Thunderbolt stopped working after recent update?

upve

New Member
Jun 29, 2024
14
3
3
solo10gsfp+ was working fine until recent update.

Output from dmesg
[ 1424.374480] thunderbolt 0-2: new device found, vendor=0x8 device=0x36
[ 1424.374486] thunderbolt 0-2: Sonnet Technologies, Inc Solo 10G SFP+ Thunderbolt 3 Edition

Output from journal -b -r
pve kernel: thunderbolt 0000:c9:00.5: 0:2: failed to reach state TB_PORT_UP. Ignoring port...
 
Ouch. Many people have been having issues with the recent 6.8 series of kernels.

Probably best to drop back to the older 6.5 kernel instead. There have been several threads about doing so in the forums over the past few weeks, so you should be able to find exact instructions fairly easily if you need to. :)
 
  • Like
Reactions: tomcat41 and upve
Ouch. Many people have been having issues with the recent 6.8 series of kernels.

Probably best to drop back to the older 6.5 kernel instead. There have been several threads about doing so in the forums over the past few weeks, so you should be able to find exact instructions fairly easily if you need to. :)
That was fast :)
Thanks for the info!
 
Wow, I've been scratching my head for 2 days trying to figure out what happend to my SONNET, and stumbled upon this post by accident. I'm seeing the exat same thing on my machine.
 
Wow, I've been scratching my head for 2 days trying to figure out what happend to my SONNET, and stumbled upon this post by accident. I'm seeing the exat same thing on my machine.
So far no official fix yet. Or no update of the package that resolve this yet.
 
So far no official fix yet. Or no update of the package that resolve this yet.
Wondering if there's a list of features that will be broken or potentially buggy with moving to an older kernel? Reverting to 6.5 worked and now I can use the nic.

Do you know how usually we can get notified if this is fixed so we can move back to the 6.8 kernel ? :)
 
Wondering if there's a list of features that will be broken or potentially buggy with moving to an older kernel? Reverting to 6.5 worked and now I can use the nic.

Do you know how usually we can get notified if this is fixed so we can move back to the 6.8 kernel ? :)
why upgrading.. you will put brand new hardware or so ? As for a test to install and commit to code sure, but a server is not a phone; no update for stability.
 
why upgrading.. you will put brand new hardware or so ? As for a test to install and commit to code sure, but a server is not a phone; no update for stability.
Not sure what you mean. Software updates are not always about new HW - stability, security are big part of linux kernel updates - instead of staying behind on a kernel that is no longer updated, I'd like to be able to update to latest again to make sure my promox machine is up to date with all the current stability and security patches possible.
 
For what it's worth, I use Thunderbolt for networking between Proxmox nodes. The last version of the 6.5 kernel (6.5.13-5-pve) works great. Anything in the 6.8 series of kernels causes all sorts of tries and slower throughputs when I use iperf3 across the Thunderbolt network. I hope they get this stuff fixed, I'd prefer to not stay on a previous major version.
 
Just curious has this been possible resolved? I was working on installing proxmox from scratch and have the Sonnettech twin25G adapters over usbc... I for some reason can't get them to work or even be recognized.
 
Just curious has this been possible resolved? I was working on installing proxmox from scratch and have the Sonnettech twin25G adapters over usbc... I for some reason can't get them to work or even be recognized.
Still not resolved by any official resources. Don't want to used old kernel either.
Though it seems to be Linux Kernel itself, not much to do with PVE itself.
 
Last edited:
Did you try as I suggested above?
It seems promising. As other Linux distros seems to having similar issue and used this method to get thunderbolt working.
But I'm concerned that it will have side effects in the future. So just kept waiting in hope it will be fixed form the root.
 
I have the same error:

[Mo, 28. Okt 2024, 14:42:46] ACPI: bus type thunderbolt registered
[Mo, 28. Okt 2024, 14:42:46] thunderbolt 0000:88:00.0: host router reset successful
[Mo, 28. Okt 2024, 14:42:46] thunderbolt 0000:88:00.0: device links to tunneled native ports are missing!
[Mo, 28. Okt 2024, 14:42:46] thunderbolt 0000:88:00.0: 0: DROM data CRC32 mismatch (expected: 0x51000000, got: 0x51e76a5f), continuing
[Mo, 28. Okt 2024, 14:42:49] thunderbolt 0-3: new device found, vendor=0x3d device=0x26
[Mo, 28. Okt 2024, 14:42:49] thunderbolt 0-3: CalDigit, Inc. TS4
[Mo, 28. Okt 2024, 14:42:50] thunderbolt 0-303: new device found, vendor=0x3d device=0x25
[Mo, 28. Okt 2024, 14:42:50] thunderbolt 0-303: CalDigit, Inc. Element Hub
[Mo, 28. Okt 2024, 14:42:50] thunderbolt 0-1: new device found, vendor=0x3d device=0x26
[Mo, 28. Okt 2024, 14:42:50] thunderbolt 0-1: CalDigit, Inc. TS4
[Mo, 28. Okt 2024, 14:42:51] thunderbolt 0-301: new device found, vendor=0x3d device=0x25
[Mo, 28. Okt 2024, 14:42:51] thunderbolt 0-301: CalDigit, Inc. Element Hub
[Mo, 28. Okt 2024, 14:42:52] thunderbolt 0-70301: new device found, vendor=0x8 device=0x34
[Mo, 28. Okt 2024, 14:42:52] thunderbolt 0-70301: Sonnet Technologies, Inc Solo 10G Thunderbolt 3 Edition
[Mo, 28. Okt 2024, 14:43:54] thunderbolt 0000:88:00.0: 0:1: failed to reach state TB_PORT_UP. Ignoring port...
[Mo, 28. Okt 2024, 14:43:54] thunderbolt 0000:88:00.0: 0:1: lost during suspend, disconnecting
[Mo, 28. Okt 2024, 14:43:55] thunderbolt 0000:88:00.0: 0:3: failed to reach state TB_PORT_UP. Ignoring port...
[Mo, 28. Okt 2024, 14:43:55] thunderbolt 0000:88:00.0: 0:3: lost during suspend, disconnecting
[Mo, 28. Okt 2024, 14:43:55] thunderbolt 0-70301: device disconnected
[Mo, 28. Okt 2024, 14:43:55] thunderbolt 0-301: device disconnected
[Mo, 28. Okt 2024, 14:43:55] thunderbolt 0-1: device disconnected
[Mo, 28. Okt 2024, 14:43:55] thunderbolt 0-303: device disconnected
[Mo, 28. Okt 2024, 14:43:55] thunderbolt 0-3: device disconnected

Any News here?
 
So the latest version, seems got this issue fixed.
PVE 8.3.1
pve-manager/8.3.1/fb48e850ef9dde27 (running kernel: 6.8.12-5-pve)
 
  • Like
Reactions: dakralex
So somehow, it stopped working again recently?
Anyone know what happened?
 
Last edited:

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!