As per above my issue appears to have resolved itself, but for the record, when I did have the issue pre 7pm Saturday - very similar, ie:
- multiple browsers no difference - disable adblocker no difference - using 8.8.8.8 DNS no difference - browsing over VPN and the problem wasnt observed, disconnect VPN problem returns
Again, as mentioned, since Saturday night this issue appears resolved for me
Talkiet: HI guys - Sorry I haven't had a chance to respond to this specific issue yet... A few questions before I look at escalating internally... Apologies if these have been answered or implied in answers so far...
- What plan are you on? (Note this shouldn't make a difference because we don't do any traffic shaping on any plans - but if you are on a metered plan then I am going to ask about your up and download usage to see if it's higher than expected) - Do you use adblockers on all your machines? Can you try using an alternate default browser (I imagine you don't use IE, so could you try using that with no plugins enabled) - Have you detected any IP level dropouts? I see you picked up on pinging 219.88.188.4 (one of our Auckland based Google caches). Have you ever detected connectivity issues to that? - You mention that this can occur at any time of day - have you noticed that it's more frequent or more severe at peak times? This is a really important issue as the answer directs us down very different troubleshooting paths - Do you have a static IP address? - Do you host any services on your home connection (webserver, FTP, game servers, RDP, etc)? - When did the issue first become apparent? Was it sudden onset or did it seem to get worse over days or weeks? - From your VDSL connect speeds, clearly you have a decent line - but have you noticed any disconnects? What is a typical DSL uptime?
Cheers - N
Plan is VDSL, 150GB cap. Adblock and browsers (and OS) makes no difference. No issues pinging any IP. I think it can be worse do during peak times but it is marginal. No hosting of services. Sudden onset. No disconnects, only time uptime was reset was when I manually rebooted the router testing, its been very stable.
I've also been using VPN to get around the issue, connecting to machine in Orcon DC, very fast speeds browsing via that method.
When talking with L2 support at Spark we tested DNS performance, using 8.8.8.8 made it worse for me, local Spark DNS servers seemed to responding fine.
I have been running resource monitor / networking looking to see if it reports lost packets or anything interesting, nothing of note to report there however.
Update. - this has now gone from unusable to intermittent. I am now getting 10-15 min periods where things seem resolved. Then it goes back to unusable again.
It's a strange one
Hi guys, just wanted to say a +1 on the same issue, I'm based in Torbay on the North Shore, running VDSL with a wired connection. Much the same symptoms except I'm also getting intermittent issues with local traffic (eg geekzone.co.nz failing sometimes). The woes started about 5-7 days ago, before that I've been running VDSL at this address with this hardware/software setup for over 2 years without a problem. Tried most of the other suggestions listed above (changed DNS etc) but no improvements. I've got multiple devices both wired and wireless experiencing the issue, from my Win8 PC to an iPad. To try and isolate the issue I've gone out and purchased a new VDSL modem/router (TPLINK TD-W9980) and while my WAN connection speed seems slightly improved, I'm still getting intermittent page loading issues. Something has most definitely changed quite recently and it's interesting a whole lot of North Shore people all saying the same thing.......
Hi, Reading through the last 4 pages, this seems to be isolated to the North Shore. Myself (Greenhithe) and my colleague (Bayview) have both had issues since Thursday afternoon. His problem he feels was resolved by Saturday, but mine has continued to be an issue. Like others in the thread, International appears to be the main problem. CDN's that use Akamai (Microsoft), Imgur (AWS) and GFYcat (uses AWS S3) that load balance the traffic across various geolocation's, but in most cases there is not a local peer in NZ, so the nearest one will be Sydney, but a lot of stuff is Singapore This could be something really gnarly like a AS peer or BGP route.
However some testing brought some interesting results. My NZ norm is 20 down 7 up 5ms latency 1) Download/upload tests to various Australia locations was good 2) Download tests to the US were poor. Ping tests were okay https://wondernetwork.com/pings suggests auckland to la to be 126ms.
Cruzio in Santa Cruz
Time Warner Cable in LA
This is a Traceroute 4 8 ms 10 ms 8 ms ae8-10.akbr6.global-gateway.net.nz [122.56.116.5 ] 5 * * * Request timed out. 6 7 ms 8 ms 6 ms ae1-10.tkbr12.global-gateway.net.nz [202.50.232. 37] 7 131 ms 262 ms 196 ms xe8-0-2-0.lebr7.global-gateway.net.nz [210.55.20 2.194] 8 136 ms 176 ms 135 ms ae3-10.sjbr3.global-gateway.net.nz [122.56.127.2 5] 9 163 ms 163 ms 163 ms ae0.pabr5.global-gateway.net.nz [203.96.120.74]
10 149 ms 170 ms 204 ms ix-8-2-5-0.tcore1.PDI-Palo-Alto.as6453.net [66.1 98.127.49] 11 210 ms 203 ms 204 ms if-2-2.tcore2.PDI-Palo-Alto.as6453.net [66.198.1 27.2] 12 219 ms 174 ms 175 ms if-5-2.tcore2.SQN-San-Jose.as6453.net [64.86.21. 1] 13 204 ms 205 ms 202 ms 64.86.21.62 14 142 ms 142 ms 142 ms et-5-3-0.lax21.ip4.gtt.net [89.149.129.206] 15 141 ms 177 ms 141 ms gtt-gw-cr1.lax.ipv4.hugeserver.com [173.205.61.2 2] 16 142 ms 204 ms 142 ms cs1.lax3.hugeserver.com [199.244.116.2] 17 142 ms 141 ms 141 ms ts-a4.lax3.hugeserver.com [199.244.116.22] 18 159 ms 141 ms 141 ms hugeserver.com [107.161.48.7]
Trace complete.
I don't know if it is a coincidence but Time Warner Cable one of the biggest US peers that Spark uses https://downdetector.com/status/time-warner-cable/news/6233-problems-at-time-warner-cable-2
This is a RIPEstat of the telecom AS number AS4771 that my current public IP is assigned to https://stat.ripe.net/AS4771
This shows there were nearly 15000 route changes which is huge compared to the normal route change frequency. Nothing that large has happened since late september. https://stat.ripe.net/AS4771#at-a-glancerouting
All my traffic goes through AS4648 of which 60% goes through Level3 and about 20% through NTT http://bgp.he.net/AS4648#_peers .
Yep just wanted to say things are noticeably better today so thanks for investigating this issue. Will keep an eye on it and advise if the problem reoccurs.
Are you subscribed to our RSS feed? You can download the latest headlines and summaries from our stories directly
to your computer or smartphone by using a feed reader.
