Really strange ping...

Had an Aquiss FTTP (1000/110) line put in, using a UniFi UDM-SE. All seems largely fine up until recently, but am seeing some very strange latency patterns.

We use a Nest cam as a baby monitor which seems to be *incredibly* packet loss sensitive (Live stream stopped on display every time theres even a tiny bit of lag / loss. Bad design.) Checked my SmokePing and TBB monitor, and it seems to have taken a turn recently. UDM has been restarted.

Is this likely to be an OpenReach transit issue, a CityFibre routing issue or a me issue? UDM seems fine on resources (low CPU / Memory usage, no other issues). It looks like its slowly been ramping up in max ping time, seeing a lot more random packet loss, and when I sit it on my (wired) desktop pinging, I see a lot more 2-3s+ pings than I would expect to. The line is totally untaxed, theres more headroom than you could imagine there.

Any ideas which bit is most likely to be falling down here?

Imgur link of my smokeping - Ignore the 360 day one and the missing gap at begining of october!

TBB Graph

Just to add, restarted UDM / ONT etc and no visible change in ping distribution.

For comparison, when it went in they all looked like this: link

Has your UDM done an auto-update recently?

I don't believe so - I've purposefully turned off auto updates and only do them manually now as they do cause issues fairly regularly!

If you’ve got another box, can you swap it out with your UDM, to rule out that it’s “your” side of the connection?

Are Aquiss still using CityFibre/Entanet?
They do traffic shape, maybe this is the cause?

In reply to a post by devonkev:

Are Aquiss still using CityFibre/Entanet?
They do traffic shape, maybe this is the cause?

Please don't post FUD, that's really not been the case for quite some time and neither reflective of our commercial relationships with CityFibre Wholesale.

Good Morning

Not currently aware of any other reports hitting the support team, so I don't believe this is a CF issue (and we would get reports within minutes of something not being right).

Can't rule any Openreach issues at this stage as we would need to test your circuit to see if anything was being flagged.

Likewise, have you plugged directly into your ONT with some CPE hit attached, just to try and rule out any LAN equipment?

I would ask you to open a case with us directly so we can take a look as I'm doing a bit of guess work with who you are

In reply to a post by bdav:

Had an Aquiss FTTP (1000/110) line put in, using a UniFi UDM-SE. All seems largely fine up until recently, but am seeing some very strange latency patterns.

We use a Nest cam as a baby monitor which seems to be *incredibly* packet loss sensitive (Live stream stopped on display every time theres even a tiny bit of lag / loss. Bad design.) Checked my SmokePing and TBB monitor, and it seems to have taken a turn recently. UDM has been restarted.

Is this likely to be an OpenReach transit issue, a CityFibre routing issue or a me issue? UDM seems fine on resources (low CPU / Memory usage, no other issues). It looks like its slowly been ramping up in max ping time, seeing a lot more random packet loss, and when I sit it on my (wired) desktop pinging, I see a lot more 2-3s+ pings than I would expect to. The line is totally untaxed, theres more headroom than you could imagine there.

Any ideas which bit is most likely to be falling down here?

Imgur link of my smokeping - Ignore the 360 day one and the missing gap at begining of october!

TBB Graph

I was seeing strange performance issues (download would drop to a few mbs but upload was generally at 110mbs) and when I checked my TBB's monitor it showed latency spikes and packet loss on going back a while – even in periods when nothing else should have been using connection

While digging around router I realised the DoS protection was off (probably because it restricts ping rates to 1 a second) - switching it on (this was Wednesday) pretty much stopped the latency spikes on TBB (and cured performance issues) – still seeing some dropped packets but I suspect might be ping limits

I’m on Zen so may not be same issue, but it may be worth looking to see if the issue is related to some form of DoS/bot probing your router.