[TelstraClear] Johnsonville Micro-outages for past week

Forums › Vodafone New Zealand › [TelstraClear] Johnsonville Micro-outages for past week - support too slow

1 | 2

jpollock

600 posts

Ultimate Geek
+1 received by user: 5

Trusted

Reply # 126683 28-Apr-2008 07:52

I've got the cronjob checking google with a ping every minute. If the ping isn't answered within a second, I follow it up with a traceroute.

Since then, I've had the job trigger several times, but no smoking gun. (of course)

It does trigger moderately frequently. You can pretty much track the busy hours on my connection with it. ;)

---
http://blog.jason.pollock.ca

timmmay

12871 posts

Uber Geek
+1 received by user: 2111

Trusted

Subscriber

Reply # 126729 28-Apr-2008 10:27

I had a 3 pings running for a few hours, one to the gateway, one to the dns server, one to google. They all dropped packets every now and then, sometimes for what will have been a microoutage. TC said they want to know if I can ping the gateway during an outage, if not they'll send out a technician. That seems like a waste of time to me, it's an intermittant problem, they might not see it.

AWS Certified Solution Architect Professional, Sysop Administrator Associate, and Developer Associate
TOGAF certified enterprise architect
Professional photographer

jpollock

600 posts

Ultimate Geek
+1 received by user: 5

Trusted

Reply # 128158 3-May-2008 12:57

Just had one, lasted for 5 minutes.

The missing hop seems to be:

 6  i-4-0.wil-core02.net.reach.com (202.84.142.129)  161.950 ms  171.548 ms  168.385 ms

I frequently see Bertha really, really slow to respond to pings, but this is the first outage I've seen.

traceroute to 72.14.253.147 (72.14.253.147), 30 hops max, 38 byte packets
 1  pe14-bertha-loopback.paradise.net.nz (203.96.153.96)  16.573 ms  12.622 ms  13.722 ms
 2  ae0-840.ie4.telstraclear.net (203.167.223.114)  8.544 ms  10.484 ms  9.526 ms
 3  ge-1-2-0-843.ie1.telstraclear.net (218.101.61.105)  20.842 ms  22.232 ms  19.312 ms
 4  ge-0-2-0-1.xcore1.acld.telstraclear.net (203.98.50.251)  18.987 ms  18.792 ms  19.218 ms
 5  203.167.233.10 (203.167.233.10)  19.715 ms  19.864 ms  19.216 ms
 6  * * *
 7  * * *
 8  * * *
 9  * * *
10  * * *
11  * * *
12  * * *
13  * * *
Icmp checksum is wrong
14  * * *
15  * * *
16  * * *
17  * * *
18  * * *
19  * po-in-f147.google.com (72.14.253.147)  223.118 ms  221.310 ms

---
http://blog.jason.pollock.ca

JAYJAY

117 posts

Master Geek
+1 received by user: 6

Reply # 128248 3-May-2008 18:10

I've had three today. The technician was supposed to be here today also but called his work as he was sick.

JAYJAY

117 posts

Master Geek
+1 received by user: 6

Reply # 130402 12-May-2008 21:57

Tech has been out. Replaced the modem. Two hours later the connection died so I'm doing some more digging myself. I've removed the router and tested for a day and it's fine. Just put the router back on to transfer files across my network and I'll leave it on the night. I'm guessing when I wake up it'll be down.

timmmay

12871 posts

Uber Geek
+1 received by user: 2111

Trusted

Subscriber

Reply # 130408 12-May-2008 22:06

I'd love to hear what you find, i've had a few outages today as well. I have to try swapping out the router, but that's not a good way to work it out, it still relies on me noticing them. It also cuts off everyone running off wireless.

AWS Certified Solution Architect Professional, Sysop Administrator Associate, and Developer Associate
TOGAF certified enterprise architect
Professional photographer

BarTender

2206 posts

Uber Geek
+1 received by user: 623

Trusted

Reply # 130418 12-May-2008 22:48

I was out in JVill today as well, and we had an outage that lasted for ~1 1/2 mins ~12:20pm. Unfourtunatly I didn't have access behind their router to traceroute etc (nat is turned off and we proxy everything). But I just realised now that the customer I was working at is on TCL Cable, and that could explain why we lost internet for a few mins, only to have it randomly reappear, without touching or rebooting anything.... Strange things are afoot at the Circle K tonight.

jpollock

600 posts

Ultimate Geek
+1 received by user: 5

Trusted

Reply # 130477 13-May-2008 08:37

I've been having terrible performance the past couple of weeks. The pings to WoW have been really flakey, and there appears to be a lot of congestion between us and Bertha.

Last night at 2:15-2:50, the network was taken down:

traceroute to 72.14.253.147 (72.14.253.147), 30 hops max, 38 byte packets
 1  * * *
 2  203-97-96-78.cable.telstraclear.net (203.97.96.78)  2740.080 ms !H  2998.105 ms !H  2999.943 ms !H

Probably a planned outage since it happened at 2:15 AM. :)

---
http://blog.jason.pollock.ca

sbiddle

25183 posts

Uber Geek
+1 received by user: 5035

Moderator

Trusted

Biddle Corp

Subscriber

Reply # 130482 13-May-2008 09:03

jpollock: I've been having terrible performance the past couple of weeks. The pings to WoW have been really flakey, and there appears to be a lot of congestion between us and Bertha.

Last night at 2:15-2:50, the network was taken down:
traceroute to 72.14.253.147 (72.14.253.147), 30 hops max, 38 byte packets
 1  * * *
 2  203-97-96-78.cable.telstraclear.net (203.97.96.78)  2740.080 ms !H  2998.105 ms !H  2999.943 ms !H
Probably a planned outage since it happened at 2:15 AM. :)

There is work being done every night at present moving upgrading nodes and moving them over to the new CMTS's. This has started a few months ago and is ongoing.

timmmay

12871 posts

Uber Geek
+1 received by user: 2111

Trusted

Subscriber

Reply # 130515 13-May-2008 11:05

jpollock: I've been having terrible performance the past couple of weeks. The pings to WoW have been really flakey, and there appears to be a lot of congestion between us and Bertha.

Last night at 2:15-2:50, the network was taken down:
traceroute to 72.14.253.147 (72.14.253.147), 30 hops max, 38 byte packets
 1  * * *
 2  203-97-96-78.cable.telstraclear.net (203.97.96.78)  2740.080 ms !H  2998.105 ms !H  2999.943 ms !H
Probably a planned outage since it happened at 2:15 AM. :)

The 2700ms ping is because they're servicing a router. To bypass the router during service they have a team of trained monkeys writing down the contents of packets coming in on one link, then typing them into another link. 2700ms is fairly impressive for a bunch of monkeys.

AWS Certified Solution Architect Professional, Sysop Administrator Associate, and Developer Associate
TOGAF certified enterprise architect
Professional photographer

jpollock

600 posts

Ultimate Geek
+1 received by user: 5

Trusted

Reply # 130516 13-May-2008 11:08

IPoM... Is that a new version of IPoAC?

http://en.wikipedia.org/wiki/IP_over_Avian_Carriers

---
http://blog.jason.pollock.ca

1 | 2