10Gb network card


Recommended Posts

7 hours ago, alturismo said:

should a ubuntu live do it to see ? from USB ?

i think so.. i have not this card, but i recently added Brocade BR-1010 (10Gbit SFP+) to my main Ubuntu workstation and it detects it correctly without any additional drivers needed. 

Link to comment

ok, live usb running, also not loaded ... access denied in Capabilities, so also no driver loaded

 

also same, router says connected (pic) ... so hardware seems ok, no driver support.

 

03:00.0 Ethernet controller: Intel Corporation Device 0001 (rev 01)
    Subsystem: Device 1b52:10fb
    Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
    Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
    Latency: 0, Cache Line Size: 64 bytes
    Interrupt: pin A routed to IRQ 11
    Region 0: Memory at 2000000000 (64-bit, prefetchable)
    Region 2: I/O ports at d020
    Region 4: Memory at 2000100000 (64-bit, prefetchable)
    Expansion ROM at df580000 [disabled]
    Capabilities: <access denied>

03:00.1 Ethernet controller: Intel Corporation Device 0001 (rev 01)
    Subsystem: Device 1b52:10fb
    Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
    Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
    Latency: 0, Cache Line Size: 64 bytes
    Interrupt: pin B routed to IRQ 10
    Region 0: Memory at 2000080000 (64-bit, prefetchable)
    Region 2: I/O ports at d000
    Region 4: Memory at 2000304000 (64-bit, prefetchable)
    Expansion ROM at df500000 [disabled]
    Capabilities: <access denied>

 

Bildschirmfoto vom 2017-06-11 06-52-37.png

Edited by alturismo
Link to comment
58 minutes ago, uldise said:
are you running this command with sudo?

 


the modprobe command, yes

the lspci -v not

now sudo lspci -v

 

i added the onboard ethernet in paste, there is the kernel driver loaded, but not for the intel one,

i guess i just have no luck with NICs ...

 

03:00.0 Ethernet controller: Intel Corporation Device 0001 (rev 01)
    Subsystem: Device 1b52:10fb
    Flags: bus master, fast devsel, latency 0, IRQ 11
    Memory at 2000000000 (64-bit, prefetchable)
    I/O ports at d020
    Memory at 2000100000 (64-bit, prefetchable)
    Expansion ROM at df580000 [disabled]
    Capabilities: [40] Power Management version 3
    Capabilities: [50] MSI: Enable- Count=1/1 Maskable+ 64bit+
    Capabilities: [70] MSI-X: Enable- Count=64 Masked-
    Capabilities: [a0] Express Endpoint, MSI 00
    Capabilities: [e0] Vital Product Data
    Capabilities: [100] Advanced Error Reporting
    Capabilities: [140] Device Serial Number 00-25-90-ff-ff-97-10-98
    Capabilities: [150] Alternative Routing-ID Interpretation (ARI)
    Capabilities: [160] Single Root I/O Virtualization (SR-IOV)

03:00.1 Ethernet controller: Intel Corporation Device 0001 (rev 01)
    Subsystem: Device 1b52:10fb
    Flags: bus master, fast devsel, latency 0, IRQ 10
    Memory at 2000080000 (64-bit, prefetchable)
    I/O ports at d000
    Memory at 2000304000 (64-bit, prefetchable)
    Expansion ROM at df500000 [disabled]
    Capabilities: [40] Power Management version 3
    Capabilities: [50] MSI: Enable- Count=1/1 Maskable+ 64bit+
    Capabilities: [70] MSI-X: Enable- Count=64 Masked-
    Capabilities: [a0] Express Endpoint, MSI 00
    Capabilities: [e0] Vital Product Data
    Capabilities: [100] Advanced Error Reporting
    Capabilities: [140] Device Serial Number 00-25-90-ff-ff-97-10-98
    Capabilities: [150] Alternative Routing-ID Interpretation (ARI)
    Capabilities: [160] Single Root I/O Virtualization (SR-IOV)

05:00.0 Ethernet controller: Qualcomm Atheros Device e0b1 (rev 10)
    Subsystem: Micro-Star International Co., Ltd. [MSI] Device 7a78
    Flags: bus master, fast devsel, latency 0, IRQ 19
    Memory at df400000 (64-bit, non-prefetchable)
    I/O ports at c000
    Capabilities: [40] Power Management version 3
    Capabilities: [58] Express Endpoint, MSI 00
    Capabilities: [c0] MSI: Enable- Count=1/16 Maskable+ 64bit+
    Capabilities: [d8] MSI-X: Enable+ Count=16 Masked-
    Capabilities: [100] Advanced Error Reporting
    Capabilities: [180] Device Serial Number ff-b3-82-ac-4c-cc-6a-ff
    Kernel driver in use: alx
    Kernel modules: alx

 

 

Edited by alturismo
added sudo paste from ubuntu live
Link to comment

thanks for the links, but thats not my skill to check the unraid kernel if that driver is loaded or if its in the right place ;)

 

also i dont see any error message that the sfp module would be unsupported ... all i see in syslog is the modprobe command ... but no respond or anything.

 

so i ll try to get another replacement with an original not branded Intel card, may that will help if its not from supermicro ... cant really imagine that there should be any difference,

but who knows ;)

 

thanks again for the advices, sadly no luck here

Link to comment

tried now all kinda usb live dists ... none really working properly on pc, i assume the 2 GPU´s are not nice for live ;)

 

stick working fine on laptop ... i ll check again when i get the stick running.

 

when i check your link, all i see is the modprobe sfp edit wich i already tried, anything else i could test in ubuntu enviroment when im in there ?

 

i get the stick running without persistent mode, so i could check some more exept the modprobe ...

Link to comment

This issue illustrates a problem we run into all the time.  Many vendors insist on maintaining their own open source branches of drivers which already have open source linux in-tree equivalents.  The latest Intel 10Gbit PCIe driver (called ixgbe) on the Intel download site is version 5.0.4 dated 2/1/2017.  However this driver will not compile on the linux 4.11 kernel.  It ges an error:

 

/usr/src/drivers/intel/ixgbe-5.0.4/src/ixgbe_main.c:1913:10: error: ‘LL_FLUSH_FAILED’ undeclared (first use in this function); did you mean ‘MF_FAILED’?
   return LL_FLUSH_FAILED;
          ^~~~~~~~~~~~~~~

This is due to a change in the linux kernel introduced in 4.11:

https://patchwork.ozlabs.org/patch/726542/

 

The way development works in the linux kernel is this: it's "ok" to change internal API's, but the developer has the responsibility to fix it in all kernel code.  If Intel maintained their "latest" drivers in the linux kernel, then this "bug" would have automatically been fixed by a linux kernel developer.

 

This is the reason why our policy is "no out of tree drivers".  This is because, if we needed to move to the latest kernel for some other reason (such as xfs/btrfs fixes), we couldn't because we wouldn't be able to compile the kernel.

 

I'm going to say "probably" this latest Intel driver will fix your problem, however we have to wait for one of two things.  Either enough people have to bitch at Intel to get them to fix their driver so it will compile on 4.11, or we have to wait for the in-tree ixgbe driver to be modified to support the latest Intel devices.  But I have to say, I'm not too crazy about the first choice because historically Intel is dog-slow to fix issues due to new kernel changes.

 

Anyway, I don't see why Intel just doesn't maintain the in-tree driver, seems idiotic to me for them not to.  Same with a lot of other vendor drivers (looking at you Realtek, and you too Highpoint and LSI).

 

  • Upvote 1
Link to comment

@limetech, thanks for the info, so i can stop trying atm.

 

just to clarify for me, when i read this now correctly, currently no intel will probably work, also not an "unbranded" one.

 

another option would be the Synology E10G15-F1, its the same as Tehuti TN9310: 10GbE SFP+, just different color.

 

may i ask about the brocade one (i had before but also no driver support), may gets added in 6.4 ?

roughly i already bought me passive cables and i think the brocade only runs with active ;)

 

cause the mellanox ones ... are not compatible with my board, system stalls on boot completely.

 

thanks ahead.

Link to comment
1 minute ago, alturismo said:

just to clarify for me, when i read this now correctly, currently no intel will probably work, also not an "unbranded" one.

 

Answer is: I don't know.  That's just my "guess" that there is something in the driver that is preventing from being loaded with a non-Intel SPI - but I could be wrong.  The reason I suggested to try a standard linux distro is that you might be able to find a kernel version/driver combination that works for your controller.

  • Upvote 1
Link to comment

thanks, i ll rather stay on 1500 ... dunno if all clients etc can take 9000 properly.

 

now last step is somehow get the card boot faster, takes > 2min to pass init screen, no real big deal cause server shouldnt reboot to often, but ... ;)

 

thanks again for all the advices and help.

Link to comment
  • 1 year later...

Apologies for the recro bump. But to close the loop on the OP's original request, with the most current at the moment (6.7.0 rc2), the Supermicro AOC-STGN-I2S card works fine (version 2.0). I swapped from an X540-T2 to this today since I am using SFP+ adapters. Which by the way, the X540-T2 also worked fine.

Edited by cybrnook
Added some back story
  • Like 1
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.