New 1GB FTTP with strange regular latency spikes, is it normal?

[Lordcast0r]

Hi folks, I got my new 1gb FTTP installed the other day, VM is new in my area, the connection is maybe 3 days old at this point.

I ran some line tests for latency and jitter, as I will be using the line for gaming, straight away I noticed the latency wasn't as good as my VDSL BT, it was okay but definitely slower latency wise to the same servers.

here is the BQM, I wonder if with the connection being new that this is some sort of testing phase?

[Riche90]

Just want to say, you're a hero. Not joking I logged in just now to post this graph and ask exactly the same. I'm on a new XGS PON install and my BQM is identical. 

Looks to be 100ms spikes exactly an hour apart? 

[Riche90]

I am wondering actually, as due to a few dropouts I was having since install. I used the websites checker which runs tests against your connection. I did notice however, that this appears to test regularly automatically (potentially hourly?). If you put your postcode and surname in HERE, you can see the timestamp of the last test. It may be that while this test is running over 10 or so seconds, ICMP isn't prioritised and results in what looks like an increase in latency?

I pinged OUTWARDS to google.co.uk with constant and left it running while manually running a test and my latency didn't increase. But obviously this could be a representation of how ping is prioritised for incoming requests when something else is happening. It would certainly make sense to me!

[Lordcast0r]

Hi Riche90, thanks for the reply, may I ask how new is your FTTP?

[Riche90]

Was installed last Tuesday. So I've got the Hub 5X.. and all the headache that comes with it. I had disconnects every day or so since Tuesday and a complete outage yesterday between 8am-4pm. 

[Riche90]

Doesn’t line up with that time actually and seems to update when I click but I swear earlier it had an earlier timestamp. Still with the frequency being so solid, I have to expect it’s something causing ICMP responses to be delayed or deprioritised. Have you managed to track poor network performance at these specific times? I’ll try tail a constant ping out to a file, track the times and line them up. 

[Tudor]

I saw something like this reported on the board with a Hub5 when they first came out, I think VM were doing some kind of monitoring. Believe it either just went away or there was a firmware update.

[Riche90]

Yeah it seems too perfect to not be on purpose.. at first I wondered if there was correlation between the average ping riding up to the peak and then dropping at this spike but my graph has them in the middle. I think they're separate things. Mine has continued since. 

[Riche90]

Well mine was consistent, now the graph will be red as my connection is down AGAIN. They cancelled my engineer visit the other day because they said the 8am-4pm issue was widespread and being reported by many. I asked if it was related to my daily outages since install though, to which I was told "maybe it built up to it". Now I have to book an engineer for 2 days away and likely won't have internet again all tonight. 😑

[Lordcast0r]

Awe man that sucks, nothing worse I can think of than no internet 😂

I still have my VDSL connection ATM, and I won't get rid of it until I know for sure VM is gonna be stable at all hours of the day...

I'm hoping these spikes are just some sort of monitoring that's going on, I didn't do any kind of research to see if others have noticed anything. I haven't had any drops or lag while in game, I did however notice the slightly unstable ping, which led me to doing the tests, but so far no real issues.

I hope you get yours sorted soon