Bug#573133: arbtt: idle time computation does not work as expected

2010-07-20 Thread Helmut Grohne
reopen 573133
reassign 573133 xserver-xorg-core
thanks

On Sun, Jul 18, 2010 at 05:56:53PM +0200, Joachim Breitner wrote:
 Am Sonntag, den 18.07.2010, 17:56 +0200 schrieb Helmut Grohne:
  On Fri, Mar 12, 2010 at 01:49:39PM +0100, Joachim Breitner wrote:
   the wrong data passed via the XScreenSaver extension is produced by the
   X server, so probably xserver-xorg-core.
  
  It looks like some xorg upload fixed this issue: It is no longer
  reproducible.
 
 thanks, closing this bug.

Actually the issue is reproducible. One only needs to run X and arbtt
long enough. For a fresh X accurate numbers are reported by xprintidle.
However running X and/or arbtt long enough break the XScreenSaver
extension again. xprintidle would the report numbers = 40
(milliseconds), which is wrong. This bug is probably hard to reproduce.
Any suggestions on work arounds?

Helmut



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Bug#573133: arbtt: idle time computation does not work as expected

2010-07-18 Thread Helmut Grohne
Hi Joachim,

On Fri, Mar 12, 2010 at 01:49:39PM +0100, Joachim Breitner wrote:
 the wrong data passed via the XScreenSaver extension is produced by the
 X server, so probably xserver-xorg-core.

It looks like some xorg upload fixed this issue: It is no longer
reproducible.

Helmut



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Bug#573133: arbtt: idle time computation does not work as expected

2010-03-12 Thread Joachim Breitner
Hi,

Am Donnerstag, den 11.03.2010, 21:55 +0100 schrieb Helmut Grohne:
 On Thu, Mar 11, 2010 at 08:31:40PM +0100, Joachim Breitner wrote:
   However the whole night appears in the stats. I therefore had a look at
   the ouput of arbtt-dump, specifically the cLastActivity numbers. They
   seem to range between 0 and 14. During the time I was asleep the
   numbers were around 0 to 40.
  
  This is strange. Do you maybe have some kind of wobbly input device,
  like a joystick or anything that would make X believe there was user
  input?
 
 This was the right question to ask. As the idle time is computed
 correctly when I sit in front of my machine it must have to do with my
 way of locking my screen. I first switch to another vt and then use
 vlock. And indeed, after switching to the other vt the idle time (it is
 also reported by xprintidle) is always below 50. The next thing I tried
 was to switch away from X using chvt (as root). Interestingly the enter
 key used to issue that command is not released until I switch back to X
 again. The wobbly input device therefore is my (PS/2) keyboard. Do you
 know which package to assign this bug to?

the wrong data passed via the XScreenSaver extension is produced by the
X server, so probably xserver-xorg-core.

Greetings,
Joachim

-- 
Joachim nomeata Breitner
Debian Developer
  nome...@debian.org | ICQ# 74513189 | GPG-Keyid: 4743206C
  JID: nome...@joachim-breitner.de | http://people.debian.org/~nomeata


signature.asc
Description: This is a digitally signed message part


Bug#573133: arbtt: idle time computation does not work as expected

2010-03-11 Thread Joachim Breitner
Dear Helmut,

Am Dienstag, den 09.03.2010, 10:48 +0100 schrieb Helmut Grohne:
 I finally found some time to try arbtt. It looks like a great tool, but
 somehow does not produce reasonable stats for me.
 
 I tried to use the inactive tag as documented:
 
 $idle  60 == tag inactive,
 
 However the whole night appears in the stats. I therefore had a look at
 the ouput of arbtt-dump, specifically the cLastActivity numbers. They
 seem to range between 0 and 14. During the time I was asleep the
 numbers were around 0 to 40.

This is strange. Do you maybe have some kind of wobbly input device,
like a joystick or anything that would make X believe there was user
input?

Do screensavers activate as expected?

 This made me digg in the source. The
 numbers seem to be deduced from Graphics.X11.XScreenSaver.getXIdleTime
 and the unit is said to be milli-seconds. Now comparing with my data
 this sounds wrong.

Here, it looks ok. Note that $idle refers to seconds, while the data is
(should be) stored as milli seconds.

 Do I need xscreensaver packages for arbtt? I don't have those installed.

No, XScreenSaver is also the name of an X protocol extension and is not
directly related to the program xscreensaver.

Greetings,
Joachim

-- 
Joachim nomeata Breitner
Debian Developer
  nome...@debian.org | ICQ# 74513189 | GPG-Keyid: 4743206C
  JID: nome...@joachim-breitner.de | http://people.debian.org/~nomeata


signature.asc
Description: This is a digitally signed message part


Bug#573133: arbtt: idle time computation does not work as expected

2010-03-11 Thread Helmut Grohne
Hi Joachim,

On Thu, Mar 11, 2010 at 08:31:40PM +0100, Joachim Breitner wrote:
  However the whole night appears in the stats. I therefore had a look at
  the ouput of arbtt-dump, specifically the cLastActivity numbers. They
  seem to range between 0 and 14. During the time I was asleep the
  numbers were around 0 to 40.
 
 This is strange. Do you maybe have some kind of wobbly input device,
 like a joystick or anything that would make X believe there was user
 input?

This was the right question to ask. As the idle time is computed
correctly when I sit in front of my machine it must have to do with my
way of locking my screen. I first switch to another vt and then use
vlock. And indeed, after switching to the other vt the idle time (it is
also reported by xprintidle) is always below 50. The next thing I tried
was to switch away from X using chvt (as root). Interestingly the enter
key used to issue that command is not released until I switch back to X
again. The wobbly input device therefore is my (PS/2) keyboard. Do you
know which package to assign this bug to?

Helmut



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org



Bug#573133: arbtt: idle time computation does not work as expected

2010-03-09 Thread Helmut Grohne
Package: arbtt
Version: 0.4.5.1-1
Severity: important

Hi,

I finally found some time to try arbtt. It looks like a great tool, but
somehow does not produce reasonable stats for me.

I tried to use the inactive tag as documented:

$idle  60 == tag inactive,

However the whole night appears in the stats. I therefore had a look at
the ouput of arbtt-dump, specifically the cLastActivity numbers. They
seem to range between 0 and 14. During the time I was asleep the
numbers were around 0 to 40. This made me digg in the source. The
numbers seem to be deduced from Graphics.X11.XScreenSaver.getXIdleTime
and the unit is said to be milli-seconds. Now comparing with my data
this sounds wrong.

Do I need xscreensaver packages for arbtt? I don't have those installed.

Helmut



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org