@Seth:
I just saw that I missed several of your messages above. In particular 98 and 99. I'm sorry. Before I say it you coded it and had it building on launchpad. Great.
I see no problem with your code. I built your change here and am running it successfully. Since I didn't see the bug I can't say any more.
You said: > I have seen evidence in Xorg.*.log that this appeared to work even though I had four X session launch attempts within about a second (the first failing because of nvidia not being around).
Sounds like you are close to the limit. Do you think there was another reason for the other 4? I do get one message about a session not lasting long, but not two or more like you and Jamie. I assume that's normal but I don't know.
Even if we go no farther here I want to note in the record that driver failures are happening because X is being started prematurely.
@Seth:
I just saw that I missed several of your messages above. In particular 98 and 99. I'm sorry. Before I say it you coded it and had it building on launchpad. Great.
I see no problem with your code. I built your change here and am running it successfully. Since I didn't see the bug I can't say any more.
You said: > I have seen evidence in Xorg.*.log that this appeared to work even though I had four X session launch attempts within about a second (the first failing because of nvidia not being around).
daemon/ gdm-local- display- factory. c:#define MAX_DISPLAY_ FAILURES 5
Sounds like you are close to the limit. Do you think there was another reason for the other 4? I do get one message about a session not lasting long, but not two or more like you and Jamie. I assume that's normal but I don't know.
Even if we go no farther here I want to note in the record that driver failures are happening because X is being started prematurely.