On Thu, Jun 23, 2005 at 06:17:42PM -0400, Balbir Thomas wrote: > On Fri, Jun 17, 2005 at 01:04:23PM +0930, Ron wrote: > > Hi, I'm afraid you are going to have to clarify a few things, > > and/or do some debugging of your own here. I've just been > > forwarded a separate report where you claim(ed) this is the > > solution? I gather you no longer believe that?? > > After uncommenting the "Cursor" line in /etc/TextConig > I find by restarting (/etc/init.d/svgatextmode restart) > I am able to activate the cursor on tty2 to tty6.
And what does it do if you restart it without changing anything in TextConig? > > I will take one punt though, now that I have some inkling this > > may be with kernel 2.6, and suggest you look at the initialisation > > order of related things after a cold boot. > > > > I have not changed anything about the initialization order in any > way. In fact if I create /etc/rc2.d/S99local -> /etc/init.d/local > and add a line "/etc/init.d/svgatextmode restart", this does > not solve the problem. I need to log in from tty1 and manually > do the restart to get the cursor working. Well. that is arguably, _some_ point in the boot process, though a very late one. > This is very odd and I have no clue what is happening here. +1 Aside from the fact it is clear that something is now happening out of order, or not performing the correct initialisation. > Permit me to repeat this for emphasis restarting SVGATextMode > at no point in the boot process, no matter how late (such as > through /etc/init.d/local) does not fix the problem. It needs > to be started manually after a console login from the working > tty1. > > Further logging in through tty2-tty6 and restarting causes a > problem in that only half my monitor screen seems to be used. > > > If simply restarting stm makes things better, it is possible > > that we are initialising something too early (or late) now, > > before udev and/or other mechanisms fully configure the all > > console vt's or so. > > > > As mentioned restarting from /etc/init.d/local after udev > has created the device nodes does not seem to help. > > > But I'm not going to be able to do much here without a tested > > patch, or a much more detailed analysis of exactly what is > > going wrong on what sort of systems. > > > > Ok. I'll try looking into this as and when I get a new idea. I > have run out of them for now :-(. You could try adding some logging to various parts you suspect and see what happens (or not) and when. I'm afraid I don't have a 2.6 box I can test this with, so I'm going to have to act on advice and patches. sorry, Ron -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
