Here's a problem I'm hoping to solve: When idle for some amount of time, X shuts down the screen. When a program like VLC is playing a video (or even paused) this does not happen. How can I tell X to never sleep? I just want it to send output down that cable, constantly, forever. I manage screen burn and power wasting by turning off the monitor, but when the screen goes to sleep, my inexpensive monitor (an LG) sometimes has trouble noticing and won't display anything until X is restarted, which is a pain. Thanks! PS The system is Debian testing.
On 17 de março de 2015 at 08:56:01, William Witteman (wwitteman@gmail.com) wrote: Here's a problem I'm hoping to solve: When idle for some amount of time, X shuts down the screen. When a program like VLC is playing a video (or even paused) this does not happen. How can I tell X to never sleep? I just want it to send output down that cable, constantly, forever. I manage screen burn and power wasting by turning off the monitor, but when the screen goes to sleep, my inexpensive monitor (an LG) sometimes has trouble noticing and won't display anything until X is restarted, which is a pain. Probably not a X configuration, but a WM. What WM are you using?? You need to check the Power Save options, or something like that (depending on which WM). -- Marcelo Cavalcante Rocha - Kalib Fortaleza - Ceará - Brazil Mobile: +55 085 87620983 Certificações: ITIL V3 | CSM | LPI-C1 | LPI-C2 | LPI-C3 | Novell CLA Minha Pessoa: Blog =================================================== Proteja meu endereço como estou protegendo o seu. Não revele e-mail dos correspondentes: use Cco (Copia Carbonada Oculta). Retire os endereços antes de reenviar. Dificulte assim a disseminação de vírus e spam.
I'm using alpine. It runs in an xterm. In the message I'm replying to, I cannot tell the quoted text from the new text. This may be a "feature" of gmail. I'm not sure what "X-Mailer: Airmail (286)" signifies in Marcelo's email. I had the same problem with Russell Reiter's postings. Russell's postings may have been fixed by his recent switch to k9. I don't know since now he's "top posting" (which I think is unfortunate in a mailing list). He previously responded-by-chunk, which I think is great EXCEPT when quote and response are indistinguishable ("red" doesn't work in plain ASCII). Here's what I recommend: - quote relevant chunks of what you are replying to - add your response after the relevant chunk - make sure that the quotes and replies are distinct, even in ASCII - don't quote parts that you are not responding to This is one way to be considerate of your audience. Regardless of what I just said, I'm top-posting for this one example so that you can see what I see. | From: Marcelo Kalib <kalibslack@gmail.com> | To: GTALUG Talk <talk@gtalug.org> | Date: Tue, 17 Mar 2015 09:01:04 -0300 | Subject: Re: [GTALUG] Keep X from going to sleep | Reply-To: GTALUG Talk <talk@gtalug.org> | | On 17 de março de 2015 at 08:56:01, William Witteman (wwitteman@gmail.com) wrote: | Here's a problem I'm hoping to solve: | | When idle for some amount of time, X shuts down the screen. When a | program like VLC is playing a video (or even paused) this does not | happen. | | How can I tell X to never sleep? I just want it to send output down | that cable, constantly, forever. | | I manage screen burn and power wasting by turning off the monitor, but | when the screen goes to sleep, my inexpensive monitor (an LG) | sometimes has trouble noticing and won't display anything until X is | restarted, which is a pain. | | Probably not a X configuration, but a WM. What WM are you using?? You need to check the Power Save options, or something like that (depending on which WM). | | | -- | Marcelo Cavalcante Rocha - Kalib | | Fortaleza - Ceará - Brazil | Mobile: +55 085 87620983 | Certificações: ITIL V3 | CSM | LPI-C1 | LPI-C2 | LPI-C3 | Novell CLA | Minha Pessoa: Blog | | =================================================== | | Proteja meu endereço como estou protegendo o seu. | Não revele e-mail dos correspondentes: use Cco (Copia Carbonada Oculta). | Retire os endereços antes de reenviar. Dificulte assim a | disseminação de vírus e spam.
On March 17, 2015 1:45:15 PM EDT, "D. Hugh Redelmeier" <hugh@mimosa.com> wrote: <snip previous>
I had the same problem with Russell Reiter's postings. Russell's postings may have been fixed by his recent switch to k9. I don't know since now he's "top posting" (which I think is unfortunate in a mailing list). He previously responded-by-chunk, which I think is great EXCEPT when quote and response are indistinguishable ("red" doesn't work in plain ASCII).
Sorry I tend to top post when I hurry around and try to reply to topics as well. I often forget that others besides the intended recipient get the messages too. Hopefully my replies from this mobile device are now colourless as well. -- Sent from my Android device with K-9 Mail. Please excuse my brevity.
There are two primary controls which can blank the screen. One is the screen saver widget the other is power management. Obviously the screen saver can display an image but it can also just blank the screen. Power management can spin down drives etc. but it also can blank the screen. If you access both of these and set each one to never, your screen should stay lit. One caveat, for some reason updates of either of these resets the time to blank the screen to 10 min or so. Or at least it has done this in the past. With everyone using at least LCD displays now I don't know why the screen savers ship enabled by default. They were originally intended to stop CRT burn in. That is if you left a static image on your screen for a significant amount of time you would actually burn the image onto your display. I don't believe burn in is a problem these days. My screen never shut off when my computer is running, sometimes weeks of uptime before I reboot for one reason or another. Haven't toasted my LCD monitor, yet! Hope this helps. Russell On Mar 17, 2015 7:56 AM, "William Witteman" <wwitteman@gmail.com> wrote:
Here's a problem I'm hoping to solve:
When idle for some amount of time, X shuts down the screen. When a program like VLC is playing a video (or even paused) this does not happen.
How can I tell X to never sleep? I just want it to send output down that cable, constantly, forever.
I manage screen burn and power wasting by turning off the monitor, but when the screen goes to sleep, my inexpensive monitor (an LG) sometimes has trouble noticing and won't display anything until X is restarted, which is a pain.
Thanks!
PS The system is Debian testing. --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
On March 17, 2015 8:18:52 AM EDT, Russell Reiter <rreiter91@gmail.com> wrote:
With everyone using at least LCD displays now I don't know why the screen savers ship enabled by default.
Because they evolved into screen lockers. Much like a key lock on you front door, it's there now to deter invasion of privacy and malicious use of property. -- Scott Sullivan
On 03/17/2015 09:01 AM, Scott Sullivan wrote:
Because they evolved into screen lockers.
Back when I was at IBM, we were required to have our desktops lock up automatically. There were security people wandering around to ensure this was done. For my computer, I made a bitmap of my desktop and had it display when the "screen saver" kicked in. So, to anyone passing by, it looked as though my computer wasn't locked and they'd have to check. ;-)
On Tue, Mar 17, 2015 at 08:18:52AM -0400, Russell Reiter wrote:
There are two primary controls which can blank the screen. One is the screen saver widget the other is power management. Obviously the screen saver can display an image but it can also just blank the screen. Power management can spin down drives etc. but it also can blank the screen. If you access both of these and set each one to never, your screen should stay lit.
One caveat, for some reason updates of either of these resets the time to blank the screen to 10 min or so. Or at least it has done this in the past. With everyone using at least LCD displays now I don't know why the screen savers ship enabled by default. They were originally intended to stop CRT burn in. That is if you left a static image on your screen for a significant amount of time you would actually burn the image onto your display. I don't believe burn in is a problem these days. My screen never shut off when my computer is running, sometimes weeks of uptime before I reboot for one reason or another. Haven't toasted my LCD monitor, yet! Hope this helps.
Well at least with DPMS, making the screen go into power saving mode does save power, so having it do that by default does sound like a reasonable thing. -- Len Sorensen
I agree saving power is a reasonable thing. That's one reason I don't recommend disabling DPMS completely. Spinning down the drives saves power and also servs to extend the lifespan of the drive, if only marginally. On Tue, Mar 17, 2015 at 9:52 AM, Lennart Sorensen < lsorense@csclub.uwaterloo.ca> wrote:
There are two primary controls which can blank the screen. One is the screen saver widget the other is power management. Obviously the screen saver can display an image but it can also just blank the screen. Power management can spin down drives etc. but it also can blank the screen. If you access both of these and set each one to never, your screen should stay lit.
One caveat, for some reason updates of either of these resets the time to blank the screen to 10 min or so. Or at least it has done this in the
On Tue, Mar 17, 2015 at 08:18:52AM -0400, Russell Reiter wrote: past.
With everyone using at least LCD displays now I don't know why the screen savers ship enabled by default. They were originally intended to stop CRT burn in. That is if you left a static image on your screen for a significant amount of time you would actually burn the image onto your display. I don't believe burn in is a problem these days. My screen never shut off when my computer is running, sometimes weeks of uptime before I reboot for one reason or another. Haven't toasted my LCD monitor, yet! Hope this helps.
Well at least with DPMS, making the screen go into power saving mode does save power, so having it do that by default does sound like a reasonable thing.
-- Len Sorensen --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
On 17 March 2015 at 10:20, Russell Reiter <rreiter91@gmail.com> wrote:
I agree saving power is a reasonable thing. That's one reason I don't recommend disabling DPMS completely. Spinning down the drives saves power and also servs to extend the lifespan of the drive, if only marginally.
That's a good point. My drives don't spin, but it's still a good point. If my monitor would acknowledge the signal coming in when I am typing and staring at a blank screen, I wouldn't bother with this.
You mean your using solid state drives and not getting wake on keypress either? On Tue, Mar 17, 2015 at 10:25 AM, William Witteman <wwitteman@gmail.com> wrote:
On 17 March 2015 at 10:20, Russell Reiter <rreiter91@gmail.com> wrote:
I agree saving power is a reasonable thing. That's one reason I don't recommend disabling DPMS completely. Spinning down the drives saves power and also servs to extend the lifespan of the drive, if only marginally.
That's a good point. My drives don't spin, but it's still a good point. If my monitor would acknowledge the signal coming in when I am typing and staring at a blank screen, I wouldn't bother with this. --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
On 17 March 2015 at 10:37, Russell Reiter <rreiter91@gmail.com> wrote:
You mean your using solid state drives and not getting wake on keypress either?
I am using SSDs, and the computer wakes up fine on keypress, but the monitor stays stubbornly dark. What I currently do is Alt-Tab over to a terminal or fire one up using F1 (mapped in my Openbox config) and type pkill Xorg. Then the screen wakes up and shows me the console, and I restart X and firefox, which asks me if I want to restore 71 tabs :-)
71 tabs you had pinned or 71 new tabs? For the longest time SElinux, which a friend uses in Fedora, was choking on colourd. Notwithstanding that the Nvidia driver was a little problematic to install, it was better than the native nouveau drivers Fedora was shipping at the time. Around the time of Fedora 18, chrome would start and then download a 1mg or more file to a folder. Then an automatic update fixed the problem so I didn't have to look into it too much. Colord is responsible, in part, for defining the neutral palate that all else is rendered against in the display. You are on testing so I'd either regress to stable unless there is a good reason not to or, have a look for colord debug notes in the logs. On Tue, Mar 17, 2015 at 10:41 AM, William Witteman <wwitteman@gmail.com> wrote:
On 17 March 2015 at 10:37, Russell Reiter <rreiter91@gmail.com> wrote:
You mean your using solid state drives and not getting wake on keypress either?
I am using SSDs, and the computer wakes up fine on keypress, but the monitor stays stubbornly dark. What I currently do is Alt-Tab over to a terminal or fire one up using F1 (mapped in my Openbox config) and type pkill Xorg. Then the screen wakes up and shows me the console, and I restart X and firefox, which asks me if I want to restore 71 tabs :-) --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
I just checked some exploits from the last year. There is one WiFi attack which targets the red wire of video, in effect capturing frames which are carried on it. It is a directed attack, but what I'm getting at is that there may be a possibility of RF interference. I remember back in the say that running cable for IBM token ring networks was problematic if the wiring ran parallel to AC carried over Romex wiring and completely FUBAR crossing near florescent lighting. So your cables are all neat and tidy and not close to ac cables, lighting ballasts or compact fluorescent lights, and in a place with relatively modern and properly grounded power, right. If so I think its possible that on wake events, the palate Damon doesn't release the blank screen, which is a signal of its own, rather than the lack of a signal, as we tend to assume that's what a nothing is. On March 17, 2015 11:47:46 AM EDT, William Witteman <wwitteman@gmail.com> wrote:
On 17 March 2015 at 10:58, Russell Reiter <rreiter91@gmail.com> wrote:
71 tabs you had pinned or 71 new tabs?
Just however many tabs were open when killing X took out Firefox, so it asks me (very pleasantly, I think) if I want them back. I usually do. --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
-- Sent from my Android device with K-9 Mail. Please excuse my brevity.
I have an old generic laptop with a dead screen, and I wired it to my TV using the hdmi port. Some times the screen does not come back too, and I discovered that pressing ctrl-alt-f1 and ctrl-alt-f8 it comes back without having to kill X. On Mar 17, 2015 6:10 PM, "R Russell Reiter" <rreiter91@gmail.com> wrote:
I just checked some exploits from the last year. There is one WiFi attack which targets the red wire of video, in effect capturing frames which are carried on it. It is a directed attack, but what I'm getting at is that there may be a possibility of RF interference. I remember back in the say that running cable for IBM token ring networks was problematic if the wiring ran parallel to AC carried over Romex wiring and completely FUBAR crossing near florescent lighting. So your cables are all neat and tidy and not close to ac cables, lighting ballasts or compact fluorescent lights, and in a place with relatively modern and properly grounded power, right. If so I think its possible that on wake events, the palate Damon doesn't release the blank screen, which is a signal of its own, rather than the lack of a signal, as we tend to assume that's what a nothing is.
On March 17, 2015 11:47:46 AM EDT, William Witteman <wwitteman@gmail.com> wrote:
On 17 March 2015 at 10:58, Russell Reiter <rreiter91@gmail.com> wrote:
71 tabs you had pinned or 71 new tabs?
Just however many tabs were open when killing X took out Firefox, so it asks me (very pleasantly, I think) if I want them back. I usually do. --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
-- Sent from my Android device with K-9 Mail. Please excuse my brevity. --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
On Tue, Mar 17, 2015 at 10:20:59AM -0400, Russell Reiter wrote:
I agree saving power is a reasonable thing. That's one reason I don't recommend disabling DPMS completely. Spinning down the drives saves power and also servs to extend the lifespan of the drive, if only marginally.
Some of the drives I have had last the longest never spun down. I had an 18GB quantum fireball that ran for 82000 power on hours before the bearings started sounded so bad you didn't want to be in the same building as the drive. -- Len Sorensen
There is that factor as well. The hardest thing on any vehicle is that first start in the morning. Little lubrication, not at optimum operating temperature etc. I don't think anything in life is really either or, it's more like if and when. On Tue, Mar 17, 2015 at 10:56 AM, Lennart Sorensen < lsorense@csclub.uwaterloo.ca> wrote:
On Tue, Mar 17, 2015 at 10:20:59AM -0400, Russell Reiter wrote:
I agree saving power is a reasonable thing. That's one reason I don't recommend disabling DPMS completely. Spinning down the drives saves power and also servs to extend the lifespan of the drive, if only marginally.
Some of the drives I have had last the longest never spun down. I had an 18GB quantum fireball that ran for 82000 power on hours before the bearings started sounded so bad you didn't want to be in the same building as the drive.
-- Len Sorensen --- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
On March 17, 2015 7:55:37 AM EDT, William Witteman <wwitteman@gmail.com> wrote:
Here's a problem I'm hoping to solve:
When idle for some amount of time, X shuts down the screen. When a program like VLC is playing a video (or even paused) this does not happen.
How can I tell X to never sleep? I just want it to send output down that cable, constantly, forever.
The monitor sleep functionality is called DPMS. At run time you can disable it with the xset command. xset -dpms The more permanent option is to create a conf snippet in your xorg.conf.d. Once again the arch wiki has great docs. And this specific information will readily translate to debian or any other distro. https://wiki.archlinux.org/index.php/Display_Power_Management_Signaling -- Scott Sullivan
On Tue, Mar 17, 2015 at 07:55:37AM -0400, William Witteman wrote:
Here's a problem I'm hoping to solve:
When idle for some amount of time, X shuts down the screen. When a program like VLC is playing a video (or even paused) this does not happen.
How can I tell X to never sleep? I just want it to send output down that cable, constantly, forever.
I manage screen burn and power wasting by turning off the monitor, but when the screen goes to sleep, my inexpensive monitor (an LG) sometimes has trouble noticing and won't display anything until X is restarted, which is a pain.
Thanks!
PS The system is Debian testing.
As already suggested, xset -dpms and perhaps xset s off, can help, but you would need to do them after each login, given most window manager/desktop environments tend to change the values. I just fixed xfce on one machine at home yesterday by going into xfce4's power management settings and turning off putting the display to sleep or off, and no more screen going to sleep problem. -- Len Sorensen
On Tue, Mar 17, 2015 at 07:55:37AM -0400, William Witteman wrote
Here's a problem I'm hoping to solve:
When idle for some amount of time, X shuts down the screen. When a program like VLC is playing a video (or even paused) this does not happen.
How can I tell X to never sleep? I just want it to send output down that cable, constantly, forever.
I manage screen burn and power wasting by turning off the monitor, but when the screen goes to sleep, my inexpensive monitor (an LG) sometimes has trouble noticing and won't display anything until X is restarted, which is a pain.
That may not be necessary. I experience "similar" behaviour, but the answer to that is *NOT* to restart X. Basically, X has told the monitor to not merely blank the screen, but to shut down altogether. When that happens to me, and I come back from a distraction elsewhere, I merely reach for the "Power" button on the monitor, and toggle it back on again. It's that simple. Try it. -- Walter Dnes <waltdnes@waltdnes.org>
On 17 March 2015 at 17:41, Walter Dnes <waltdnes@waltdnes.org> wrote:
That may not be necessary. I experience "similar" behaviour, but the answer to that is *NOT* to restart X. Basically, X has told the monitor to not merely blank the screen, but to shut down altogether. When that happens to me, and I come back from a distraction elsewhere, I merely reach for the "Power" button on the monitor, and toggle it back on again. It's that simple. Try it.
The monitor if usually off when the described behaviour occurs. Even when it is on, power cycling it has no effect. It is not that simple. "xset -dpms" doesn't work, nor does "xset s off", nor does "setterm -blank 0 -powersave off -powerdown 0" If these commands are achieving ant result, it is being overridden elsewhere. Ctrl-Alt-F* followed by Ctrl-Alt-F1 *does* work - thanks to Mauro Sousa. It is a nicer kludge, and I appreciate it. However, I'd still love to figure out a real, permanent solution, if possible. The WM is OpenBox, which isn't doing any power management, but I will see if can get it to invoke a different power manager (XFCE looks promising, based on Lennart's experience) and try that. I will report back.
Ctrl-Alt-F* followed by Ctrl-Alt-F1 *does* work - thanks to Mauro Sousa. It is a nicer kludge, and I appreciate it. However, I'd still love to figure out a real, permanent solution, if possible.
Probably SYN & ACK requests- IMHO.
The WM is OpenBox, which isn't doing any power management, but I will see if can get it to invoke a different power manager (XFCE looks promising, based on Lennart's experience) and try that. I will report back
Remember that if you are testing a virtualized environment your profiles should match real world hardware. For example it might look advantageous to set the baud rate of a virtual disk to 115000, however 9600 is a supported real world rate that won't flood the bus with sync or other requests in difficult circumstances, most especially in a testing environment. The baud rate on the planar is different than the baud rate of modem communications in dealing with the bus ie. Perhaps it is device limited to requesting and syncing clock tics. I know this may not seem to relate to the problem at hand but I hear testing, I think testing.
--- Talk Mailing List talk@gtalug.org http://gtalug.org/mailman/listinfo/talk
-- Sent via K-9 Mail.
I did some more reading on openbox and it appears to conform w/ http://www.freedesktop.org/software/colord/ I note that colord was implemented to solve certain discrete problems and uses PolicyKit to effect finer grained user control and further that explicit device profiles are required for each display in order to ensure stability and persistence. I confess to having my own motives for pointing you in this direction, as your problems are quite similar to certain display anomalies experienced by some of the end users I support on a variety of spins. Most recently trying to tame SElinux for some cloud using Fedora Linux newbies. I hope you continue to track this issue rather than switch. Although I'm a big fan of XFCE on LDM myself. You can tell me to buzz off at any time, I've got a pretty thick skin and a sense of humor, which serve me well most of the time. FYI I'm most definitely not claiming to be a Linux super guru. From an enterprise perspective, I'm at best a semi-skilled sysop however, I am an excellent hardware integrator with several decades of integration experiences to draw upon. Cheers Russell -- Sent via K-9 Mail.
participants (11)
-
D. Hugh Redelmeier -
James Knott -
Lennart Sorensen -
Marcelo Kalib -
Mauro Souza -
R Russell Reiter -
R. Russell Reiter
-
Russell Reiter
-
Scott Sullivan -
Walter Dnes -
William Witteman