[Nut-upsuser] Tripp-Lite USB constantlly disconnecting.

Mon Sep 22 15:14:23 UTC 2014

Charles,
   So I installed a minimal CentOS 6.5 installation on the Fedora box this
morning and I got the same issue.

Below are the kernel versions used.

CentOS 6.5 Minimal -
[root at nemo tmp]# uname -a
Linux nemo 2.6.32-431.el6.x86_64 #1 SMP Fri Nov 22 03:15:09 UTC 2013 x86_64
x86_64 x86_64 GNU/Linux

[root at nemo tmp]# rpm -qa | grep nut
nut-2.6.5-2.el6.x86_64
nut-client-2.6.5-2.el6.x86_64

CentOS 6.5 -
[root at localhost ~]# uname -a
Linux localhost.localdomain 2.6.32-431.23.3.el6.x86_64 #1 SMP Thu Jul 31
17:20:51 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

[root at localhost ~]# rpm -qa | grep nut
nut-client-2.6.5-2.el6.x86_64
nut-2.6.5-2.el6.x86_64

Fedora 20 -
[root at nemo ~]# uname -a
Linux nemo 3.11.10-301.fc20.x86_64 #1 SMP Thu Dec 5 14:01:17 UTC 2013
x86_64 x86_64 x86_64 GNU/Linux

[root at nemo ~]# rpm -qa | grep nut
nut-2.7.2-1.fc20.x86_64
nut-client-2.7.2-1.fc20.x86_64

As mentioned earlier I have tried another USB port and cable. Below are the
/var/log/messages - I also got a new behavior this time.

Sep 22 10:35:50 nemo upsd[12011]: Data for UPS [upsunit] is stale - check
driver
Broadcast message from nut at nemo (Mon Sep 22 10:52:15 2014):.0.1] failed -
Data stale

Sep 22 10:36:34 nemo kernel: usb 8-1: USB disconnect, device number 112
Sep 22 10:36:34 nemo kernel: usb 8-1: new low speed USB device number 113
using uhci_hcd
Sep 22 10:36:35 nemo kernel: usb 8-1: New USB device found, idVendor=09ae,
idProduct=3015
Sep 22 10:36:35 nemo kernel: usb 8-1: New USB device strings: Mfr=2,
Product=3, SerialNumber=4
Sep 22 10:36:35 nemo kernel: usb 8-1: Product: TRIPP LITE SMART1500RM2UN
Sep 22 10:36:35 nemo kernel: usb 8-1: Manufacturer: Tripp Lite
Sep 22 10:36:35 nemo kernel: usb 8-1: SerialNumber: 2424AY0SM882300229
Sep 22 10:36:35 nemo kernel: usb 8-1: configuration #1 chosen from 1 choice
Sep 22 10:36:35 nemo upsmon[12015]: Poll UPS [upsunit at 127.0.0.1] failed -
Data stale
Sep 22 10:36:35 nemo kernel: generic-usb 0003:09AE:3015.0071:
hiddev96,hidraw2: USB HID v1.10 Device [Tripp Lite  TRIPP LITE
SMART1500RM2UN ] on usb-0000:00:1d.2-1/input0
Sep 22 10:36:39 nemo kernel: usb 8-1: USB disconnect, device number 113
Sep 22 10:36:40 nemo upsmon[12015]: Poll UPS [upsunit at 127.0.0.1] failed -
Data stale
Sep 22 10:36:45 nemo upsmon[12015]: Poll UPS [upsunit at 127.0.0.1] failed -
Data stale
Sep 22 10:36:50 nemo upsmon[12015]: Poll UPS [upsunit at 127.0.0.1] failed -
Data stale
Sep 22 10:36:55 nemo upsmon[12015]: Poll UPS [upsunit at 127.0.0.1] failed -
Data stale
Sep 22 10:36:56 nemo kernel: usb 7-1: new low speed USB device number 2
using uhci_hcd
Sep 22 10:36:56 nemo kernel: usb 7-1: New USB device found, idVendor=09ae,
idProduct=3015
Sep 22 10:36:56 nemo kernel: usb 7-1: New USB device strings: Mfr=2,
Product=3, SerialNumber=4
Sep 22 10:36:56 nemo kernel: usb 7-1: Product: TRIPP LITE SMART1500RM2UN
Sep 22 10:36:56 nemo kernel: usb 7-1: Manufacturer: Tripp Lite
Sep 22 10:36:56 nemo kernel: usb 7-1: SerialNumber: 2424AY0SM882300229
Sep 22 10:36:56 nemo kernel: usb 7-1: configuration #1 chosen from 1 choice
Sep 22 10:36:57 nemo kernel: generic-usb 0003:09AE:3015.0072:
hiddev96,hidraw2: USB HID v1.10 Device [Tripp Lite  TRIPP LITE
SMART1500RM2UN ] on usb-0000:00:1d.1-1/input0
Sep 22 10:37:00 nemo upsmon[12015]: Poll UPS [upsunit at 127.0.0.1] failed -
Data stale
Sep 22 10:37:05 nemo upsmon[12015]: Poll UPS [upsunit at 127.0.0.1] failed -
Data stale
Sep 22 10:37:05 nemo upsd[12011]: UPS [upsunit] data is no longer stale
Sep 22 10:37:10 nemo upsmon[12015]: Communications with UPS
upsunit at 127.0.0.1 established
Sep 22 10:37:10 nemo wall[12035]: wall: user nut broadcasted 1 lines (55
chars)

[root at nemo tmp]#
Broadcast message from nut at nemo (Mon Sep 22 10:40:00 2014):

Communications with UPS upsunit at 127.0.0.1 lost

These machines are exact replicas of each other and are fairly old and have
a Core 2 Duo E8400 CPU in them. Is there anything else you need or anything
I can do to help fix this? Thanks!

Shade

On Sun, Sep 21, 2014 at 1:12 PM, Shade Alabsa <shade34321 at gmail.com> wrote:

> Soon as I get back to that computer I will let you know which kernel
> version. Both were just recently installed, Fedora was installed
> specifically for this actually, though updates have been applied.
>
> I've tried it with other USB ports already across 5 different machines.
> I've also swapped out the USB cable for another one that I knew worked fine.
>
> I'll see what I can do about installing a minimal build onto these
> machines to see if something changes.
>
> Thanks!
> Shade
>
> On Sat, Sep 20, 2014 at 8:47 AM, Charles Lepple <clepple at gmail.com> wrote:
>
>> On Sep 19, 2014, at 7:48 PM, Shade Alabsa <shade34321 at gmail.com> wrote:
>>
>> > So I was wondering does NUT do anything which would cause the USB to
>> disconnect?
>>
>> Not intentionally, no. The reconnection code is meant to clean up after
>> these disconnection events, but it is something that isn't expected all
>> that frequently.
>>
>> > I didn't think so and I couldn't find anything yet it's quite possible
>> I missed it. I've tested this with CentOS 6.5 and Fedora 20. With CentOS we
>> were using nut-client-2.6.5-2.el16.x86_64 and nut-2.6.5-2.el6.x86_64. With
>> Fedora we were using nut-2.7.2-1.fc20.x86_64 and
>> nut-client-2.7.2-1.fc20.x86_64.
>>
>> What kernel versions?
>>
>> We have had a number of recent reports of issues with USB 3.0 host
>> controllers. While it seems like your UPS is getting picked up by the
>> uhci_hcd driver, you might want to check if another USB port works better.
>> Also, make sure you are using a decent-quality cable (no passive USB
>> extenders, etc.).
>>
>> I am also slightly suspicious of mtp-probe. We haven't done any testing
>> with other programs trying to access the UPS at the same time, but there is
>> some code in NUT to try and catch two NUT drivers from tripping over each
>> other. But it only shows up in the Fedora output, and its messages seem
>> closer to when the device is attached, rather than when it disconnects. I'm
>> not sure if Fedora or CentOS have "server" builds, but I would recommend
>> starting from the minimum software necessary and building up.
>>
>> --
>> Charles Lepple
>> clepple at gmail
>>
>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140922/39ac7283/attachment.html>