@websquadza Last night we saw on VumaTel - from me (Northern Suburbs) and a colleague (Gardens/CityBowl) - some spurious packet losses (Fortigate SDWAN SLA reporting ~1-2% packet losses) to Zoom.us, some AWS hosted site and to smtp for Gmail and 8.8.8.8 for DNS responses. It did cause Zoom and Skype some connection drops - also seemed DNS was impacted during this time though pings might've been fine
-
Join the MyBroadband community
South Africa’s biggest forum. Discuss, discover, and connect with thousands of members.
-
Question of the Day: What's the most interesting small town you've visited in South Africa?
Web Squad ISP
- Thread starter Gazed416
- Start date
- Status
You didn't get any MTRs for us to look at by any chance?
We're working on a theory here; you're on Vuma Aerial/Village as far as I remember?@websquadza I am getting pretty low download speeds internationally compared to some of the other people I see in the thread. Any idea what the issue is?
Looks like someone borked up something on the VLAN configs there. Pinged the L2 guy we know on their team and sorted it out (fairly) quickly.
Thanks for that!
The Vuma Villages/SADV one, I am still on trenched fibre, same same but different.
BouncyNinja
Executive Member
- Joined
- Oct 15, 2009
- Messages
- 5,225
- Reaction score
- 2,522
Evening,
Quick question, would this packet loss be normal? or is that the vooooma cpe dropping ?
Only loss if it persists to the last hop. That one server just isn't responding properly.
Common factor is the L2 network. Will advise once we’ve tested our theories a little further
As @tehan said, it’s only packet loss if it carries through to the last hop. Explained the technical side a few posts up.
rvanwyk
Expert Member
Any one else have no connectivity (red light flashing on CPE) in Wilgeheuwel (Vuma Aerial)?
rvanwyk
Expert Member
Last edited:
rvanwyk
Expert Member
Is it possible to get an update and ETR from Vumatel perhaps?
@websquadza - quite a few usenet providers seem to use the same DC in the Netherlands. Seems to be 81.171.92.xxx range. The trace / latency to these servers seems more indicative of going London > USA > NL as opposed to London > NL. I think this is a recent thing as I can't remember this being the case before. An example is news-nl.newshosting.com, eu.news.astraweb.com, eunews.frugalusenet.com etc. All seem to take a very roundabout route and have a latency over 230ms.
Any reason for this? Pretty sure this would affect download speeds if packet loss were involved wouldn't it?
Any reason for this? Pretty sure this would affect download speeds if packet loss were involved wouldn't it?
rvanwyk
Expert Member
Any update from Vumatel?
rvanwyk
Expert Member
Still nothing
Our network status page will have any updates that we receive from Vumatel regarding NWIs. Vumatel don’t supply ETRs for network incidents. From what I see, your outage is related to vandalism of infrastructure and the Vumatel team are working on replacing and repairing the damage- I really can’t give you a guess as to how long this will take.
International speed today ain't looking great, also noticed an extra 20ms latency to London on Speed Test, Fortnite servers which are usually defaulting to the middle east at 120ms latency are now hitting EU instead which much higher latency. Something seems off.
Any MTRs to these servers we can check? Remember latency changes for every single destination on the planet. Helps narrow it down when we can see what paths may be affected.
Fortnite is hosted on AWS’s network and we have no control over this routing. The moment we hand over (and pick up from) to AWS’s network in JHB, it’s in their hands, this is the same for all networks that peer with AWS in JHB - it could be one of AWS’s paths (east coast) is down.
- Status