I'm on Fedora 6.3. After a reboot, some proprietary software didn't want to run. I found out that the startup script for said software manually sets DISPLAY to :0.0, which I know is not a good idea, and I can fix.
However, this still doesn't explain why my default X DISPLAY is suddenly :3.0.
On Thu, 25 Apr 2013, Joakim Ziegler wrote:
I'm on Fedora 6.3. After a reboot, some proprietary software didn't want to run. I found out that the startup script for said software manually sets DISPLAY to :0.0, which I know is not a good idea, and I can fix.
I'd thought that :0.0 was the norm.
However, this still doesn't explain why my default X DISPLAY is suddenly :3.0.
On 2013-04-26, Michael Hennebry hennebry@web.cs.ndsu.nodak.edu wrote:
On Thu, 25 Apr 2013, Joakim Ziegler wrote:
I'm on Fedora 6.3. After a reboot, some proprietary software didn't want to run. I found out that the startup script for said software manually sets DISPLAY to :0.0, which I know is not a good idea, and I can fix.
I'd thought that :0.0 was the norm.
It is, but it's not a hard and fast rule. If you are running multiple local X consoles, for example, they can't all be :0.0. I've also seen different identifiers when one X session is hung and/or doesn't finish cleanly before a new one is brought up.
If DISPLAY is consistently set to :3.0, perhaps it is a Fedora-specific thing, and it'd be best to ask people more familiar with Fedora. If they have made changes, those changes may not have propagated to RHEL/ CentOS yet.
--keith
Keith Keller wrote:
I'd thought that :0.0 was the norm.
It is, but it's not a hard and fast rule. If you are running multiple local X consoles, for example, they can't all be :0.0. I've also seen different identifiers when one X session is hung and/or doesn't finish cleanly before a new one is brought up.
I've seen this on CentOS 6 - X (re-)starts up on something other than :0
I didn't spend any time investigating why at the time - but I guess the reasons why X didn't start on :0 should be in /var/log/Xorg.0.log ?
James Pearson
On Fri, Apr 26, 2013 at 10:36 AM, Keith Keller kkeller@wombat.san-francisco.ca.us wrote:
I'm on Fedora 6.3. After a reboot, some proprietary software didn't want to run. I found out that the startup script for said software manually sets DISPLAY to :0.0, which I know is not a good idea, and I can fix.
I'd thought that :0.0 was the norm.
It is, but it's not a hard and fast rule. If you are running multiple local X consoles, for example, they can't all be :0.0. I've also seen different identifiers when one X session is hung and/or doesn't finish cleanly before a new one is brought up.
I don't know why the session on the first console wouldn't be :0.0, but I'd consider any X application that cared what display it uses to be badly broken. I'm much more likely to run things in a freenx/NX session than on the main console (if there even is one).
-- Les Mikesell lesmikesell@gmail.com
On 26/04/13 10:36, Keith Keller wrote:
On 2013-04-26, Michael Hennebry hennebry@web.cs.ndsu.nodak.edu wrote:
On Thu, 25 Apr 2013, Joakim Ziegler wrote:
I'm on Fedora 6.3. After a reboot, some proprietary software didn't want to run. I found out that the startup script for said software manually sets DISPLAY to :0.0, which I know is not a good idea, and I can fix.
I'd thought that :0.0 was the norm.
It is, but it's not a hard and fast rule. If you are running multiple local X consoles, for example, they can't all be :0.0. I've also seen different identifiers when one X session is hung and/or doesn't finish cleanly before a new one is brought up.
If DISPLAY is consistently set to :3.0, perhaps it is a Fedora-specific thing, and it'd be best to ask people more familiar with Fedora. If they have made changes, those changes may not have propagated to RHEL/ CentOS yet.
Sorry, brain fart, I'm running CentOS 6.3, not Fedora. The weird thing is that this changed after a reboot. I haven't done any updates that seem relevant lately either.
And yes, I know :0.0 shouldn't be depended on, but it seems weird that it'd change like that for no good reason.
On 2013-04-26, Joakim Ziegler joakim@terminalmx.com wrote:
Sorry, brain fart, I'm running CentOS 6.3, not Fedora. The weird thing is that this changed after a reboot. I haven't done any updates that seem relevant lately either.
And yes, I know :0.0 shouldn't be depended on, but it seems weird that it'd change like that for no good reason.
Agreed. You might take James' advice and check the X logs. If you're lucky, if the issue is a couple of failures to start X, the successful start won't have written over the previous logs. (I'm not sure how you'd be able to tell if you were unlucky and X had to restart a few times but overwrote the log.)
--keith
On 04/26/2013 07:06 PM, Keith Keller wrote:
On 2013-04-26, Joakim Ziegler joakim@terminalmx.com wrote:
Sorry, brain fart, I'm running CentOS 6.3, not Fedora. The weird thing is that this changed after a reboot. I haven't done any updates that seem relevant lately either.
And yes, I know :0.0 shouldn't be depended on, but it seems weird that it'd change like that for no good reason.
Agreed. You might take James' advice and check the X logs. If you're lucky, if the issue is a couple of failures to start X, the successful start won't have written over the previous logs. (I'm not sure how you'd be able to tell if you were unlucky and X had to restart a few times but overwrote the log.)
--keith
I've seen this from time to time. It always seems to happen when I change run levels without a reboot. That makes me think it may have something to do with an earlier post, when an old X session does not exit completely before the new one.
In any case, it seems to have little impact on how it all works.
On 2013-04-28, Jay Leafey jay.leafey@mindless.com wrote:
On 2013-04-26, Joakim Ziegler joakim@terminalmx.com wrote:
Sorry, brain fart, I'm running CentOS 6.3, not Fedora. The weird thing= is that this changed after a reboot.
I've seen this from time to time. It always seems to happen when I=20 change run levels without a reboot.
The OP already stated (see above) that he saw this directly after a reboot.
That makes me think it may have=20 something to do with an earlier post, when an old X session does not=20 exit completely before the new one.
That's still a possibility, which is why multiple people have mentioned checking the logs.
In any case, it seems to have little impact on how it all works.
No, and the OP seems to know this. But it can't hurt to know what's changed--it could be helpful information if the change turns out to break something else later. (Odds are slim in this case, but you never know--something already stopped working because of it.)
--keith
-
James Pearson -
Jay Leafey -
Joakim Ziegler -
Keith Keller -
Les Mikesell -
Michael Hennebry