 Fraoch
join:2003-08-01
Cambridge, ON
TP-Link EAP225
Grandstream HT502
1 recommendation |
Fraoch
Member
2017-Nov-2 11:09 am
Strange network issue I've never seen before on these forumsI'm coming here before I contact my ISP (Acanac) because I've never seen my issue discussed here before and it's hard to pin down exactly what's wrong. I'm on Acanac 60/10 on Rogers cable with a Thomson DCM476. I first saw this at least a year ago. It seems to be very very high ping (3000 ms+) to all sites...then suddenly ping is fine again, but connecting to sites takes 30 seconds+ or times out. It seems to be DNS...but it's not! Just when I think it is and I'm stalling reaching DNS servers, DNS pings leap back into life but still takes 30 seconds to a minute to load various sites. Very, very strange. It's like once a connection is initiated, it's fine, but initiating the connection takes a long time. It's almost like it's cached as well, sites I've already connected to load fine for a while, but sites I have not connected to recently take minutes. Needless to say, this wreaks havoc on my Zazeen IPTV. Each channel takes 30 seconds to a minute to appear and doesn't appear at all 50% of the time, with either a blank screen or a "cannot reach host server" error. Channels I have previously viewed come up fine, but "new" channels stall. This issue has followed me from ISPs too (indicating it's a Rogers problem - joy!!). I first noticed this when I was on TekSavvy. Now I'm on Acanac and this morning is the worst it has ever been. All the tools I'm using don't really show the problem. ping, smokeping, traceroute, I can't find the smoking gun. $ ping www.google.ca -c 15
PING www.google.ca (172.217.2.99) 56(84) bytes of data.
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=1 ttl=56 time=22.7 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=2 ttl=56 time=16.5 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=3 ttl=56 time=29.3 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=4 ttl=56 time=21.0 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=5 ttl=56 time=18.7 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=6 ttl=56 time=26.5 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=7 ttl=56 time=14.9 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=8 ttl=56 time=14.6 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=9 ttl=56 time=20.5 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=10 ttl=56 time=21.7 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=11 ttl=56 time=18.2 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=12 ttl=56 time=26.0 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=13 ttl=56 time=26.1 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=14 ttl=56 time=15.3 ms
64 bytes from yyz10s05-in-f3.1e100.net (172.217.2.99): icmp_seq=15 ttl=56 time=16.8 ms
--- www.google.ca ping statistics ---
15 packets transmitted, 15 received, 0% packet loss, time 14020ms
rtt min/avg/max/mdev = 14.682/20.645/29.361/4.573 ms
So ping looks fine.  Michigan smokeping Smokeping doesn't look great, but doesn't look catastrophically bad, and doesn't convey how bad the connection "feels". $ traceroute www.google.ca
traceroute to www.google.ca (172.217.2.99), 30 hops max, 60 byte packets
1 192.168.1.1 (192.168.1.1) 0.810 ms 1.097 ms 0.966 ms
2 * * *
3 206.80.241.124 (206.80.241.124) 15.938 ms 17.127 ms *
4 206-80-244-129.cpe.distributel.net (206.80.244.129) 21.927 ms 20.093 ms 58.241 ms
5 206.80.255.56 (206.80.255.56) 21.336 ms 21.328 ms 21.329 ms
6 74.125.147.64 (74.125.147.64) 22.540 ms * 17.832 ms
7 108.170.250.225 (108.170.250.225) 16.400 ms 17.642 ms 108.170.250.241 (108.170.250.241) 18.051 ms
8 108.170.226.213 (108.170.226.213) 16.508 ms 17.700 ms *
9 yyz10s05-in-f3.1e100.net (172.217.2.99) 19.174 ms 19.103 ms 46.692 ms
(looks fine, but just initiating the test took a minute or two until the first line 'traceroute to...' appeared) Signal levels: 
Seems OK from what I can tell. Now "namebench", a DNS benchmarking program I'm running to see if it is a DNS problem, has just come up with something: How odd! Only 7.1 percent of name servers were pingable. Trying again with 6 threads
and Google Public DNS-2 [8.8.4.4] failed test #1/5: TestARootServerResponse: a.root-servers.net.: Timeout
which should not happen. It's like "most" of the Internet is unreachable for me? Another clue: when I reset my cable modem the issue goes away for several minutes, but it returns eventually. I know from experience that once I see this issue it persists for days, then goes away for months but reappears. I don't even know what to tell my ISP to get them to open a ticket with Rogers to investigate - it seems to me the problem lies with Rogers - some sort of signal or maybe router issue, it's followed me between ISPs and Rogers is the only common denominator. Maybe I'll play dumb and say the Internet doesn't work. :p Node congestion? Wouldn't that result in consistently high pings though? |
|
| |
what kind of modem? i'd speculate Puma6 chip issue |
|
1 recommendation |
to Fraoch
First step is to bypass your router. It's the first troubleshooting step that you absolutely must do. Let your computer pull DNS addresses automatically, repeat the tests. |
|
Fraoch
join:2003-08-01
Cambridge, ON |
to prairiesky
No, it's an old faithful Thomson DCM476 as I indicated at the top of the post. |
|
| Fraoch |
to MacGyver
Good point, I don't do this very often and forgot. I plugged the computer directly into the modem and had high hopes, but it seems the problem remains - only namebench is showing something concrete again: OpenDNS [208.67.220.220] failed test #1/5: TestARootServerResponse: a.root-servers.net.: Timeout
Again, something which shouldn't happen. And perhaps packets are getting mangled? Unusual error with A:a.root-servers.net. on Telefonica Movistar ES [194.179.001.100]: SyntaxError
I've been doing updates on this computer and have noticed that a network resource that doesn't work at first will often work on a retry. More weirdness, and I don't know what to tell my ISP. |
|
| |
to Fraoch
is it occurring on multiple devices? I'm unfamiliar with zazeen TV, so i'm not sure if that's a separate box or just an app/website. |
|
Fraoch
join:2003-08-01
Cambridge, ON SmartRG SR808ac
TP-Link EAP225
Grandstream HT502
|
Fraoch
Member
2017-Nov-2 2:05 pm
Looks like the modem rebooted and didn't come back up but no idea when: Time Not Established Warning (5) DHCP WARNING - Non-critical field invalid in response ;CM-MAC...
Time Not Established Critical (3) DHCP FAILED - Request sent, No response;CM-MAC=fc:94:e3:6c:a0...
Time Not Established Critical (3) No Ranging Response received - T3 time-out;CM-MAC=fc:94:e3:6c...
Mon Oct 02 20:41:04 2017 Critical (3) Started Unicast Maintenance Ranging - No Response received - ...
Mon Sep 25 15:01:16 2017 Critical (3) SYNC Timing Synchronization failure - Failed to acquire QAM/Q...
Sun Sep 24 23:31:58 2017 Critical (3) Started Unicast Maintenance Ranging - No Response received - ...
Fri Sep 22 06:58:23 2017 Critical (3) Started Unicast Maintenance Ranging - No Response received - ...
Mon Sep 18 20:34:37 2017 Critical (3) Started Unicast Maintenance Ranging - No Response received - ...
Fri Sep 15 12:28:55 2017 Critical (3) Started Unicast Maintenance Ranging - No Response received - ...
Fri Sep 08 17:00:16 2017 Critical (3) Started Unicast Maintenance Ranging - No Response received - ...
There's indications from various network services that send me e-mails that it was down completely at 6:19 AM, up again at 6:53 AM and down again at 7 AM today. But other network services never gave a down indication... |
|
Fraoch
1 edit |
to prairiesky
said by prairiesky:is it occurring on multiple devices? I'm unfamiliar with zazeen TV, so i'm not sure if that's a separate box or just an app/website. Everything connected to the network. All computers, all services. It seems to be random whether a site/service is affected - some are, some aren't. It's like there are huge IP ranges that are affected. Zazeen's service acts like a web browser. Each "channel" is like a website with a live stream. Currently some work and some don't. All respond sluggishly and some don't come in at all, just like websites in any connected computer - some respond, some don't, some do after a few retries. |
|
1 recommendation |
to Fraoch
I noticed connectivity issues as well this morning (Acanac Cable 100/10), and didn't really have time to look into it before going to work. I was roughly around 6:15-6:30am that I noticed it.
I was going to see if it was resolved when I got home tonight. If it's still an issue I was going to call into acanac. |
|
Fraoch
join:2003-08-01
Cambridge, ON SmartRG SR808ac
TP-Link EAP225
Grandstream HT502
1 edit |
Fraoch
Member
2017-Nov-2 2:26 pm
said by kenwuwu:I noticed connectivity issues as well this morning (Acanac Cable 100/10), and didn't really have time to look into it before going to work. I was roughly around 6:15-6:30am that I noticed it.
I was going to see if it was resolved when I got home tonight. If it's still an issue I was going to call into acanac. Aha! Now that makes sense. Some sort of failing Rogers router or routing issues. Failing in a very odd way. The problem is still ongoing. Can't connect to acanac.com at the moment.  |
|
 Fergless
Premium Member
join:2008-04-19
Toronto, ON
2 recommendations |
Fergless
Premium Member
2017-Nov-2 2:32 pm
|
|
Fraoch
join:2003-08-01
Cambridge, ON SmartRG SR808ac
TP-Link EAP225
Grandstream HT502
|
Fraoch
Member
2017-Nov-2 2:35 pm
I see reports from Kitchener - close enough to me that that's probably where the CMTS is.  Thanks. |
|
1 recommendation |
to Fraoch
Can confirm that certain cable customers in Rogers and Cogeco areas are experiencing intermittent access in certain areas/websites. We have escalated to and are working with the carriers. You can monitor Network Status here: » www.acanac.com/contact/n ··· -status/ |
|
Fraoch
join:2003-08-01
Cambridge, ON |
Fraoch
Member
2017-Nov-2 3:05 pm
Thank you! |
|
| |
I experienced the same issues. Started very late last night, got intermittent. Can't use the Internet reliably. Those Rogers systemic issues are really annoying. I wish I lived in a DSL-covered building. |
|
Fraoch
join:2003-08-01
Cambridge, ON SmartRG SR808ac
TP-Link EAP225
Grandstream HT502
|
Fraoch
Member
2017-Nov-2 3:38 pm
said by FiReSTaRT:I experienced the same issues. Started very late last night, got intermittent. Can't use the Internet reliably. Those Rogers systemic issues are really annoying. I wish I lived in a DSL-covered building. Where are you located, approximately? |
|
Fergless
Premium Member
join:2008-04-19
Toronto, ON |
Fergless
Premium Member
2017-Nov-2 3:47 pm
I'm in Toronto. Started really early this morning as well. Ended up connecting my STB and VoIP to my 10/1 DSL Line. Cable isn't to bad right now. Hang in there guys.  |
|
| |
to Fraoch
said by Fraoch:Where are you located, approximately? Etobicoke. The issue seems to be pretty well distributed in Robbers land based on the outages report both for Distributel and Rogers. IME, cable is more reliable overall but when a systemic issues rear their ugly heads, it's brutal. |
|
Fraoch
join:2003-08-01
Cambridge, ON SmartRG SR808ac
TP-Link EAP225
Grandstream HT502
1 recommendation |
Fraoch
Member
2017-Nov-2 5:31 pm
said by FiReSTaRT:IME, cable is more reliable overall but when a systemic issues rear their ugly heads, it's brutal. Yeah, it's still going on. Just really random strangeness that's not entirely one thing and not entirely another. The connection is very slow but not to certain websites and certain services, until it is, then it's not. It's like routes to certain IPs go down, then up, then others go down, then up. Whole big blocks at a time. Routing? |
|
| |
to Fraoch
It just cleared up for me (Distributel in Mississauga). I was seeing a >50% fail rate for DNS packets and a high rate of three way handshake failures. It may have been routing but using MTR didn't show any weird routes popping in and out. |
|
| |
to Fraoch
In London, Ontario on Distributel Cable, having same issues as well. |
|
| |
Anon46875 to Fraoch
Anon
2017-Nov-2 6:17 pm
to Fraoch
Yeah. Itâs defintely weird. We have a Distributel Cable connection exclusively for our Zazeen TV service, and a Start.ca connection for our primary connection. We thought our Zazeen TV service was down since we just kept getting a âHost not reachableâ error. Couldnât get them to tune in at all.
Glad I decided to connect to the base station connected to Distributel before I started hammering Zazeen. Our Start DSL connection and Start rCable connection is rock solid as usual though.
Looks like the Distributel problem has been going on for the good majority of the past 16 hours or so, according to Twitter |
|
| |
to Fraoch
Aurora here..been bad since midnight. Came here to see if anyone else affected |
|
| |
Anon5891e
Anon
2017-Nov-2 7:01 pm
I'm also having similar issues in kitchener, I can connect to some IPs just fine, but others, nothing. Running TOR browser lets me access everything without issues (really does seem like a routing issue) TOR isn't a real solution but at least it's something. Been like this for 20+ hours now. |
|
| Anon5891e |
Anon5891e
Anon
2017-Nov-2 8:46 pm
Just checking in again, all of a sudden everything started working again (I am Anon5891e) on distributel (kitchener) hopefully they have resolved this issue. |
|
Fraoch
join:2003-08-01
Cambridge, ON SmartRG SR808ac
TP-Link EAP225
Grandstream HT502
1 recommendation |
to Malvane
Yes, I think things improved around 7:45 or so. Suddenly Zazeen TV started working well again, and it's usually pretty sensitive. I haven't done all the tests but this is working and my wife is happy watching HGTV again. |
|
| |
to Fraoch
Seems to have corrected itself about 20-30 minutes after I emailed support last night (one hell of a coincidence). I think it was around 6:00pm that it started to work for me (in Ottawa). |
|
2 recommendations |
to Fraoch
We are getting word from our Network team that things should be back to normal. We recommend that affected customers power cycle and call in for technical support if they are still experiencing issues. We apologize for the inconvenience- we know how central Internet is to people day-to-day and we work hard to minimize any disruptions. |
|
Fraoch
join:2003-08-01
Cambridge, ON SmartRG SR808ac
TP-Link EAP225
Grandstream HT502
|
Fraoch
Member
2017-Nov-3 10:21 am
said by Acanac:We are getting word from our Network team that things should be back to normal. We recommend that affected customers power cycle and call in for technical support if they are still experiencing issues. We apologize for the inconvenience- we know how central Internet is to people day-to-day and we work hard to minimize any disruptions. Thank you! |
|
| |
to Fraoch
said by Fraoch:It seems to be very very high ping (3000 ms+) to all sites...then suddenly ping is fine again, but connecting to sites takes 30 seconds+ or times out. It seems to be DNS...but it's not! Just when I think it is and I'm stalling reaching DNS servers, DNS pings leap back into life but still takes 30 seconds to a minute to load various sites. Very, very strange. said by Fraoch:All the tools I'm using don't really show the problem. ping, smokeping, traceroute, I can't find the smoking gun. At that point I'm of the mindset "packetsniffer" to start looking at the packet level what's going on. I presume you're running continuous pings to said website(s) and DNS server(s) you're using. I presume namebench also has some function to periodically test resolution of a particular URL against a (list of) DNS servers? If it can / is doing that, that at least confirms name resolution is working. That at least would rule out DNS as a problem. Does the namebench documentation have anything on the error messages you posted? Or any clue what they mean in the grand scheme of things? Regarding your posted modem logs, my only thought is whether those CRITICAL (3) messages line up with your expterienced probelms? T3 timeout basically means "problem between your modem and the CMTS" and no traffic would pass. My initial thoughts anyways. Regards |
|
