee anything.

[#5]

Would there be a rationale for not reducing it further?

[#6]

Bob's chasing this up, we'd have to apply this across the whole product set and not just FTTC based customers.
We'll have some answers soon if we able to bring this any lower.

[#7]

Quote from: Bob

Well two minutes was what we were hoping for so I think we can assume the work to have been successful I'll question the possibility of making the timeout lower but I can't guarantee anything.

Transpires that 30 seconds is the lowest we can go. It's a limit set by the vendor of the kit we're using.

[#8]

If I'm permitted to quote from BT SIN 495:

Quote

Unlike other WBC access technologies, WBC FTTC accesses use the PPP (“Point-to-Point Protocol”) session establishment to inform the BT Wholesale BRAS of the Openreach line rate. It is therefore essential that the PPP session is re-started every time the VDSL line retrains. To ensure this, the PPP / L2TP timeout values must be set to less than 20 seconds.

[#9]

Doesn't detract from the fact that the kit we're using can't do that.

Quote

Options seconds—Keepalive timeout period, in the range 30–64800 seconds for high-density mode, 1–64800 seconds for POS uplink interfaces in low-density mode, or 10–64800 seconds for all other HDLC interfaces in low-density mode; default value is 30 seconds

[#10]

Can't argue with that.

[#11]

Wow it is amazing how dumbed-down (no offence meant!) the configuration for the Juniper kit is.  They don't allow you to tweak the number of LCP retries either it seems (must be hard coded to 3).  Shame.  I wonder how many other ISP's can't comply due to such issues?

[#12]

Quote from: Bob

Doesn't detract from the fact that the kit we're using can't do that.

So it's not compliant then?  When is it due to be replaced with something that meets the BT spec?

[#13]

Not due to be replaced any time soon to my knowledge. We've a fair few of the things in the network and I've reason to believe that they're capable of scaling fairly well.

[#14]

I thought this problem had been fixed  or the time had been reduced, ? below is the result from being kicked from the Gateway i had been connected to for reasons unknown,  8minutes !!!!! .
But i do wish Plusnet systems that auto do this would do it a different time of the day, infact i wish  it would stop swapping me from gateway to gateway, i can do that myself, if and when i feel i need to BTW i checked the zen status for BT engineering works, nothing for my area code

Oct 07 00:50:04  daemon  pppd[16222]: No response to 6 echo-requests
  Oct 07 00:50:04  daemon  pppd[16222]: Serial link appears to be disconnected.
  Oct 07 00:50:04  daemon  pppd[16222]: Clear IP addresses.  Connection DOWN. 
  Oct 07 00:50:04  daemon  pppd[16222]: Clear IP addresses. 
  Oct 07 00:50:04  daemon  pppd[16222]: Couldn't increase MTU to 1500.
  Oct 07 00:50:04  daemon  pppd[16222]: Couldn't increase MRU to 1500
  Oct 07 00:50:05  user  syslog: begin: interface: ppp_ewan_1 go to down 
  Oct 07 00:50:05  user  syslog: end: interface: ppp_ewan_1 go to down 
  Oct 07 00:50:05  daemon  pppd[16222]: PPPoE: Terminating on signal 15.
  Oct 07 00:50:10  daemon  pppd[16222]: Connection terminated.
  Oct 07 00:50:10  daemon  pppd[16222]: Connect time 7924.3 minutes.
  Oct 07 00:50:10  daemon  pppd[16222]: Sent 6793888165 bytes, received 25838345197 bytes.
  Oct 07 00:50:11  user  kernel: dev_shutdown, dec ppp device refcnt, dev->refcnt=4
  Oct 07 00:50:11  user  kernel: unregister_netdevice: waiting for ppp_ewan_1 to become free. Usage count = -1
  Oct 07 00:50:11  user  kernel: dev->name = ppp_ewan_1, dev->refcnt=-1
  Oct 07 00:50:11  user  kernel: after reset to 0, dev->refcnt=0
  Oct 07 00:50:11  daemon  pppd[16222]: Doing disconnect
  Oct 07 00:50:14  daemon  pppd[16222]: Exit.
  Oct 07 00:50:15  daemon  pppd[16762]: pppd 2.4.1 started by admin, uid 0
  Oct 07 00:50:15  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:50:38  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:50:38  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:50:38  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:50:41  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:51:04  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:51:04  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:51:04  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:51:07  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:51:30  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:51:30  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:51:30  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:51:33  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:51:56  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:51:56  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:51:56  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:51:59  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:52:22  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:52:22  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:52:22  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:52:25  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:52:48  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:52:48  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:52:48  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:52:51  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:53:14  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:53:14  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:53:14  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:53:17  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:53:40  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:53:40  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:53:40  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:53:43  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:54:06  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:54:06  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:54:06  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:54:09  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:54:32  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:54:32  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:54:32  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:54:35  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:54:58  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:54:58  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:54:58  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:55:01  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:55:24  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:55:24  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:55:24  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:55:27  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:55:50  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:55:50  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:55:50  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:55:53  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:56:16  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:56:16  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:56:16  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:56:19  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:56:42  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:56:42  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:56:42  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:56:45  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:57:08  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:57:08  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:57:08  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:57:11  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:57:34  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:57:34  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:57:34  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:57:37  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:58:01  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:58:01  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:58:01  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:58:04  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:58:27  daemon  pppd[16762]: Couldn't get channel number: Transport endpoint is not connected
  Oct 07 00:58:27  daemon  pppd[16762]: PPP: PPP Server No Response !!! 
  Oct 07 00:58:27  daemon  pppd[16762]: Doing disconnect
  Oct 07 00:58:30  daemon  pppd[16762]: PPP: Start to connect ... 
  Oct 07 00:58:44  daemon  pppd[16762]: PPP server detected. 
  Oct 07 00:58:44  daemon  pppd[16762]: PPP session established. 
  Oct 07 00:58:44  daemon  pppd[16762]: Using interface pppewan_1
  Oct 07 00:58:44  daemon  pppd[16762]: Connect: ppp_ewan_1 <--> eth0
  Oct 07 00:58:44  daemon  pppd[16762]: Couldn't increase MTU to 1500.
  Oct 07 00:58:44  daemon  pppd[16762]: Couldn't increase MRU to 1500
  Oct 07 00:58:44  daemon  pppd[16762]: Couldn't increase MRU to 1500
  Oct 07 00:58:44  daemon  pppd[16762]: PPP LCP UP. 
  Oct 07 00:58:44  daemon  pppd[16762]: Couldn't increase MTU to 1500.
  Oct 07 00:58:44  daemon  pppd[16762]: Couldn't increase MRU to 1500
  Oct 07 00:58:44  daemon  pppd[16762]: Couldn't increase MRU to 1500
  Oct 07 00:58:44  daemon  pppd[16762]: PPP LCP UP. 
  Oct 07 00:58:45  daemon  pppd[16762]: local  IP address 212.159.XXX.XXX
  Oct 07 00:58:45  daemon  pppd[16762]: remote IP address 195.166.128.192
  Oct 07 00:58:45  daemon  pppd[16762]: primary   DNS address 212.159.6.9
  Oct 07 00:58:45  daemon  pppd[16762]: secondary DNS address 212.159.6.10
  Oct 07 00:58:45  daemon  pppd[16762]: Received valid IP address from server.  Connection UP. 
  Oct 07 00:58:46  user  syslog: begin: interface: ppp_ewan_1 go to up 
  Oct 07 00:58:49  user  syslog: end: interface: ppp_ewan_1 go to up 

[#15]

I don't think it was us that disconnected you.   We have a "Termination by server side (NAS) Modem" event on our side, and no profile change or forced disconnect logged on our side.