ESXI: usb 1-1: reset high-speed USB device number 2 using ehci-pci


Recommended Posts

I am looking for some help on a trouble I am getting since going from 6-beta-12 to 6rc4. I've also tried rc5+rc6 with same results.

 

from syslog:

Jun 12 18:53:44 Tower kernel: usb 1-1: reset high-speed USB device number 2 using ehci-pci

Jun 12 18:53:44 Tower kernel: sd 0:0:0:0: [sda] 15633408 512-byte logical blocks: (8.00 GB/7.45 GiB)

this keeps repeating every 30-60seconds and doesnt stop.

 

I was running ESXI 5.5, and have also tried on ESXI 6.0 with same results.

 

I want to point the finger at UnRaid RC's(new 4.0.x kernel) but if I boot Unraid to bare metal(Supermicro X9SCM-IIF) I dont see the resets. I have seen this mentioned in a couple of threads but did not see any definite fix. 1 person changed their flash drives and it went away. I have 3 sandisk cruzer blades with pro keys and all three do it. just for a test i also tried a Sandisk fit with trial key and same results. I am getting this before the array is even started.

 

I can go back and get a syslog for 6b12 if that helps.

Is there a place to download all the different versions to test b14-15+rc1-3?

 

To narrow this down, can anyone with an X9SCM MB + ESXI let me know if they see this trouble or not. If not what type of setup do you have?

 

Thanks

-Dave

 

EDIT: A workaround for this trouble is to use a USB card reader instead of a USB Flash drive.

syslog-6-12-15_ESXI-flash_reset_error.txt

Link to comment

I am looking for some help on a trouble I am getting since going from 6-beta-12 to 6rc4. I've also tried rc5+rc6 with same results.

 

from syslog:

Jun 12 18:53:44 Tower kernel: usb 1-1: reset high-speed USB device number 2 using ehci-pci

Jun 12 18:53:44 Tower kernel: sd 0:0:0:0: [sda] 15633408 512-byte logical blocks: (8.00 GB/7.45 GiB)

this keeps repeating every 30-60seconds and doesnt stop.

 

I was running ESXI 5.5, and have also tried on ESXI 6.0 with same results.

 

I want to point the finger at UnRaid RC's(new 4.0.x kernel) but if I boot Unraid to bare metal(Supermicro X9SCM-IIF) I dont see the resets. I have seen this mentioned in a couple of threads but did not see any definite fix. 1 person changed their flash drives and it went away. I have 3 sandisk cruzer blades with pro keys and all three do it. just for a test i also tried a Sandisk fit with trial key and same results. I am getting this before the array is even started.

 

I can go back and get a syslog for 6b12 if that helps.

Is there a place to download all the different versions to test b14-15+rc1-3?

 

To narrow this down, can anyone with an X9SCM MB + ESXI let me know if they see this trouble or not. If not what type of setup do you have?

 

Thanks

-Dave

 

I'm getting the same error messages on my server as well.  Not only are the messages showing up in my unRAID logs but they are showing up in my VMware logs (vmkernal.log to be exact) as long as I have my unRAID USB drive connected to ANY active VM.  That means even if I shut down my unRAID VM and attach the USB to a Windows VM I get the same messages.  So it seems it's definitely an issue with something on the unRAID USB drive in relation to ESXi.

 

On top of the error messages, I'm also seeing terrible performance of the USB drive.  Doing something as simple as a change directory from the console or clicking different links in the webgui take 30+ seconds.  It's unbearable.

 

For me it started once I upgraded from 6b14 to RC3 and it hasn't changed all the way up to RC6.  I've asked in the previous release's threads if I could get a download for an earlier version so I could test it but haven't gotten a response.

 

P.S. If you want to confirm if these messages are showing up in your VMware logs as well, just type the following into your browser and open the vmkernel.log file:  https://"IPAddressOfESXiHost"/host.

Link to comment

I am looking for some help on a trouble I am getting since going from 6-beta-12 to 6rc4. I've also tried rc5+rc6 with same results.

 

from syslog:

Jun 12 18:53:44 Tower kernel: usb 1-1: reset high-speed USB device number 2 using ehci-pci

Jun 12 18:53:44 Tower kernel: sd 0:0:0:0: [sda] 15633408 512-byte logical blocks: (8.00 GB/7.45 GiB)

this keeps repeating every 30-60seconds and doesnt stop.

 

I was running ESXI 5.5, and have also tried on ESXI 6.0 with same results.

 

I want to point the finger at UnRaid RC's(new 4.0.x kernel) but if I boot Unraid to bare metal(Supermicro X9SCM-IIF) I dont see the resets. I have seen this mentioned in a couple of threads but did not see any definite fix. 1 person changed their flash drives and it went away. I have 3 sandisk cruzer blades with pro keys and all three do it. just for a test i also tried a Sandisk fit with trial key and same results. I am getting this before the array is even started.

 

I can go back and get a syslog for 6b12 if that helps.

Is there a place to download all the different versions to test b14-15+rc1-3?

 

To narrow this down, can anyone with an X9SCM MB + ESXI let me know if they see this trouble or not. If not what type of setup do you have?

 

Thanks

-Dave

 

i was running unRAID v6b15 with ESXI 5.5 on X9SCM-iif for a long time without any problems, just switched my motherboard to x8DTH-6F this week and no problems with USB with this mobo too, just looked at my syslog to confirm..

if this is an ESXI problem, i'm running esxi 5.5 initial release(ESXi-5.5.0-1331820-standard), not upgraded it jet since first install. 

Link to comment

I went to "vmkernel.log file:  https://"IPAddressOfESXiHost"/host" and it does show the resets also for 6rc6, but not with 6b12.  Thanks for the info that its happening on you X10 MB's, at least i know its not tied to just the X9SCM. That should make a larger pool of people that may see this trbl and narrow down the cause.

 

uldise - thanks for verifying 6b15 on the X9SCM. Have you tried any of the 6RC series on either board yet?

 

I also found an old(2009) thread http://lime-technology.com/forum/index.php?topic=4877.msg45327#msg45327 which recommended:

echo 64 >/sys/block/sdX/device/max_sectors

but that did nothing for me. I do think its just a simple setting like that causing it, but I just don't know enough about it.

-Dave

 

Link to comment

uldise - thanks for verifying 6b15 on the X9SCM. Have you tried any of the 6RC series on either board yet?

-Dave

 

No, i think i will wait for some time to v6 final or more with production server upgrade, not decided jet :)

 

i can setup test server with x9SCM-iif, but this will happen in the middle of the next week maybe, i'm not at my lab now..

but i'm sure, you can find v6b15 install(on limetech git? ) and give it a try..

 

Uldis

 

Link to comment

I found I had a backup copy of 6b15 on one of my other flash drives that updated from the GUI(Nice feature Lime-tech!). I tested it out and it worked no USB resets on ESXI 5.5 or 6.0.

 

From the change logs the kernel 4.0.x started in 6RC3 -which is giving me the resets. I cant find a copy of 6RC1 or 2 in the announcement threads, not sure if they were released to public or just in-house. If someone does have a copy of 6RC2 I would like to test it even though i'm 99% sure its something in the 4.0. kerenl that is wrong.

 

I hope someone that knows the kernel config can come up with a change that will fix the 4.0.x trouble we're seeing.

Is there an EASY way to overide the 6RCx usb driver with the one from the beta15?

 

Thanks.

Link to comment

fyi, I sent an email to "[email protected]" to bring attention to this. Hopefully there is an easy fix to make it into 6.0 final.

see here

 

I was not seeing this when I was running as a VMDK, but swapped to PLOP to enjoy the easier updating and I see this message.

 

And yet I'm seeing it while running as a VMDK (never used PLOP) so it's pretty save to say the issue isn't isolated to one or the other.  Makes sense that it's not since I get the same messages when I connect the unRAID USB to a Windows VM as well.

 

BTW, since you've started seeing these messages have you noticed any performance issues in the webgui at all?

Link to comment

fyi, I sent an email to "[email protected]" to bring attention to this. Hopefully there is an easy fix to make it into 6.0 final.

see here

 

I was not seeing this when I was running as a VMDK, but swapped to PLOP to enjoy the easier updating and I see this message.

 

And yet I'm seeing it while running as a VMDK (never used PLOP) so it's pretty save to say the issue isn't isolated to one or the other.  Makes sense that it's not since I get the same messages when I connect the unRAID USB to a Windows VM as well.

 

BTW, since you've started seeing these messages have you noticed any performance issues in the webgui at all?

 

Honestly.. the WebGUI for me is slow/fast at random times.  One time it will take 2 minutes to load a page, others its instant.  So I don't know if its PLOP vs VMDK vs Bare Metal.  Its inconsistent everywhere as far as speed.

Link to comment

fyi, I sent an email to "[email protected]" to bring attention to this. Hopefully there is an easy fix to make it into 6.0 final.

see here

 

I was not seeing this when I was running as a VMDK, but swapped to PLOP to enjoy the easier updating and I see this message.

 

And yet I'm seeing it while running as a VMDK (never used PLOP) so it's pretty save to say the issue isn't isolated to one or the other.  Makes sense that it's not since I get the same messages when I connect the unRAID USB to a Windows VM as well.

 

BTW, since you've started seeing these messages have you noticed any performance issues in the webgui at all?

 

Honestly.. the WebGUI for me is slow/fast at random times.  One time it will take 2 minutes to load a page, others its instant.  So I don't know if its PLOP vs VMDK vs Bare Metal.  Its inconsistent everywhere as far as speed.

 

It was never that way for me on bare metal and my backup server is still bare metal and is never slow in the webgui.  But ever since I setup my main as a guest in VMware and started seeing these error messages, it's painfully slow at times.

Link to comment

I do agree that it seems slower. I don't think the usb is used after the initial boot. It all stays in ram, except to write config changes.  I figured the slowness has to do with the never ending reads on the usb drive causing the VM to struggle(just a guess).

 

For some reason while testing a few things, the usb completely disappeared from ESXI 6, had to unplug and  plug back in to be seen by any VM. It's very odd, I wouldn't think a usb in a VM could affect the host this much.

 

I have been booting with PLOP. Will try to get time to try with VMDK.

 

 

Link to comment

Trying something out. I unattached the Unraid USB from the VM and removed the USB controller; I added in a XHCI USB controller instead of the EHCI+UHCI USB controller and re-attached my Unraid USB to the VM.

 

Plop does not support booting from XHCI devices so I used PlopKexec (same author as Plop) and was able to boot off the USB. It actually boots A LOT faster than Plop, like ~30 seconds vs the 5 minutes of Plop.

 

So far I have not seen any of these USB reset errors. I'll report back tomorrow with more results.

Link to comment

Trying something out. I unattached the Unraid USB from the VM and removed the USB controller; I added in a XHCI USB controller instead of the EHCI+UHCI USB controller and re-attached my Unraid USB to the VM.

 

Plop does not support booting from XHCI devices so I used PlopKexec (same author as Plop) and was able to boot off the USB. It actually boots A LOT faster than Plop, like ~30 seconds vs the 5 minutes of Plop.

 

So far I have not seen any of these USB reset errors. I'll report back tomorrow with more results.

 

Nice find!! Did you just use the default iso? I don't see a GUI builder like for PLOP.

Link to comment

Trying something out. I unattached the Unraid USB from the VM and removed the USB controller; I added in a XHCI USB controller instead of the EHCI+UHCI USB controller and re-attached my Unraid USB to the VM.

 

Plop does not support booting from XHCI devices so I used PlopKexec (same author as Plop) and was able to boot off the USB. It actually boots A LOT faster than Plop, like ~30 seconds vs the 5 minutes of Plop.

 

So far I have not seen any of these USB reset errors. I'll report back tomorrow with more results.

 

Nice find!! Did you just use the default iso? I don't see a GUI builder like for PLOP.

 

I used the default iso.

 

I spoke too soon; I see the following errors:

Jun 15 22:53:13 DeathStar kernel: usb 1-2.1: reset high-speed USB device number 5 using xhci_hcd

Jun 15 22:53:28 DeathStar kernel: usb 1-2.1: device descriptor read/64, error -110

 

I read that the -110 error means that the USB device is trying to pull too much power from the USB bus, something very strange is going on and I am definitely thinking we need to point fingers at the new kernel.

Link to comment

uldise - thanks for verifying 6b15 on the X9SCM. Have you tried any of the 6RC series on either board yet?

-Dave

 

No, i think i will wait for some time to v6 final or more with production server upgrade, not decided jet :)

 

i can setup test server with x9SCM-iif, but this will happen in the middle of the next week maybe, i'm not at my lab now..

but i'm sure, you can find v6b15 install(on limetech git? ) and give it a try..

 

Uldis

 

ok, today i assembled back my test server with x9SCM-iif.

esxi6 with unRAID 6b15 are working ok, then upgrade to unRAID V6 final and full syslog with similar messages as OP.

and unRAID then is so unresponsive.. so i'm staying on b15 for a while....

 

Link to comment

So, yall are seeing this error in your unRAID syslog?  I am not, but this (below) appeared twice in the last week in my vmkernel log.

 

This doesn't seem to be causing me an issues.

 

2015-06-17T10:33:25.989Z cpu2:35855)<6>usb 2-1.1: device is available for passthrough
2015-06-17T10:33:26.072Z cpu2:35855)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd
2015-06-17T10:33:26.299Z cpu1:35855)<6>usb 2-1.1: Device is allocated for USB passthrough use; not available for VMkernel use
2015-06-17T10:33:26.300Z cpu5:33152)<6>hub 2-1:1.0: resumed
2015-06-17T10:33:26.300Z cpu5:33152)<6>hub 2-1:1.0: suspended
2015-06-17T10:33:41.418Z cpu5:33152)<6>hub 2-1:1.0: resumed
2015-06-17T10:33:41.418Z cpu5:33152)<6>hub 2-1:1.0: suspended
2015-06-17T10:33:41.444Z cpu2:35858)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd
2015-06-17T10:33:45.851Z cpu0:35645)FSS: 5327: No FS driver claimed device 'control': No filesystem on the device
2015-06-17T10:33:45.859Z cpu0:35645)VC: 3551: Device rescan time 9 msec (total number of devices 7)
2015-06-17T10:33:45.859Z cpu0:35645)VC: 3554: Filesystem probe time 11 msec (devices probed 4 of 7)
2015-06-17T10:33:45.859Z cpu0:35645)VC: 3556: Refresh open volume time 4 msec
2015-06-17T10:33:56.918Z cpu7:35858)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd
2015-06-17T10:33:57.034Z cpu5:33152)<6>hub 2-1:1.0: resumed
2015-06-17T10:33:57.034Z cpu5:33152)<6>hub 2-1:1.0: suspended
2015-06-17T10:33:57.393Z cpu3:35858)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd
2015-06-17T10:33:58.570Z cpu5:33152)<6>hub 2-1:1.0: resumed
2015-06-17T10:33:58.570Z cpu5:33152)<6>hub 2-1:1.0: suspended
2015-06-17T10:34:13.864Z cpu6:35858)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd
2015-06-17T10:34:13.930Z cpu5:33152)<6>hub 2-1:1.0: resumed
2015-06-17T10:34:13.930Z cpu5:33152)<6>hub 2-1:1.0: suspended
2015-06-17T10:34:29.290Z cpu5:33152)<6>hub 2-1:1.0: resumed
2015-06-17T10:34:29.290Z cpu5:33152)<6>hub 2-1:1.0: suspended
2015-06-17T10:34:29.337Z cpu3:35858)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd
2015-06-17T10:34:29.803Z cpu6:33152)<6>hub 2-1:1.0: resumed
2015-06-17T10:34:29.803Z cpu6:33152)<6>hub 2-1:1.0: suspended
2015-06-17T10:34:29.812Z cpu3:35858)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd
2015-06-17T10:34:32.449Z cpu1:35855)<6>usb 2-1.1: device is available for passthrough

Link to comment

I'm not getting those errors.  ESXi 5.5, unRAID 6, vmdk boot method.  Build details in my sig.

 

If we want to get to the root cause of this, we may need to start attaching screen shots of the unRAID VM config within the vpshere client.  I can't do it from work, but if needed, I'll post it later today.

Link to comment

So, yall are seeing this error in your unRAID syslog?  I am not, but this (below) appeared twice in the last week in my vmkernel log.

 

Yes, those messages are on unRAID syslog - and keep repeating every 30 secs..

Quoted again myself - i can see those messages in vmkernel.log too..

 

2015-06-17T13:17:23.385Z cpu2:41053)<6>usb 2-1.1: device is available for passthrough

2015-06-17T13:17:23.469Z cpu2:41053)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd

2015-06-17T13:17:23.591Z cpu6:41053)<6>usb 2-1.1: Device is allocated for USB passthrough use; not available for VMkernel use

2015-06-17T13:17:23.591Z cpu6:33149)<6>hub 2-1:1.0: resumed

2015-06-17T13:17:23.591Z cpu6:33149)<6>hub 2-1:1.0: suspended

2015-06-17T13:17:23.790Z cpu6:41053)<6>usb 2-1.1: reset high speed USB device number 3 using ehci_hcd

 

Link to comment

I tried VMDK boot and still seeing the errors(only left the config folder on the usb drive).

my VM config is very generic using VM version 8 and add echi usb controller+device, dont even need to add the sata cntlr to the vm. About a minute after boot completes the errors start -dont even need to start the array.

 

kaiguy - if you get a chance can you create a test vm with a plop boot and check unraid and vmkernel.log?

              Also do you have any patches applied to your ESXI 5.5?

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.