Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Ronski]

Since January we seem to be having periods of high packet loss, this is noticeable as websites occasionally don't load, and my daughters complain the wi-fi has gone off. When we get the issues if I look in pFsense then it shows a very high RTT, lots of loss and states the connection is down or has latency (see attached). It's not completely down, as the modems not dropped the connection, but the pings are very high, unless very bad speed tests still run and record good speeds.

The test immediately before this one actually recorded a high ping and 0/0Mbps speeds, but didn't get saved to my results



21/01/2020


04/04/2020



Pinging bbc.co.uk [151.101.128.81] with 32 bytes of data:
Reply from 151.101.128.81: bytes=32 time=427ms TTL=58
Reply from 151.101.128.81: bytes=32 time=13ms TTL=58
Reply from 151.101.128.81: bytes=32 time=440ms TTL=58
Reply from 151.101.128.81: bytes=32 time=493ms TTL=58

I have asked another user who's just down the road from me and on Virgin, and he's not had any complaints from his family, who've been hammering the connection recently, he's going to set up a ping monitor to see what the connections like.

In another thread d2d4j suggested running a pathping, so here's what I get

Code: [Select]

pathping bbc.co.uk

Tracing route to bbc.co.uk [151.101.64.81]
over a maximum of 30 hops:
  0  MY-PC [192.168.0.11]
  1  pfSense.changed.co.uk [192.168.0.1]
  2  10.213.48.1
  3  brig-core-2a-xe-830-0.network.virginmedia.net [80.3.65.181]
  4     *        *        *
Computing statistics for 75 seconds...
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
  0                                           MY-PC [192.168.0.11]
                                0/ 100 =  0%   |
  1    0ms     0/ 100 =  0%     0/ 100 =  0%  pfSense.changed.co.uk [192.168.0.1]
                                9/ 100 =  9%   |
  2  365ms    98/ 100 = 98%    89/ 100 = 89%  10.213.48.1
                                0/ 100 =  0%   |
  3  186ms     9/ 100 =  9%     0/ 100 =  0%  brig-core-2a-xe-830-0.network.virginmedia.net [80.3.65.181]

Trace complete.

That was run when things were working OK but my ping was a little high, but numbers 2 & 3 don't look good, and suggest the problem is on Virgins network??

And this

Code: [Select]

tracert bbc.co.uk

Tracing route to bbc.co.uk [151.101.192.81]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  pfSense.changed.co.uk [192.168.0.1]
  2   592 ms   536 ms   587 ms  10.213.48.1
  3   577 ms   549 ms   492 ms  80.3.65.221
  4     *        *        *     Request timed out.
  5    24 ms    19 ms    19 ms  eislou2-ic-3-ae0-0.network.virginmedia.net [94.174.238.226]
  6    16 ms    25 ms    16 ms  157.52.127.6
  7    16 ms    15 ms    17 ms  151.101.192.81

Trace complete.
I watched over two hours of Netflix last night & Friday night, didn't have any issues, so I'm guessing it's buffered enough to keep playing with out problems.

Is this something on the Virgin network, or an issue on my network given the other user I've asked doesn't appear to have any issues?

[d2d4j]

Hi ronski

Many thanks

Yes 2 and 3 don’t look good but could be edge dropping pings if busy servicing other things

I would shutdown your pfsense, and vm modem, wait 2 minutes (with power lead unplugged from power socket completely), and then turn on the vpn modem and pfsense

Test - is this any better

Many thanks

John

[d2d4j]

Hi

Sorry I would change pfsense dns servers to 1.1.1.1 and 8.8.4.4

Many thanks

John

[Ronski]

Thanks John, currently using 9.9.9.9 and 149.112.112.112

I will try your suggestions now.

[Ronski]

Unfortunately neither made any difference, currently RTT and RTTsd are fluctuating into the 200 to 400ms range.

[d2d4j]

Hi ronski

Many thanks

Could I ask if you unplugged power lead as well (there is a reason for this - allows stacks to fully reset to zero)

Is your first hop (10.) still the same as hops past that ping fine from here

If so, I would report to VM as it’s only vm who can fix

Many thanks

John

[Ronski]

I did initially forget to unplug the power leads, so did it again, this time unplugging the power leads.

Not sure what you mean by "Is your first hop (10.) still the same as hops past that ping fine from here"

Looks like I'll have to put the modem back to router mode, and run it like that for a while, as I know full well Virgin will ask me to do that. Was hoping not to have to deal with their support again, will use the forum.

[Ronski]

Realised what you meant

tracert bbc.co.uk

Tracing route to bbc.co.uk [151.101.128.81]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  pfSense.changed.co.uk [192.168.0.1]
  2    10 ms     9 ms    13 ms  10.213.48.1
  3    21 ms    11 ms    13 ms  brig-core-2a-xe-833-0.network.virginmedia.net [80.3.65.193]
  4     *        *        *     Request timed out.
  5    13 ms    13 ms    15 ms  eislou2-ic-2-ae0-0.network.virginmedia.net [62.254.84.62]
  6    16 ms    14 ms    15 ms  157.52.127.8
  7    14 ms    15 ms    16 ms  151.101.128.81

Need to try it really when things are being problematic

[d2d4j]

Hi ronski

Many thanks and sorry, should have been clearer sorry

The pings looks fine on the last test

Have you changed modem settings. If not, I would leave and periodically test as it appears intermittent

I am thinking the 10. Is your gateway so if so, it should affect all users using that gateway, which hopefully vm should be aware and fix

If checking with neighbours, could you ask if they use same first hop 10.

Many thanks

John

[niemand]

The neighbours will use the same router. If on a different cabinet a slim chance the immediate neighbours will use a different port but VM kit is pretty geographic in separation much as Openreach kit is even if you may not be able to see that on traceroute.

[Ronski]

So currently back to fiddling with the broadband. This morning I switched the modem back to router mode, and therefore took pfSense out of the equation

This is in modem mode and pfSense



This is in router mode, no pfSense, and so far it looks much better.



My IP address has changed, but everything else seems the same:

Code: [Select]

pathping bbc.co.uk

Tracing route to bbc.co.uk [151.101.128.81]
over a maximum of 30 hops:
  0  MY-PC [192.168.0.11]
  1  192.168.0.1
  2  10.213.48.1
  3  brig-core-2b-xe-832-0.network.virginmedia.net [80.3.65.221]
  4     *        *        *
Computing statistics for 75 seconds...
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
  0                                           X99-PC [192.168.0.11]
                                0/ 100 =  0%   |
  1    1ms     0/ 100 =  0%     0/ 100 =  0%  192.168.0.1
                                0/ 100 =  0%   |
  2    8ms    98/ 100 = 98%    98/ 100 = 98%  10.213.48.1
                                0/ 100 =  0%   |
  3   10ms     0/ 100 =  0%     0/ 100 =  0%  brig-core-2b-xe-832-0.network.virginmedia.net [80.3.65.221]

Trace complete.

C:\tracert bbc.co.uk

Tracing route to bbc.co.uk [151.101.128.81]
over a maximum of 30 hops:

  1     1 ms     1 ms     2 ms  192.168.0.1
  2     9 ms     9 ms     7 ms  10.213.48.1
  3    11 ms    11 ms    11 ms  brig-core-2b-xe-832-0.network.virginmedia.net [80.3.65.221]
  4     *        *        *     Request timed out.
  5    23 ms    21 ms    23 ms  eislou2-ic-3-ae0-0.network.virginmedia.net [94.174.238.226]
  6    18 ms    17 ms    17 ms  157.52.127.6
  7    18 ms    19 ms    18 ms  151.101.128.81

Trace complete.
I'll see what it looks like tomorrow morning.

[Ronski]

So 24 hours later, and I still have a very good ping graph, and no complaints about the broadband.



So where do I go from here in tracking down the issue?

As I'm not having issues in router mode is it definitely something in my setup causing it, or could it still be something on Virgins network (different routing perhaps)?

Edited to add, this is Friday's graph when running in modem mode and pfSense as the router

[PhilipD]

Hi

Have you looked in the pfSense various system logs to see if there are any errors reported or perhaps something constantly restarting (DNS Resolver perhaps).  Have you recently updated to pfSense 2.4.5 which is causing problems for some along these lines.

Regards

Phil 

[Ronski]

Hi Phil,

No I haven't looked in any logs, given VMs history it was expected to be them at fault and I deliberately hadn't updated to 2.4.5, IIRC I'm on v 2.4.4

I'll switch back to modem mode and pfSense and have a look at the logs, at some point today or tomorrow.

Thanks

[Alex Atkin UK]

Hi

Have you looked in the pfSense various system logs to see if there are any errors reported or perhaps something constantly restarting (DNS Resolver perhaps).  Have you recently updated to pfSense 2.4.5 which is causing problems for some along these lines.

Regards

Phil

I had DNS resolver crash just today, but that's been an ongoing thing with pfSense sometimes, not specific to 2.4.5.  If anything, 2.4.5 was a huge improvement on older versions, its so much faster.

What does tend to trip up pfSense is packet loss and high pings in general, as it will mark the gateway as down and restart the firewall.  Its why I found hooking up 4G to pfSense unusable, as when one link goes down, the whole thing goes down as the firewall reloads.

It may be you need to fiddle with the gateway settings so its more tolerant of high latency.  System, Routing, Edit your WAN gateway, Advanced settings, Latency/Packet loss thresholds.  Also using a different Monitor IP might help, if the default one is not responding to ICMP well.  But this is assuming you are seeing it reporting high latency/packet loss.

This is where DNS Resolver is also prone to fail, it occasionally does not restart correctly with the firewall.  Heck, even the firewall itself sometimes glitches as I find my load balancing rules are missed and I get stuck with routing out only one ISP.  I did kinda assume maybe it was my unique configuration, but it does suggest its something that can happen.

Its not a huge problem in general, but I can see how having just the wrong scenario it could be.