LATENCY!!!

Pinging telkom.co.za [196.43.22.222] with 32 bytes of data:
Reply from 196.43.22.222: bytes=32 time=147ms TTL=245
Reply from 196.43.22.222: bytes=32 time=170ms TTL=245
Reply from 196.43.22.222: bytes=32 time=174ms TTL=245
Reply from 196.43.22.222: bytes=32 time=170ms TTL=245

Ping statistics for 196.43.22.222:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 147ms, Maximum = 174ms, Average = 165ms

Whatever improvement there was, it was short-lived..
 
What he said^


Pinging www.webafrica.co.za [196.220.58.66] with 32 bytes of data:
Reply from 196.220.58.66: bytes=32 time=160ms TTL=121
Reply from 196.220.58.66: bytes=32 time=138ms TTL=121
Reply from 196.220.58.66: bytes=32 time=157ms TTL=121
Reply from 196.220.58.66: bytes=32 time=154ms TTL=121

Ping statistics for 196.220.58.66:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 138ms, Maximum = 160ms, Average = 152ms
 
Well if nothing else the body of evidence we have gathered is so strong that this simply cannot be ignored anymore so let's call that the first small victory. I have been pushing pretty much continuously for answers this week and will continue to do so next week.

Hang in there people.

Regards
Will
 
Well I'm look at 40-50 ping @ 11 PM, but I'm going to assume it's coz it's Friday and everyone is out partying :/

*Cancel that. It's still borked.
 
Last edited:
teRRRier said:
Yep, unfortunately you tested right in the dip. This was the full graph for the whole day - http://www.imagehost.co.za/share-47F3_4CB93AC5.html
Click to expand...

Perfect chart, showing lulls at 5pm and 10.30pm!

Will: I was just wondering if the 'router' that is congested is an old 155Mbps ATM switch, which is being used in place of a faulty Metro-Ethernet 1-10Gbps switch. This would certainly explain the 'sharp' cutoff in capacity...
 
Started again around 9:26 - looks like it coincides with CINX hitting 180Mpbs

Edit: But that doesn't explain the large dips to normal yesterday which occurred at between 260Mbps and 300Mbps. Wierd.
 
Last edited:
I think the CINX traffic probably serves as a yardstick for how busy the Cpt Infrastructure is and the latency problems therefore more, or less follows the same curve.
 
Will@MWEB said:
@WAJeff I'm glad to see you guys are running with this as well - please feel free to drop me a PM if you'd like to swap notes - we're all on the same side on this one :)
Click to expand...

Agreed, we are all in the same boat, working together should help come to a resolution quicker. I've posted an update from Web Africa's end here and will follow up on the Souther Suburbs issue on Monday when I'm back at the office - will drop you a PM then :)
 
Tracing route to www.mweb.co.za [196.2.63.110]
over a maximum of 30 hops:

1 1 ms 1 ms <1 ms 192.168.0.1
2 23 ms 22 ms 21 ms 41-133-112-1.dsl.mweb.co.za [41.133.112.1]
3 176 ms 178 ms 174 ms 196-28-178-202.adsl.mweb.co.za [196.28.178.202]

4 159 ms 137 ms 136 ms 196.41.144.36
5 173 ms 173 ms 170 ms www.mweb.co.za [196.2.63.110]

Trace complete.
 
I only logged on at 8am, so I didn't see any spikes. Overall traffic is still low, but I will keep checking...

Update #1: Latency starting to creep up to 50ms around 11.15am. (CINX traffic graph also creeping up to 200Mbps).

Update #2: Latency briefly rocketed up to 150ms after 1.30pm. (CINX graph around 250Mbps).
 
Last edited:
Gordon_R said:
I only logged on at 8am, so I didn't see any spikes. Overall traffic is still low, but I will keep checking...

Update #1: Latency starting to creep up to 50ms around 11.15am. (CINX traffic graph also creeping up to 200Mbps).

Update #2: Latency briefly rocketed up to 150ms after 1.30pm. (CINX graph around 250Mbps).
Click to expand...

Yeah pretty much the same.
 
Same old thing as usual in Rondebosch:

Tracing route to www.mweb.co.za [196.2.63.110]
over a maximum of 30 hops:

1 <1 ms <1 ms 1 ms 192.168.0.1
2 23 ms 22 ms 24 ms 41-133-112-1.dsl.mweb.co.za [41.133.112.1]
3 144 ms 146 ms 150 ms 196-28-178-202.adsl.mweb.co.za [196.28.178.202]

4 169 ms 170 ms 161 ms 196.41.144.36
5 156 ms 168 ms 169 ms www.mweb.co.za [196.2.63.110]

Trace complete.
 
another victim

hey there, im joining late but i have the same problem as everyone here.

im in rondebosch, cape town - 021 685 xxxx.

i have a 4meg line and uncapped from mweb and at the beginning of last month my line got upgraded to 10meg and shortly after my pings got really bad most of the time. just like everyone else my local ping is in the 200ms range almost all day till about midnight so i cant play any games. i lag on mweb, web africa, and telkom accounts. i havent got any other accounts to test.

i phoned telkom and they told me they arent aware of any problems in my area and it must be my line so they dropped my line to 4meg and hung up on me, which didnt fix the problem. i dont know what else to do, but im glad i found an active thread about it. please tell me there is a solution in sight. i havent been able to play games in a month and i am getting so angry i want to kill someone.

ps. there is also occasional packet loss for me. my brother in rondebosch also has the same ping as me. so does my clan-mate in rondebosch. my friend in observatory has lag too but not as often as mine. plenty of people on the quake live servers have the same problem but telkom arent aware of any problems? bull****.
 
Last edited:
I'm surprised, with a fault of this nature affecting so many people, a news article hasn't surfaced from MyBB. Telkom acknowledges the problem to ISP techies yet their call centres are not informed when dealing with customers. Sounds to me like someone is playing ping pong with us.
 
It seems that as far as Telkom is concerned, as long as you can surf the web and send e-mails, everything is OK. Latency is directly related to congestion (or contention ratio), so its always easy to shift the blame onto someone else...

The 10Mbps rollout generated huge amounts of hype, but the downside has now become clear. If the infrastructure isn't in place, it just creates bottlenecks further up the chain (no matter how fast your ISP's links to SEACOM are).

I'm seriously wondering whether it is worth having a 4Mbps line (7Mbps in my case), when latency limits the throughput for simple browsing (not counting downloads). The line rental of R413pm has never been good value for money, and there is no sign of price reductions. Unfortunately the other choices aren't much good, and 384Kbps seems like a horrible thought...
 
After my lovely experience on Friday I managed to speak to one of the supervisors at the call centre and he admitted that they have been getting a lot of calls from the Southern Suburbs and are aware of the problem.As a temporary fix to deal with the congestion at the Rondebosch Exchange everyone is going to be or already has been downgraded to 6MB.He couldn't give any indication on timelines for a permanent fix but I wouldn't hold my breath.
Apparently these upgrades are all a trial which will be re-assessed in March.
 
As a 384k user I can assure you that it is horrible. :(

4Meg is just very expensive and 512 is too small a jump for the price. Argh.

Also, latency this morning hasn't picked up yet, although it will by 10am:

Tracing route to www.mweb.co.za [196.2.63.110]
over a maximum of 30 hops:

1 1 ms <1 ms <1 ms 192.168.0.1
2 23 ms 22 ms 22 ms 41-133-112-1.dsl.mweb.co.za [41.133.112.1]
3 25 ms 25 ms 27 ms 196-28-178-202.adsl.mweb.co.za [196.28.178.202]

4 24 ms 24 ms 27 ms 196.41.144.36
5 26 ms 28 ms 27 ms www.mweb.co.za [196.2.63.110]

Trace complete.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X