Throughput on WAN took a noose dive back in Dec

[nullspace]

So my WAN down and up started to tank sometime in december and all I can think is an update caused it. Today I had some time and tweaked the MTU a little with middling to no change:

ersion   Date   
26.1.5 (installed)   2026-03-24   
26.1.4   2026-03-11   
26.1.3   2026-03-04   
26.1.2   2026-02-12   
26.1.1   2026-02-04   
26.1   2026-01-28   
25.7.11   2026-01-15   
25.7.10   2025-12-18   
25.7.9   2025-12-04   
25.7.8   2025-11-26   
25.7.7   2025-11-06   
25.7.6   2025-10-22   
25.7.5   2025-10-08   
25.7.4   2025-09-30   
25.7.3   2025-09-09


Timestamp (GMT)   Server id   Server name   Download   Upload   Latency
2026-03-26T16:40:47   69076   Ripplefiber, Detroit, MI   154.27   119.56   34.63
2026-03-26T16:39:46   53778   123NET, Southfield, MI   321.78   162.60   23.07
2026-03-26T16:39:26   53778   123NET, Southfield, MI   427.84   150.26   18.03
2026-03-26T16:30:47   53778   123NET, Southfield, MI   303.88   158.49   23.59
2026-03-26T16:30:18   69076   Ripplefiber, Detroit, MI   175.65   118.15   35.37
2026-03-26T15:55:38   69076   Ripplefiber, Detroit, MI   282.70   66.56   46.06
2026-03-26T15:54:56   53778   123NET, Southfield, MI   309.22   197.95   22.45
2026-03-26T15:18:56   53778   123NET, Southfield, MI   298.12   178.34   31.09
2026-03-26T15:12:58   15342   Jefferson County Cable, Toronto, OH   532.03   87.85   81.72
2026-03-26T15:09:05   69076   Ripplefiber, Detroit, MI   337.01   80.93   33.71
2026-03-26T15:08:41   69076   Ripplefiber, Detroit, MI   269.19   96.78   36.84
2026-03-26T14:56:52   69076   Ripplefiber, Detroit, MI   147.98   103.90   31.88
2026-03-26T14:56:14   69076   Ripplefiber, Detroit, MI   258.65   82.40   33.43
2026-03-26T14:48:14   69076   Ripplefiber, Detroit, MI   132.60   117.18   39.13
2026-03-26T14:12:39   73050   Omni Fiber, Detroit, MI   135.47   69.68   62.72
2026-03-26T13:59:46   73050   Omni Fiber, Detroit, MI   156.37   80.17   50.09
2026-03-26T13:56:54   69076   Ripplefiber, Detroit, MI   207.24   98.28   45.51
2026-03-26T13:53:49   25491   Cable Co-op, Oberlin, OH   43.46   5.73   1035.91
2026-03-25T11:43:25   16973   Spectrum, Livonia, MI   193.45   98.21   46.64
2026-03-25T11:42:35   74288   EZEE Fiber, Detroit, MI   144.72   96.05   61.52
2026-03-25T11:42:05   5709   Merit Network, Inc, Ann Arbor, MI   288.97   93.82   70.66
2026-03-25T11:41:17   71374   Wyandotte Municipal Services, Wyandotte, MI   310.37   120.34   49.01
2025-12-28T14:45:18   6124   Merit Network, Inc, Southfield, MI   306.94   94.94   64.64
2025-12-13T14:27:32   1778   Comcast, Detroit, MI   517.04   191.64   43.55
2025-12-13T14:26:55   64415   Surf Internet, Southfield, MI   144.70   100.25   55.62
2025-07-12T20:47:45   1778   Comcast, Detroit, MI   1846.12   260.60   13.40
2025-06-28T09:38:00   53778   123NET, Southfield, MI   1472.07   346.66   14.02
2025-06-28T09:36:58   1778   Comcast, Detroit, MI   2310.24   351.93   13.62
2025-06-20T14:30:56   1778   Comcast, Detroit, MI   2304.69   349.62   13.60
2025-06-06T16:46:48   69076   Ripplefiber, Detroit, MI   2144.83   317.85   31.49
2025-06-06T16:45:58   6124   Merit Network, Inc, Southfield, MI   2204.34   347.66   26.89
2025-06-06T16:45:21   1778   Comcast, Detroit, MI   2279.34   352.00   13.46
2025-05-28T14:58:20   1778   Comcast, Detroit, MI   2311.01   339.32   13.22
2025-04-20T20:51:52   53778   123NET, Southfield, MI   1252.25   314.12   9.47
2025-04-20T20:47:59   1778   Comcast, Detroit, MI   2017.36   351.83   14

[pfry]

The difference is significant, but not on an obvious (bandwidth) boundary. A rather asymmetric service... DOCSIS? Comcast? I'd expect a (more) symmetric offering these days.

Not much to go on. Start by characterizing performance through the firewall (that is not clear here), then (as necessary) without it, watching for errors and/or packet loss.

[meyergru]

Did you actually try to keep the server ID fixed? Or try to use one of the fast Comcast servers now? Obviously, those were never used in 2026 for whatever reasons, thus the numbers are not comparable as such.