High latency on Cybersmart backbone

R

ReubenL

Banned
Joined
Jan 15, 2010
Messages
396
Reaction score
0
Location
Nelson Mandela Bay
Some clients on our network reported slow connectivity to our servers. Upon investigation it was found that they are using ADSL accounts from Cybersmart. Latency range from 800ms to 900ms on shaped and unshaped accounts. However there are no notices on their website.

Is anyone else experiencing the same hi latency problem? What backbone does Cybersmart use? I know that maintenance is being done on the Seacom cable.
 
I am using a 384kb line which I pay like R1000/month for because I am dumb enough to buy the expensive cap everytime.

Cybersmart sucks. My connection is terrible and has always been. I have always had a ton of issues.


I will probably switch to Mweb or something else. Definitely won't be a customer next month. They have perpetually sexually violated me from behind.

No more, thank you :)
 
CyberSmartSucks said:
Cybersmart sucks. My connection is terrible and has always been. I have always had a ton of issues.
Click to expand...

Are you an existing forum member that registered a new account to post all these posts? You might have a legitimate complaint but it also might come across like you have a personal agenda.

Posts like yours are pretty rare when it comes to CS, not saying you don't have issues.
 
Well, my issues certainly aren't rare. I rely on the connectivity and stability of my internet in order to make money. I can't have it just suddenly die out on me for a few minutes, or a few hours. It sucks... Today is just useless. Yesterday was the same.

It's pretty upsetting to be honest. I am posting here, obviously hoping to gain some attention. Perhaps some users will realize Cybersmart might not be the smart option after all. Perhaps they point out they aren't having similar issues.

I figure I could help some people out by voicing my opinion and being part of the vote.


I am not a member of this forum. It does feel personal when they keep giving me a bad line. Even when I have called them multiple times politely asking what is wrong, and how we can address the issue.

Note that I post anything with the consideration that overall internet in this country will be bad no matter what. My connection is unreasonable. And if they say I get x connection for Rx rand. Then I want that connection. Do not lie to me. Don't give me less. JUST TELL ME STRAIGHT FROM THE START that my connection will not be what I pay for. Just tell me so I can buy the better thing and see if I can afford it. Don't waste my time, cybersmart. This is what bother me the most. I pay for 384kb line, but it definitely does not perform like a 384kb line should. The amount of packages lost is outrageous. What a terrible connection.


Cybersmart Sucks.
 
Didn't you guys receive these emails from cybersmart:

Dear Customer

We are pleased to announce that we are opening another point of presence in Jhb. We are doing this primarily
so that we make use of Seacom Capacity that terminates there. This move adds another
layer of redundancy to our network as it gives us international and local break out points in Cape Town ,
Johannesburg and London as well as access to both international cable systems.

What this means to you as a customer in the long term is that the latencies between Cape Town and Johannesburg
should improve, particularly for customers in Johannesburg playing online games. Johannesburg customers will
further enjoy improved international latency as the traffic will no longer need to go all the way to Cape Town before
it can be routed to London. This is a fairly big project and has been in planning stages for more than 6 months. Work
will commence on this project on Saturday 17 April and should be complete by Friday 23 April. There should
be no outages during this time, however , there will be significant changes done to routing which may result slow
internet speeds both locally and internationally.

Should there be any changes to this schedule we will post these changes on http://faq.cybersmart.co.za.
Thank you for your indulgence, we believe the change will be worth it

Regards

Team Cybersmart
Click to expand...

and

Dear Customer,

We have suspended our processes as per the previous warning because of
this warning that we have just received.

You are not expected to understand all the jargon but the essential point is
that there will be delays and problems during the next few days which are
completely out of our control.

Team Cybersmart

Please note:

All non-Restorable traffic as well as all Restorable on the
Spurs to all the cable stations on the Southern SAT3 Segment excepting Accra
and Abidjan, will be down for the duration of the power reconfiguration (4-6
Hours) from 00:01 UTC (02:01 SA Time) on the morning of 17 April 2010. In
addition to the below communication from Global Capacity Business, the NNOC
Plan Work communication process as used on previous such occasions is in the
process of being implemented.

Background You might recall that on 2nd February 2010 we had an approximately
2-hour loss traffic loss for both Restorable and non-Restorable traffic in the
middle of the Normalisation process that was at the time in progress following
the resolution of the problem on the Lagos Branching Unit (BU) off the coast of Nigeria.
The cause behind the 2nd Feb failure was a faulty Abidjan (Ivory Coast) Power feeding
Equipment (PFE) The Vendor (Alcatel-Lucent Submarine Networks - ASN), in
conjunction with the relevant SAT3 Landing Parties have since repaired the
PFE which must now be re-introduced into the SAT3 system. We have since the
2nd Feb PFE failure been feeding all power for the SAT3 Southern Segment
(Melkbosstrand to Abidjan) from Melkbosstrand Cable Station which is a very
risky situation as under such conditions the PFE redundancy status across
the full SAT3 Southern Segment significantly reduced. We unfortunately had
to wait for similar work on SMW3 (cable running between the Africa and
Europe Continents to which we Restore SAT3 traffic), which activity only now
gave the SAT3 Consortium the maintenance window being used below.

It is of absolute importance that this be done now so we can establish/fix all
problems currently existing on SAT3 prior to the FIFA event. Dear All As
indicated below this is the very short notification in order to localise the
fault on S3WS Southern Trunk. SMW3 granted us a window in order to perform
the necessary work to determine where on the P3 segment is the problem.
Below follow the sequence of events: 1. 15 &16 April 2010: Combined RET &
Restoration implementation - Patching breaks to all restorable traffic on
P3, except for festooning station traffic. Window: 0:01 UTC until completed
. 2. 17 April 2010: Powering of ABJ PFE & Reconfigure BU i.e.: * Power down
the trunk * Switch the BU * Powering up as follows: Dual end feed between
MLK and ACC on the trunk. ABJ feed to the BU. * All NR traffic as well as
all (R and NR) festooning traffic down for duration of the power
reconfiguration (4-6 Hours): start 0:01 UTC on 17 April 2010. 3. 18 April
2010: Determine the location of the fault on the P3 trunk of S3WS and the
next maintenance window. 4. 19 April 2010: Normalisation of restored traffic
- Patching breaks to all restorable traffic on P3, except for festooning
station traffic. Please note: For simplicity sake and to save some time, the
restoration implementation for each step in the plan will be performed
immediately after successful testing of the step. All cable station staff is
required to be at the station for the whole duration of the restoration
schedule and NOT just at the time their step is scheduled for. This is very
important and will ensure that no time is wasted, since the ORCO will move
to the next step in the plan if station staff is not available or cannot be
contacted. Attached are the following documents: 1. Restoration (and
testing) schedule to restore all P3 traffic, excluding festooning stations.
2. List of TSA NR and festooning traffic on P3, which will be down for the
duration of P3 trunk powering exercises. 3. Complete list of TSA Traffic on
P3 segment. 4. P3 URM Plan dated 26 February 2010. Sorry for the short
notification period, but this planned work is critical and we only have a
narrow window to perform it in. SA Time = UTC Time + 2 Hours Posted: 16 Apr 10

Network Notice
Dear Customer UPDATE: Telkom have informed us that due to emergency
maitenance, traffic on SAT3 has been rerouted onto a restoration path. The
target date for normalisation of traffic onto primary SAT3 path is 19 April
2010. IMPACT: Increased latency ATTENDING: Telkom engineers STATUS:
Monitoring We are still experiencing the following service event, as per our
initial notification: Dear IS Customer We are experiencing the following
service event: AFFECTED SERVICES: International ADSL, Access, MPLS VPN
AFFECTED POPS: INTL traffic between SA (Cape Town, Rosebank) and London, New
York GRADE OF SERVICE IMPACT: Major SERVICE IMPACT: Increased latency CAUSE
OF EVENT: Telkom SAT3 circuits on restoration path TIME: 15-4-10 15h15 OUR
REF: N/A CURRENT STATUS: Monitoring ATTENDING: Telkom engineers ADDITIONAL
INFORMATION: Telkom have informed us that due to emergency maitenance,
traffic on SAT3 has been rerouted onto a restoration path. The target date
for normalisation of traffic onto primary SAT3 path is 19 April 2010.
Click to expand...
 
It usually helps to read the network notice that is sent your cybersmart email address. If you did you would know there is maintenance from this Friday till the next Friday while we are preparing for a breakout in jhb. So yes, latency and responsiveness will suck till then.

Regards
Laurie

CyberSmartSucks said:
Well, my issues certainly aren't rare. I rely on the connectivity and stability of my internet in order to make money. I can't have it just suddenly die out on me for a few minutes, or a few hours. It sucks... Today is just useless. Yesterday was the same.

It's pretty upsetting to be honest. I am posting here, obviously hoping to gain some attention. Perhaps some users will realize Cybersmart might not be the smart option after all. Perhaps they point out they aren't having similar issues.

I figure I could help some people out by voicing my opinion and being part of the vote.


I am not a member of this forum. It does feel personal when they keep giving me a bad line. Even when I have called them multiple times politely asking what is wrong, and how we can address the issue.

Note that I post anything with the consideration that overall internet in this country will be bad no matter what. My connection is unreasonable. And if they say I get x connection for Rx rand. Then I want that connection. Do not lie to me. Don't give me less. JUST TELL ME STRAIGHT FROM THE START that my connection will not be what I pay for. Just tell me so I can buy the better thing and see if I can afford it. Don't waste my time, cybersmart. This is what bother me the most. I pay for 384kb line, but it definitely does not perform like a 384kb line should. The amount of packages lost is outrageous. What a terrible connection.


Cybersmart Sucks.
Click to expand...
 
My connection has always sucked. This just further annoyed me.

I hope Mweb will provide better service.
 
It usually helps to try and not be smart. It also usually helps to read my entire post. I have always has issues.

laurie said:
It usually helps to read the network notice that is sent your cybersmart email address. If you did you would know there is maintenance from this Friday till the next Friday while we are preparing for a breakout in jhb. So yes, latency and responsiveness will suck till then.

Regards
Laurie
Click to expand...
 
wish saol would send the same notices to there customers since they on the same backbone ....
 
Anyone notice the improvement in latency lately? It's frigging low, getting about 215ms to the UK ;)
 
I phoned Cybersmart yesterday, ENRAGED that I have had to reinstall my router to the default firmware, losing all my NAT configs, reconfiguring my network, totally changing around my PS3's network config, and having to research till VERY late at night as to why exactly Modern Warfare2 on PS3 is not logging on to the Matchmaking Servers anymore. They allegedly "reset" my connection, and luckily before i reformatted my PS3, losing all my downloaded games / data, I thought i'd try my MTN (VERY HIGH Latency) 3G router. Usually it has pretty terrible latency, and only works "sometimes"... fortunately for me it was lying in my car... and VOILA! It worked, Latency was low as usual, but my network had no issues. And i realized that the PlaystationNetwork is taking unusually long to log in via Cybersmart, as well as most websites.

Whatever they did last week with the submarine cable, and how they subsequently routed everything, i have now decided that i've had enough... They're obviously routing traffic via Satellite, bouncing via 23 GPS Birds, then down to Abhu Dhabi, via Fibre to London and then SNAIL MAIL to NYC!

LlewellynsMac:~ llewellyn$ ping www.yahoo.com
PING any-fp.wa1.b.yahoo.com (67.195.160.76): 56 data bytes
64 bytes from 67.195.160.76: icmp_seq=0 ttl=52 time=754.747 ms
64 bytes from 67.195.160.76: icmp_seq=1 ttl=52 time=508.794 ms
64 bytes from 67.195.160.76: icmp_seq=2 ttl=52 time=675.841 ms
64 bytes from 67.195.160.76: icmp_seq=3 ttl=52 time=564.724 ms
64 bytes from 67.195.160.76: icmp_seq=4 ttl=52 time=703.860 ms
^C
--- any-fp.wa1.b.yahoo.com ping statistics ---
6 packets transmitted, 5 packets received, 16.7% packet loss
round-trip min/avg/max/stddev = 508.794/641.593/754.747/90.993 ms

--------- PS Server 1----------

LlewellynsMac:~ llewellyn$ ping 63.84.59.18
PING 63.84.59.18 (63.84.59.18): 56 data bytes
64 bytes from 63.84.59.18: icmp_seq=0 ttl=52 time=303.955 ms
64 bytes from 63.84.59.18: icmp_seq=1 ttl=52 time=304.068 ms
Request timeout for icmp_seq 2
64 bytes from 63.84.59.18: icmp_seq=3 ttl=52 time=304.321 ms
64 bytes from 63.84.59.18: icmp_seq=4 ttl=52 time=303.723 ms
64 bytes from 63.84.59.18: icmp_seq=5 ttl=52 time=300.559 ms
64 bytes from 63.84.59.18: icmp_seq=6 ttl=52 time=325.160 ms
^C

------ PS Server 2 -------

LlewellynsMac:~ llewellyn$ ping 69.192.17.215
PING 69.192.17.215 (69.192.17.215): 56 data bytes
Request timeout for icmp_seq 0
64 bytes from 69.192.17.215: icmp_seq=1 ttl=54 time=279.382 ms
Request timeout for icmp_seq 2
Request timeout for icmp_seq 3
64 bytes from 69.192.17.215: icmp_seq=4 ttl=54 time=319.855 ms
Request timeout for icmp_seq 5
Request timeout for icmp_seq 6
64 bytes from 69.192.17.215: icmp_seq=7 ttl=54 time=340.065 ms
64 bytes from 69.192.17.215: icmp_seq=8 ttl=54 time=363.502 ms
64 bytes from 69.192.17.215: icmp_seq=9 ttl=54 time=387.259 ms
64 bytes from 69.192.17.215: icmp_seq=10 ttl=54 time=308.114 ms
^C
--- 69.192.17.215 ping statistics ---
12 packets transmitted, 6 packets received, 50.0% packet loss
round-trip min/avg/max/stddev = 279.382/333.030/387.259/35.570 ms

-------- PS Server 3 -------

LlewellynsMac:~ llewellyn$ ping 69.192.17.215
PING 69.192.17.215 (69.192.17.215): 56 data bytes
64 bytes from 69.192.17.215: icmp_seq=0 ttl=54 time=279.139 ms
64 bytes from 69.192.17.215: icmp_seq=1 ttl=54 time=364.558 ms
64 bytes from 69.192.17.215: icmp_seq=2 ttl=54 time=489.970 ms
64 bytes from 69.192.17.215: icmp_seq=3 ttl=54 time=309.329 ms
64 bytes from 69.192.17.215: icmp_seq=4 ttl=54 time=300.573 ms
64 bytes from 69.192.17.215: icmp_seq=5 ttl=54 time=357.203 ms
64 bytes from 69.192.17.215: icmp_seq=6 ttl=54 time=310.736 ms
^C
--- 69.192.17.215 ping statistics ---
7 packets transmitted, 7 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 279.139/344.501/489.970/65.796 ms
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X