Anthro
Expert Member
Web Africa - You need to get more datacentres STAT !
Just some background on the issue:
I work for a medical company, which provides real-time claim submissions for Doctors, Trauma units, Radiologists and various other types of medical practices.
Our technology works as follows, and has a rough turnaround for each transaction of 6-8 seconds.
http://www.mededi.co.za/mededi-claims-processing-system/how-it-works/
My problem is as follows:
Dr A (based in JHB) - a fairly large trauma unit, submits claim to said server(s).
When this claim reaches our servers, a time out is set by a 'gatekeeper' for return packets, so as to prevent 'choking'.
When using WebAfrica, this client gets CONSTANT timeouts, and when doing a tracert the following results are produced:
Tracert Sample #1
Tracing route to live.mededi.co.za [196.34.148.83]
over a maximum of 30 hops:
1 2 ms <1 ms <1 ms 192.168.0.254
2 22 ms 25 ms 29 ms ndnn-ip-esr-2.ipnet.wa.co.za [41.185.67.1]
3 46 ms 46 ms 47 ms vl108.cr.gw.cpt.za.wa.co.za [196.220.59.238]
4 45 ms 51 ms 46 ms vl34.er.gw.cpt.za.wa.co.za [41.185.0.34]
5 44 ms 45 ms 49 ms upstream.vl101.gw.cpt.za.wa.co.za [196.220.59.26]
6 54 ms 44 ms 45 ms 196.43.11.234
7 163 ms 273 ms 89 ms 196.43.25.206
8 368 ms 281 ms 81 ms 196.25.247.26
9 177 ms 162 ms 59 ms core2a-ctn-gi0-1.ip.isnet.net [168.209.2.6]
10 69 ms 65 ms 68 ms core2a-pkl-gi0-0-3-1-90.ip.isnet.net [168.209.10.153]
11 583 ms 572 ms 74 ms 168.209.100.242
12 70 ms 68 ms 70 ms csw3-bry-vl2.ip.isnet.net [168.209.217.250]
13 71 ms 73 ms 72 ms medicaledipix.fw.is.co.za [196.35.71.98]
14 * * * Request timed out.
15 * * * Request timed out.
**the 196.35.71.98 is our MPLS network, and it gets forwarded correctly from there
Tracert Sample #2
1 <1 ms <1 ms <1 ms 192.168.0.254
2 34 ms 88 ms 190 ms ndnn-ip-esr-2.ipnet.wa.co.za [41.185.67.1]
3 106 ms 97 ms 187 ms vl108.cr.gw.cpt.za.wa.co.za [196.220.59.238]
4 * 46 ms 52 ms vl34.er.gw.cpt.za.wa.co.za [41.185.0.34]
5 437 ms 55 ms 229 ms upstream.vl101.gw.cpt.za.wa.co.za [196.220.59.226]
6 80 ms 134 ms 64 ms 196.43.11.234
7 47 ms * 54 ms 196.43.25.206
8 54 ms 62 ms 62 ms 196.25.247.26
9 49 ms 67 ms 52 ms core2a-ctn-gi0-1.ip.isnet.net [168.209.2.6]
10 76 ms 74 ms 68 ms core2a-pkl-gi0-0-3-1-90.ip.isnet.net [168.209.100.153]
11 71 ms 71 ms 80 ms 168.209.100.242
12 70 ms 75 ms 78 ms csw3-bry-vl2.ip.isnet.net [168.209.217.250]
13 71 ms 71 ms 77 ms medicaledipix.fw.is.co.za [196.35.71.98]
14 * * * Request timed out.
15 * * * Request timed out.
Once again, will timeout from here - since our MPLS does necessary forwarding from here
I can provide you with about 50 tracerts, each with varying results for turnaround time, that it makes me nauseous.
TL;DR
Client is having timeout issues with Web Africa, but with no other ISP
Our server is based on the IS Backbone, in Rosebank
We have clients in each of the 9 provinces - so this problem will persist, untill the client cancels - or WA puckers up.
Just some background on the issue:
I work for a medical company, which provides real-time claim submissions for Doctors, Trauma units, Radiologists and various other types of medical practices.
Our technology works as follows, and has a rough turnaround for each transaction of 6-8 seconds.
http://www.mededi.co.za/mededi-claims-processing-system/how-it-works/
My problem is as follows:
Dr A (based in JHB) - a fairly large trauma unit, submits claim to said server(s).
When this claim reaches our servers, a time out is set by a 'gatekeeper' for return packets, so as to prevent 'choking'.
When using WebAfrica, this client gets CONSTANT timeouts, and when doing a tracert the following results are produced:
Tracert Sample #1
Tracing route to live.mededi.co.za [196.34.148.83]
over a maximum of 30 hops:
1 2 ms <1 ms <1 ms 192.168.0.254
2 22 ms 25 ms 29 ms ndnn-ip-esr-2.ipnet.wa.co.za [41.185.67.1]
3 46 ms 46 ms 47 ms vl108.cr.gw.cpt.za.wa.co.za [196.220.59.238]
4 45 ms 51 ms 46 ms vl34.er.gw.cpt.za.wa.co.za [41.185.0.34]
5 44 ms 45 ms 49 ms upstream.vl101.gw.cpt.za.wa.co.za [196.220.59.26]
6 54 ms 44 ms 45 ms 196.43.11.234
7 163 ms 273 ms 89 ms 196.43.25.206
8 368 ms 281 ms 81 ms 196.25.247.26
9 177 ms 162 ms 59 ms core2a-ctn-gi0-1.ip.isnet.net [168.209.2.6]
10 69 ms 65 ms 68 ms core2a-pkl-gi0-0-3-1-90.ip.isnet.net [168.209.10.153]
11 583 ms 572 ms 74 ms 168.209.100.242
12 70 ms 68 ms 70 ms csw3-bry-vl2.ip.isnet.net [168.209.217.250]
13 71 ms 73 ms 72 ms medicaledipix.fw.is.co.za [196.35.71.98]
14 * * * Request timed out.
15 * * * Request timed out.
**the 196.35.71.98 is our MPLS network, and it gets forwarded correctly from there
Tracert Sample #2
1 <1 ms <1 ms <1 ms 192.168.0.254
2 34 ms 88 ms 190 ms ndnn-ip-esr-2.ipnet.wa.co.za [41.185.67.1]
3 106 ms 97 ms 187 ms vl108.cr.gw.cpt.za.wa.co.za [196.220.59.238]
4 * 46 ms 52 ms vl34.er.gw.cpt.za.wa.co.za [41.185.0.34]
5 437 ms 55 ms 229 ms upstream.vl101.gw.cpt.za.wa.co.za [196.220.59.226]
6 80 ms 134 ms 64 ms 196.43.11.234
7 47 ms * 54 ms 196.43.25.206
8 54 ms 62 ms 62 ms 196.25.247.26
9 49 ms 67 ms 52 ms core2a-ctn-gi0-1.ip.isnet.net [168.209.2.6]
10 76 ms 74 ms 68 ms core2a-pkl-gi0-0-3-1-90.ip.isnet.net [168.209.100.153]
11 71 ms 71 ms 80 ms 168.209.100.242
12 70 ms 75 ms 78 ms csw3-bry-vl2.ip.isnet.net [168.209.217.250]
13 71 ms 71 ms 77 ms medicaledipix.fw.is.co.za [196.35.71.98]
14 * * * Request timed out.
15 * * * Request timed out.
Once again, will timeout from here - since our MPLS does necessary forwarding from here
I can provide you with about 50 tracerts, each with varying results for turnaround time, that it makes me nauseous.
TL;DR
Client is having timeout issues with Web Africa, but with no other ISP
Our server is based on the IS Backbone, in Rosebank
We have clients in each of the 9 provinces - so this problem will persist, untill the client cancels - or WA puckers up.
Last edited: