2.4.5R1 Update - Internet Speed Decrease After Update



  • Hello all,

    I am in the process of migrating my homelab from ESXI 6.7 to Hyper-V. I run Pfsense 2.4.5 in a VM on esxi and decided to upgrade that and my physical (backup) computer to 2.4.5P1 prior to migration.

    My backup dedicated computer is using ASRock Motherboard & CPU Combo Motherboards J3455B-ITX with Intel PRO/1000 Pt Dual Port Server Adapter Pfsense version 2.4.4 originally prior to the upgrading to 2.4.5P1. After I upgraded in the GUI, I noticed my internet speed was roughly 90Mbps, where I usually get 450Mbps.

    I activated my exsi Pfsense VM back up and it performed as expected providing 450Mbps.

    Is anyone interested in this? If so what details or supporting documents should I provide?



  • This is mine..

    Normal is 800+ both ways...

    26dc6052-1804-469a-808b-1199ed06d1d7-billede.png

    Load while testing

    11b18b03-5e0b-4d20-8141-57ebcfa8ba3a-billede.png

    Testing on another server. Danish backbone provider.

    1773312e-9d83-4bda-b1a2-ecb8f83337b2-billede.png



  • Finished my migration to Hyper-v 2019 internet speed is working great @450Mbps. the standalone still only ht 90ish Mbps



  • I reverted my stand alone box to 2.4.4 p3 and it works perfectly. Internet is back to 450Mbps. I visualize anyways so its not really affecting me but it is interesting the speed drop on 2.4.5p1



  • @Cool_Corona

    do you think, if there was such a slowdown after a update ...
    wouldn't this whole forum be full of disgruntled "posters" (members)?

    the fact is certainly not the versions that make the difference, but until you dont talk about your system, no one knows how it can help

    PS:
    nor is it the last thing that we are talking about virtual machines here

    edit++++:

    for example, on our dedicated hardwares for NGFW (not Hypervisor based / Dell R210II / Supermicro Epyc / Lenovo ThinkSystem SE350, HP ProLiant DL360 Gen9, APU 4D4 (for client) , etc.), I have not experienced for a years, such a degree of speed change



  • I really dont care. Its a fact with similar rules/IDS.

    And the 2.2.6 runs only 6 vCPU's whereis the test facility runs 16 vCPU.

    I can do nothing more than report my experience and what I see.

    And to be fair...

    Not many has low latency (0-1ms) connections but home fiber. Bandwidth is not the important factor but latency is.



  • Speed test is fine for me, I am on cable and pfSense is running in hyper-v.

    Capture.JPG



  • @Bob-Dig said in 2.4.5R1 Update - Internet Speed Decrease After Update:

    Speed test is fine for me, I am on cable and pfSense is running in hyper-v.

    Capture.JPG

    Suricata or snort enabled??



  • @Cool_Corona Yes.

    End of discussion, for me at least. 👋



  • @DaddyGo said in 2.4.5R1 Update - Internet Speed Decrease After Update:

    @Cool_Corona

    do you think, if there was such a slowdown after a update ...
    wouldn't this whole forum be full of disgruntled "posters" (members)?

    the fact is certainly not the versions that make the difference, but until you dont talk about your system, no one knows how it can help

    PS:
    nor is it the last thing that we are talking about virtual machines here

    edit++++:

    for example, on our dedicated hardwares for NGFW (not Hypervisor based / Dell R210II / Supermicro Epyc / Lenovo ThinkSystem SE350, HP ProLiant DL360 Gen9, APU 4D4 (for client) , etc.), I have not experienced for a years, such a degree of speed change

    Honestly, I am just sharing information to see if anyone was interested. Your point is well taken and only why I am sharing not shouting the sky is falling. my virtualized Pfsense is running like gangbusters and I am good. My stand alone backup PC that has the exact same settings and configuration with exception to the hardware used experienced a slowdown in internet speed. It works fine with 2.4.4p3 and 2.4.5 but not 2.4.5p1. Reverting back to 2.4.4p3 and internet speed is restored. This is using the exact same backup configuration file between my VM and stand alone PC. If no one is interested meh, im good just sharing.

    Note I asked what exact information would be needed if someone thought it could be an issue and since no one really seemed to see an issue, why my minimal effort into providing anything further without being asked.

    I have since shutdown the backup pc, I could reconnect it and get information off of it such as logs or what have you. Just let me know what is needed.



  • @sheen73

    Hi,
    sorry, if I did not state it clearly
    really, what I wrote, - I intended for "@Cool_Corona "

    we had a conversation in a few sentences in the past
    how sensible it is to run pfSense on a 16 core VM at home on a 10Gig capable NICs
    https://forum.netgate.com/topic/154450/sg-3100-pfsense-v2-4-4-p3-many-sudden-issues/15

    thank you for reporting your experience, this is normal behavior in a community affected thing ✋ THX again!

    for my part, this discussion is also closed

    PS: @Cool_Corona
    "Not many has low latency (0-1ms) connections but home fiber. Bandwidth is not the important factor but latency is."

    bb9ada0b-e149-485e-9704-9c22414017e9-image.png



  • I cannot confirm any performance degradation on 2.4.5_p1 compared to 2.4.4_p3. Both, Bare Metal and VM (on ESXi) performance is good at the latest Release.

    Main Site (Bare Metal pfSense) connected by Fiber (500Mbps down / 200 Mbps up)

    72c6aaba-1b48-43d6-9c11-9dc2cf520507-grafik.png

    Here a result of a Speedtest connected from Main Site and policy routed through IPsec to a VM hosted at OVH London, then NAT'ed out to a Test Server in London.

    6d1fe528-7ecf-4528-ba51-9b5494e58f4d-grafik.png

    Main Site has IDS/IPS running, VM at OVH doesn't.



  • @DaddyGo

    haha no worries I spaced on who it was directed to. Appreciate the response.



  • @sheen73

    Okay, I just wanted to know for sure... ✋



  • @Cool_Corona hi! Another Dane here ;-)
    Usually when I test up against 'Fiberby ApS' in speedtest.net I get several hundred Mbps.

    I can see (from the page of the result ID of your screenshot) that we use the same provider (Nianet and I guess now it must be GlobalConnect for both of us). But of course we both might have some intermediary ISP in between (I have at least).

    Please could I ask you to test up against 'Fiberby ApS' (and maybe rerun your original test again)? But only of course if you are currently running pfSense 2.4.5-p1 as I understand if you might needed to roll back to get full speed or do not have time for more tests atm.

    I haven't updated yet to 2.4.5-p1 due to some preparations I have to make.


Log in to reply