Just with last update, new kernel, made my Logitech K520 do strange things. Every now and then writes alone number 5 to open window, like skype etc. Anyone else noticed?
Jarmo
On 04-03-16 07:56, jarmo wrote:
Just with last update, new kernel, made my Logitech K520 do strange things. Every now and then writes alone number 5 to open window, like skype etc. Anyone else noticed?
I have seen that too with my Logitech K520 keyboard. Not only with the 4.4.3-300 kernel but also with the older ones. Hitting the 5 key usually makes it stop. Disabling "Repeat Keys" in the Keyboard config (in Gnome) seems to make it stop. Haven't had time to further investigate. If you find out what's causing it please update this thread.
Cheers, Patrick
Fri, 4 Mar 2016 09:44:02 +0100 Patrick Laimbock patrick@laimbock.com kirjoitti:
I have seen that too with my Logitech K520 keyboard. Not only with the 4.4.3-300 kernel but also with the older ones. Hitting the 5 key
Yes, same with me, couple of older kernels wrote number 9 :) Hope somebody finds reason...
Jarmo
On Fri, Mar 04, 2016 at 09:44:02AM +0100, Patrick Laimbock wrote:
On 04-03-16 07:56, jarmo wrote:
Just with last update, new kernel, made my Logitech K520 do strange things. Every now and then writes alone number 5 to open window, like skype etc. Anyone else noticed?
I have seen that too with my Logitech K520 keyboard. Not only with the 4.4.3-300 kernel but also with the older ones. Hitting the 5 key usually makes it stop. Disabling "Repeat Keys" in the Keyboard config (in Gnome) seems to make it stop. Haven't had time to further investigate. If you find out what's causing it please update this thread.
It seems to be a recurring defect. Some kernel releases exhibit it others do not. Last I looked in to it (maybe 9 months ago) it wasn't solved and was being called the Marching Fives.
jl
Fri, 04 Mar 2016 11:21:14 -0500 Jon LaBadie jonfu@jgcomp.com kirjoitti:
It seems to be a recurring defect. Some kernel releases exhibit it others do not. Last I looked in to it (maybe 9 months ago) it wasn't solved and was being called the Marching Fives.
jl
Then I have met Marching Nines also :D
Jarmo
On 03/04/2016 09:04 AM, jarmo wrote:
Fri, 04 Mar 2016 11:21:14 -0500 Jon LaBadie jonfu@jgcomp.com kirjoitti:
It seems to be a recurring defect. Some kernel releases exhibit it others do not. Last I looked in to it (maybe 9 months ago) it wasn't solved and was being called the Marching Fives.
jl
Then I have met Marching Nines also :D
You chaps aren't using the keyboards with KVMs, are you? I've seen lots of really strange behaviors with KVMs and different kernels. ---------------------------------------------------------------------- - Rick Stevens, Systems Engineer, AllDigital ricks@alldigital.com - - AIM/Skype: therps2 ICQ: 226437340 Yahoo: origrps2 - - - - I was married by a judge. I should have asked for a jury. - - -- Groucho Marx - ----------------------------------------------------------------------
On Fri, Mar 04, 2016 at 09:49:48AM -0800, Rick Stevens wrote:
On 03/04/2016 09:04 AM, jarmo wrote:
Fri, 04 Mar 2016 11:21:14 -0500 Jon LaBadie jonfu@jgcomp.com kirjoitti:
It seems to be a recurring defect. Some kernel releases exhibit it others do not. Last I looked in to it (maybe 9 months ago) it wasn't solved and was being called the Marching Fives.
jl
Then I have met Marching Nines also :D
You chaps aren't using the keyboards with KVMs, are you? I've seen lots of really strange behaviors with KVMs and different kernels.
Nope. Two systems, Fedora 22 and CentOS 7.1. Both use the "unifying" USB dongle from Logitech that handles both mouse and keyboard. But the mice and kbds are different. One keyboard acted up on Fedora, kbd swap cured the problem. Swap back, problem back. Moving mice and usb receivers did not affect the problem. It stayed with the one kbd and Fedora.
Next F22 kernel update, no problems with either kbd. Typically I don't reboot right after a kernel update. When I do reboot it is usually into a different kernel. On one reboot the problem reappeared. Another kernel update appeared just a day or two later; problem gone.
jl
Allegedly, on or about 04 March 2016, Jon LaBadie sent:
One keyboard acted up on Fedora, kbd swap cured the problem. Swap back, problem back. Moving mice and usb receivers did not affect the problem.
Just for curiosity's sake:
Have you tried different USB ports? Some may not be supplying sufficient power (particularly those on flyleads, rather than a port directly on a motherboard or daughtercard), and a peripheral may act up.
Are you using USB extension leads?
... Side topic ...
In the past, I had continual problems with PS/2 mice and Linux, that worked perfectly fine on Windows running on the same PC (the mouse would suddenly start wildly whizzing about the screen, and clicking on things, like it was possessed). My eventual solution was to use USB peripherals. It made me wonder just what was wrong with how Linux dealt with common I/O hardware. Surely there's nothing unknown about handling keyboards and mice, by this time?
On Sat, Mar 05, 2016 at 06:06:08AM +1030, Tim wrote:
Allegedly, on or about 04 March 2016, Jon LaBadie sent:
One keyboard acted up on Fedora, kbd swap cured the problem. Swap back, problem back. Moving mice and usb receivers did not affect the problem.
Just for curiosity's sake:
Have you tried different USB ports? Some may not be supplying sufficient power (particularly those on flyleads, rather than a port directly on a motherboard or daughtercard), and a peripheral may act up.
Motherboard has headers for 8 USB ports, none go to a hub or splitters except for one header feeding a front panel memory card reader. I'm only using 2 other USB port, both of which are wired directly to a header.
Are you using USB extension leads?
The "unifying" dongle normally is on a 30 inch extension cable. But I did try it directly in a USB port without the extension and the problem persisted.
Jon
On Sat, Mar 05, 2016 at 06:06:08AM +1030, Tim wrote:
roblem.
Just for curiosity's sake:
Have you tried different USB ports? Some may not be supplying sufficient power (particularly those on flyleads, rather than a port directly on a motherboard or daughtercard), and a peripheral may act up.
And what makes here difference? If all works ok with 4.4.2-301.fc23.x86_64, but not with 4.4.3, change usb port?
Jarmo
Tim:
Just for curiosity's sake:
Have you tried different USB ports? Some may not be supplying sufficient power (particularly those on flyleads, rather than a port directly on a motherboard or daughtercard), and a peripheral may act up.
jarmo:
And what makes here difference? If all works ok with 4.4.2-301.fc23.x86_64, but not with 4.4.3, change usb port?
I can see three potential vectors:
It could be a corner case, where your power supply is verging on the point of being inadequate, and some change in the kernel has slightly changed the power consumption of the PC.
Your peripheral might not be correctly asking for more current than the default from the port, and a driver change has made that requirement more stringent. There's plenty of things that need more power than the default, but don't bother to negotiate for it, they just presume that a system is going to supply it (which is against the specs). And, I dare say, there's probably devices that use more current than the designers thought it did.
Alternatively, the peripheral might be asking for more current, and a new driver fault isn't complying.
But, I was only putting forward something else to consider. Since there's been quite a number of odd keyboard error reports about keyboards mysteriously typing 5 by themselves, repeatedly, it does sound more like there's something wrong with the keyboard decoding in the kernel.
On the other hand, maybe there's been a batch of badly engineered keyboard encoders, used in various different keyboards, that hasn't reared its ugly head until now.
-
jarmo -
Jon LaBadie -
Patrick Laimbock -
Rick Stevens -
Tim