scottlindner
join:2008-07-21
Colorado Springs, CO
| frequent disconnects
I have been struggling with frequent disconnects since switching to a new DSL account. Nothing physical changed, only my PPP username and password. I have contacted QWest several times without any success yet.
It happens as little as twice a week, and as often as four times a day, but generally about once per day. The disconnection lasts only a minute or so, and everything recovers automatically.
I have verified my PPP username and password with QWest several times. Each time they assure me it's correct.
I have the ActionTec M1000 DSL router. Yes, I know it sucks, but it's what I have. I have tried putting it into a transparent bridge with a Tomato based router, but I still get the disconnects regardless. When I used Tomato I received "LCP appears to be disconnected (pending: 5)." I am trying to find additional debug information from Tomato to see if that helps. Even if I could get a L2TP error code that would help a lot.
This morning I experienced another disconnect and I immediately checked the M1000 to see what it was doing. I noticed my CRC, FEC had no errors, no authorization errors, but the session time reset. I called QWest Tech Support and they registered no disconnection or authentication errors. What is interesting is they indicated I have been connected non stop for 252 days, but this DSL account is not even one month old yet.
Does anyone have any ideas how to debug this regular disconnection issue? QWest Tech Support doesn't appear to have the devices available each time I call them.
Scott
For additional information...
Here are my current stats from my M1000 WAN Status page:
WAN Status
Connection Status
Qwest Broadband: CONNECTED
Internet Service Provider (ISP) : CONNECTED
PPP Status
Status: Connected
User Name: xxx@qwest.net
LCP State: up
IPCP State: up
Authentication Failures: 0
Session Time: 1hr 15min 34sec
Packets Sent: 110556
Packets Received: 114353
Broadband Status
VPI: 0
VCI: 32
DSL Mode Setting: MMODE
DSL Negotiated Mode: G.DMT
Connection Status: Showtime
Speed (down/up): 7168 / 896 Kbps
Number of Retrains: 1
Number of Retrains Elapsed Time: 33 Hours 41 Mins 19 Secs
ATM QoS class: UBR
Near End CRC Errors (I/F) : 0/0
Far End CRC Errors (I/F) : 0/0
Near End CRC(Within last 30 mins)(I/F) : 0/0
Far End CRC(Within last 30 mins)(I/F) : 0/0
Near End RS FEC (I/F) : 0/0
Far End RS FEC (I/F) : 0/0
Near End FEC(Within last 30 mins)(I/F) : 0/0
Far End FEC(Within last 30 mins)(I/F) : 0/0
Discarded Packets(Within last 30 mins): 0
SNR Margin (Downstream/Upstream): 22/14
Attenuation (Downstream/Upstream): 6/8
I logged into the M1000 shell and pulled this from the log file:
(GMT-07:00)03:11:24 Sat Jul 25 2009 pppd[12009]: read conflict_time 86400
(GMT-07:00)03:11:24 Sat Jul 25 2009 pppd[12009]: read lease_file /var/tmp/landhcps0.leases
(GMT-07:00)03:11:24 Sat Jul 25 2009 pppd[12009]: local IP address xx.xxx.xxx.xxx
(GMT-07:00)03:11:24 Sat Jul 25 2009 pppd[12009]: remote IP address 67.42.184.210
(GMT-07:00)03:11:24 Sat Jul 25 2009 pppd[12009]: primary DNS address 205.171.3.25
(GMT-07:00)03:11:24 Sat Jul 25 2009 pppd[12009]: secondary DNS address 205.171.2.25
(GMT-07:00)03:11:24 Sat Jul 25 2009 logic: cmd1=/usr/bin/wrapntp clock.isc.org ntp0.nl.net ntp.actiontec.com time-nw.nist.gov pool.ntp.org &
(GMT-07:00)03:11:24 Sat Jul 25 2009 syslog: config.name_server[0]=205.171.3.25
(GMT-07:00)03:11:27 Sat Jul 25 2009 logic: got wan ip launch stunnel
(GMT-07:00)03:11:27 Sat Jul 25 2009 logic: launch stunnel 1, 0
(GMT-07:00)03:14:24 Sat Jul 25 2009 pppd[12009]: No response to 4 echo-requests
(GMT-07:00)03:14:24 Sat Jul 25 2009 pppd[12009]: Serial link appears to be disconnected.
(GMT-07:00)03:14:24 Sat Jul 25 2009 pppd[12009]: MRU: 1500
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Connection terminated.
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Connect time 3.1 minutes.
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Sent 39226 bytes, received 27639 bytes.
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Doing disconnect
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Exit.
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Plugin pppoe loaded.
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: PPPoE Plugin Initialized
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Plugin pppoe called.
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: pppd 2.4.1 started by admin, uid 0
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: setting line discipline hook
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Sending PADI
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: HOST_UNIQ successful match
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: HOST_UNIQ successful match
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Got connection: 1846
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Saved Session ID: 0
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Connecting PPPoE socket: 00:1e:4a:75:29:48 1846 br0 0x1000d538
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Using interface ppp0
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Connect: ppp0 -> br0
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: MRU: 1500
(GMT-07:00)03:14:36 Sat Jul 25 2009 syslog: DNS request timeout, server: 205.171.3.25.
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read start 192.168.1.2
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read end 192.168.1.254
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read interface br0
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read opt router 192.168.1.1
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read opt domain domain.actdsltmp
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read opt dns 192.168.1.1 205.171.3.25
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read opt subnet 255.255.255.0
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read opt lease 86400
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read conflict_time 86400
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: read lease_file /var/tmp/landhcps0.leases
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: local IP address xx.xxx.xxx.xxx
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: remote IP address 67.42.184.210
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: primary DNS address 205.171.3.25
(GMT-07:00)03:14:36 Sat Jul 25 2009 pppd[12403]: secondary DNS address 205.171.2.25
(GMT-07:00)03:14:36 Sat Jul 25 2009 syslog: config.name_server[0]=205.171.3.25
(GMT-07:00)03:14:38 Sat Jul 25 2009 logic: cmd1=/usr/bin/wrapntp clock.isc.org ntp0.nl.net ntp.actiontec.com time-nw.nist.gov pool.ntp.org &
(GMT-07:00)03:14:41 Sat Jul 25 2009 logic: got wan ip launch stunnel
(GMT-07:00)03:14:41 Sat Jul 25 2009 logic: launch stunnel 1, 0
|
|
rims
join:2000-10-22
Phoenix, AZ
| (GMT-07:00)03:14:24 Sat Jul 25 2009 pppd[12009]: No response to 4 echo-requests
(GMT-07:00)03:14:24 Sat Jul 25 2009 pppd[12009]: Serial link appears to be disconnected.
(GMT-07:00)03:14:24 Sat Jul 25 2009 pppd[12009]: MRU: 1500
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Connection terminated.
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Connect time 3.1 minutes.
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Sent 39226 bytes, received 27639 bytes.
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Doing disconnect
(GMT-07:00)03:14:30 Sat Jul 25 2009 pppd[12009]: Exit.
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: Plugin pppoe loaded.
(GMT-07:00)03:14:34 Sat Jul 25 2009 pppd[12403]: PPPoE Plugin Initialized
It looks like the your modem sent out some keep alives, got no response, so it initiated a new PPP session. There is nothing to indicate you recieved a termination packet from the ISP. This suggests that something other than PPP terminated the PPP session, since the purpose of PPP is to create, manage and tear down sessions. Usually some layer 1 event happens, so the LCP disappears and the PPP is unable to terminate gracefully. That is the purpose of the keep alives. Same thing with the tomato: "LCP appears to be disconnected": it sent out some packets and got no response from the ISP, so as far as it knows, layer 2 is gone.
I dont think you can assume that no equipment changed. Did they tell you you are on the same pair/same port, and just the ISP account was rebuilt? I would find out exactly what changed, so you can eventually work thru and swap out the changed parts (and by parts I mean physical and account configurations) |
|
scottlindner
join:2008-07-21
Colorado Springs, CO
| I know nothing physically changed from the demarcation point to anywhere within my home. I only have one pair on a Cat5 connected to Line 1 in the demarc so pairs could not have changed.
They could have changed something else within the telco but every call I have made indicated only my PPP credentials changed to the new account.
I understand what you are saying about the keep alive, but why would it not respond to a keep alive? I had run in the same physical configuration for a year non stop without an IP change or disconnection. Starting the day I switched to the new DSL account I started getting these disconnects. I can't imagine responding to a keep alive would be specific to an account, could it?
Since posting this I had another disconnection. I plan to call QWest tech support again and see if I can get someone other than Tier 1.
Scott |
|
rims
join:2000-10-22
Phoenix, AZ
| reply to scottlindner
The ISP might not respond to the keep alive because it never gets the keep alive
I've read other posts on non qwest forums where the modem remains trained (bridge mode) but the users router keeps creating new PPP sessions..... turns out it was the modem causing the problem, their was no data thruput.
So since your modem is the same, perhaps the port you are training to is different now, and for some reason you remain synched up, but their is no data thru put, your keep alives die at the dslam.
The way I understand it, if the ISP was terminating the PPP session for anyreason, it would send a terminating packet, so its hard to see this as a PPP issue, and easy to see this "a layer underneath PPP prevented the PPP communication" issue.
Nevertheless, if they are certain that the only thing that changed was the ISP account, then they need to tear that down and build another one (and make sure your old account isnt still in their system) . I would ask them nicely,,,, "how do you know the port is the same" |
|
scottlindner
join:2008-07-21
Colorado Springs, CO | What do you mean by port? |
|
rims
join:2000-10-22
Phoenix, AZ | reply to scottlindner
The modem at the dslam that syncs with your modem |
|
scottlindner
join:2008-07-21
Colorado Springs, CO | I will ask about that. I'm typically pretty knowledgable on networking stuff, but how DSL works isn't something I know much about yet. Thanks for helping!
I'll post back when I get more details from QWest tomorrow.
Scott |
|
scottlindner
join:2008-07-21
Colorado Springs, CO
| QWest tech support ran a bunch of tests today that spanned nearly an hour. They couldn't find anything wrong so they are sending me a new DSL router. I doubt that's the problem but it isn't a bad thing to try either so I'll give it a go. I will post back the results once I have the new DSL router in place.
Cheers,
Scott |
|
TalkToQwest
join:2009-03-09
| Hey Scott, this is Brian with Qwest. I ran across your posts here, and I'm sorry that you've had so much trouble with your DSL being intermittent, but I am glad to hear that we're sending you a new modem. Hopefully this will fix this issue.
However, if it doesn't, and you need more help, send me an e-mail to TalkToUs@Qwest.com. I will do more to help resolve this for you, if the new modem doesn't work.
Thanks Scott,
B
Consumer Affairs Manager
Qwest Communications
www.socialmedia.qwest.com |
|
scottlindner
join:2008-07-21
Colorado Springs, CO
| Thanks Brian!
I'll be sure to contact you regardless. If it works I'll let you know it solved the problem, and if it doesn't I'll contact you to see if there is another solution to this.
I'll also post back to the forum as well.
Scott |
|
scottlindner
join:2008-07-21
Colorado Springs, CO
| This is crazy. Since the test I have been running non stop without a disconnect. Is it possible the test somehow flushed the settings for my account on the DSLAM?
I have the new DSL router from QWest but I'm going to let this run for a while longer to see if it continues uninterrupted.
Scott |
|
rims
join:2000-10-22
Phoenix, AZ | reply to scottlindner
If they spent an hour they probably rebooted the modem on their end |
|
scottlindner
join:2008-07-21
Colorado Springs, CO | I just hit one week of continuous uptime with my Tomato router handling the PPP. Sweet. This problem is solved! |
|
scottlindner
join:2008-07-21
Colorado Springs, CO
1 edit | It turns out the problem is not solved. I don't know what happened but after a full 10 days of up time I'm back to the same problem again. I swear someone at the local QWest office is dinking around with things and doesn't understand the side affects. It appears the cycle is now around 72 hours per disconnect. Since it was always happening in mid day during primary work hours when we need our connection to not disconnect the most (our work applications crash when this happens) I did a manual disconnect in the middle of the night. Now three days later it disconnected in the middle of the night. If I can continually persuade the DSL disconnect to happen in the middle of the night I'll write a little script on my router to drop the connection at 3am every night and I'll be happy enough.
However, QWest really needs to figure this out. I have seen this posted on several forums. I had contacted Brian but he hasn't responded.
I doubt anyone can help other than someone that at QWest. I have confirmed repeatedly this is a problem within QWest and not my environment at home.
I'm updating this post since this morning to clear up some bad typos and language mistakes. Since then my DSL has dropped 4 more times in a short burst a couple of hours ago. I will try QWest once more, and if it doesn't work out I am dropping them because it is always happening during working hours and it is killing our work applications. If we cannot work, it's worth paying the $10/mo more for Cable if Cable is reliable.
Scott |
|
EJG
@qwest.net
| Scott:
Same here. Went for about 6 days without a disconnect and have now been getting disconnects for the last 2 evenings. These disconnect problems are documented all over the Internet and Qwest forums! Keep up the good work and post all over the place to get some attention! |
|
scottlindner
join:2008-07-21
Colorado Springs, CO
| We should get together to see if it is happening at the same times for all of us. I wonder if it's regional, per DSLAM, per CO, per account?
Where have you seen the most chatting about this issue so I can head over there to compare notes?
Cheers,
Scott |
|
TalkToQwest
join:2009-03-09
| Hey Scott, this is Brian with Qwest again. Dang, I'm sorry this issue is still ongoing for you. Can I help with anything? I have several DSL tech support specialists that are great with assisting our customer's with complex issues. Just let me know. E-mail at TalkToUs@Qwest.com if I can help.
Thanks Scott,
B
Consumer Affairs Manager
Qwest Communications
www.socialmedia.qwest.com |
|
kcimpulse
join:2009-08-23
Denver, CO
| reply to scottlindner
I am so glad i found others like me with this same issue. And i mean the exact same issue you described it perfectly Scott. I have tried so many things, i have even had the DSL tech come out and re patch me into a different port on the switch and have the account re trained with no luck. The best i can get is after they flush my account it seemed to work for almost a week and then the issue started over again. I Have tired it all ways with 2 different modems the 701G and the M1000 both stand alone and both transparently bridged with two both DD-WRT and Tomato. Have you guys been heard of anything else , currently i am tired of calling the tech support and am seriously considering Comcast or even Whispertel. Dial up is almost better than the constant disconnects i have been receiving. - Casey |
|
scottlindner
join:2008-07-21
Colorado Springs, CO
| Yesterday I worked with the technician Brian at QWest suggested. The tech's name is Doug and he's part of the escalation team. Doug did some tests and checked things out. He read this thread to get up to speed. He said he'd need to work with tier 3 to figure this one out because everything looked right, but he seemed to have some theories to chase down.
This may be unrelated but I had another QWest oddity yesterday. I just learned that I have two DSL accounts with QWest. Long story why I'm just finding out about it now so I won't go into that. I will take care of that on Monday to shut down the second account. The reason I mention this is that I am using my second account (the one I shouldn't have) for my PPPoE. I wonder if this could be related? Could the port on the DSLAM be specific to the account? If so, that could be the problem. I will be calling QWest Monday morning to shut down the second account and to get the PPPoE credentials for the account I should be using.
I will be sure to continue posting updates to this thread until I get this resolved. I know there are others out there having the same issue and I want to ensure there is a good write up to be found.
Scott |
|
astadtler
@qwest.net
| reply to scottlindner
I'm just going to suggest that you double check phones for filters. Only because I had this problem with someone stealthily connecting a fax machine without filters without my knowing speeds still worked fine but when we would get a phone call we would get a disconnect. I'm on an m1000 also. |
|
