Is this server officially supported by Ubuntu?
and
It should be supported: https://github.com/torvalds/linux/bl...ealtek/Kconfig

I checked my running config and it looks enabled by default:
# zgrep 8169 /proc/config.gz
CONFIG_R8169=y

As some windows box you have did 1 gig over ethernet to some router which I hope you still have, why not
*repeat that test using the exact cable you are using on your linux box? Then with your linux box instead of the router?

Somewhere in the back of my head it stuck that you actually need Cat6 for gigabit connections. There is a difference, although folks refer to it all as cat 5. For my sins I suffered a hardware Electronic Engineering degree, which went through the material for CCNA 1-4. Unusual interference (1 year out of work, 7 as an invalid) and the usual knowledge Bulimia (learn for test, then forget it after) have looked after much of that. A few bits remain.

EDIT: I don't share your concern over modules or kernel versions That 8169.ko has been working for years for everyone else, and no bugs filed. It might get by in houses, where nobody might do 1 gigabit, but not in ISPs or businesses, where that would be noticed. I'm on kernel 5.15.63, btw.

So I don't appear to have that config.gz.

I am not sure what you mean is the server supported? In that link you gave under 8169 it did show 8168

So I did try the same cable with the windows box and again with the ethernet cable that came with the new router. It's not a cable problem. Fpr some reason the card only shows 100 as the Advertised rate from the router.

Not sure about the 8169. Just know one of the drivers I was using was removing 8169 and wanted it blacklisted in the kernel. Don't recall t he reason. It's moot now.

So I got a new card. I thought it was a different chipset maybe I was confused but i'm back to an 8168 card installed. Now I have an additional problem. Something changed on the server at some point and I don't know why. It was running and when I rebooted it was just in a loop. Booting uefi BTW. After some experimenting I had to turn secure boot. Why it suddenly required secure boot is beyond me. Anyway now it no longer loads 8168 at all. I;ve confirmed installation. Here's what I get

insmod: ERROR: could not insert module /lib/modules/5.15.0-52-generic/kernel/drivers/net/ethernet/realtek/r8168.ko: Operation not permitted

Lockdown: insmod: unsigned module loading is restricted; see man kernel_lockdown.7
Kernel is locked down from EFI Secure Boot mode; see man kernel_lockdown.7

How is r8168-dkms not delivering a signed module?

Changing a nic for one of the same type could be upsetting things if anything on the network is reading the MAC address as MAC adress=host pairings are common. Your kernel mightn't even be compiled with 100Gb support. Your router mightn't was 1Gb transfers for reasons of load balancing. Can you get >100Mb out from your router. Most installations have lower speeds.

At this stage, I'm thinking the problem is you. You couldn't solve this yourself, so you asked, which is fine. But if you give good feedback, do the work and keep digging yourself, most, but not all threads can be sorted within 8 posts. This is post #20. This should have been sorted in 20 posts. What have you learned from the previous 19 posts?
I'll bet you haven't

• read How_To_Ask_a_Question
• checked the logs or shown the output. It's one thing to say something 'errored out.' But the error number tells you why: error 1 = permissions; error 2 = file not found etc. That's diagnostic information. Why did the logs say the speed dropped?
• Spent time thinking on the problem and, for example, come up with a test to reveal more.

There are known good reasons why 1Gb connections decide on a lower speed. If you have a 500Mb or so upload/download, you know your modem has a large cache, the modem might choose to increase to 1 Gb periodically.

Here's one idea that came to me while I was calling you out. The windows box does 1G. The router also acts as a switch. Why not arrange a windows <--> linux transfer and time your transfer? On a network, it's about 10bits/byte. Work it out how fast you're going. It's how fast it is, that matters, not how dast it says it is. If I'm right about you, this test won't happen. But I really hope you do it. Not everything can be solved with point & click.

Maybe there is a config-* file in your /boot directory?
file /boot/* | grep ASCII
grep 8169 /boot/config-$(uname -r)

For server support see https://ubuntu.com/certified/servers

Im not sure what you're referencing or "calling me out"about. You said you bet I hadn't shown logs. I have extensively copied log output into my posts. I gave detailed information regarding kernel disy dmesg output ethtool output lsmod etc. I heard every reply suggesting something and I replied specifically to that. What other log information do you believe I have been missing? I didn;t just say it errored help me.

It's got nothing to do with the up/down of the internet provider. If you had looked at my postings in the beginning, you would have seen the ethtool output. I'm not speaking data transfer speed. I am speaking specifically of the link connect speed which as shown above is only negotiating 100 and is only showing the ADVERTISED speed from the router as 100 full duplex. As shown in the past below:

Settings for enp5s0:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supported pause frame use: Symmetric Receive-only
Supports auto-negotiation: Yes
Supported FEC modes: Not reported
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Advertised pause frame use: Symmetric Receive-only
Advertised auto-negotiation: Yes
Advertised FEC modes: Not reported
Link partner advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
Link partner advertised pause frame use: Symmetric
Link partner advertised auto-negotiation: Yes
Link partner advertised FEC modes: Not reported
Speed: 100Mb/s
Duplex: Full
Auto-negotiation: on
Port: Twisted Pair
PHYAD: 0
Transceiver: internal
MDI-X: Unknown
netlink error: Operation not permitted
Current message level: 0x00000033 (51)
drv probe ifdown ifup
Link detected: yes

Again as you can see it's not a transfer speed thing which you seem to assume with your win to linux test.

I tried it on a brand new fios router (which again was mentioned before) with 1g and 10g ports, same advertised speed

I tried a second router by tplink, which was mentioned above, and the same advertised speed. SO whats the misunderstanding here?

Agains your cute little "If I'm right about you, this test won't happen" about a win to lin test is irrelevent since we're talking about the link speed not actual data transfer.

So before you "Call Out" anyone you might want to look at the older posts and understand the problem properly, and see the supposedly missing logs are right there. You'd also note the replies and i tried this and 8169 that and whatever else.

So explain how with two cards (both 8168 unfortunately), multiple cables and two distinct routers the problem is me? It can only connect to what it knows and it is incorrectly detecting that the routers only support 100. Explain how that is user error and not some sort of driver or oob config error?

UPDATE - Just tried it on an Xfinity supplied router. Same troubles
Link partner advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full

What I have ALWAYS been asking about was the link speed.

So it's either failing to detect a correct speed or the driver is reporting something unexpected

So what precisely have I failed to do?

You should try and help people without being snide. If you want to be a jerk and make side comments or whatever you can always simply not press the reply button and move on. But what I see here that's just not you, right?

It's moot at this point. I have ordered a card based on other than a realtek chipset. Done trying. And I bet you know where you can put my extra card, right?

Damn... that went south fast. ¯\_(ツ)_/¯