reopen 403031 severity 403031 wishlist retitle 403031 Serial console detection should happen only once. thanks
Sorry, this mail left due to a bad manipulation before i had time to finish it or even review it :/ On Tue, Dec 26, 2006 at 08:24:41AM +0100, Sven Luther wrote: > On Mon, Dec 25, 2006 at 09:36:52PM +0100, Frans Pop wrote: > > On Tuesday 19 December 2006 23:07, Sven Luther wrote: > > The second one will be the one finish-install is run from. This instance > > will have TERM_TYPE=network, and not TERM_TYPE=serial like the original > > one. So, in that case, even though the install was started over serial > > console, the installed system would not be set up to use serial console > > (while this works correctly with the current logic). > > Cool, i knew there was something. I guess the good solution here is to set a > debconf variable in rootskel which would be there to check it. Ok, I re-open this bug as a wishlist. I believe that it is plain wrong to do the check twice, and instead some way needs to be found to circumwent the problem you found above. Possibly by adding an explicit debconf variable in rootskel/detect-console or something. Notice too, that this may play in well (post-etch), with the idea to enable serial consoles even for normal installs, as this may be a useful thing to do. I don't know how to best do this, but a search for serial consoles in the dmesg output may be a solution. This has to be a low-priority interactive process though. > > /me is so glad he is conservative with patches like this and takes the > > time to think them through and really test them instead of applying them > > blindly. > > /me is so glade you took the time to investigate the issue, instead of > dismissing it out of hand as your first reply indicated. > > > Hopefully this will teach you to not dismiss my concerns out of hand in > > the future, though I won't hold my breath. > > Well, please re-read my mails, i posted the patch, asked why this was not > done, and *EXPLICITLY* asked this to be double checked, did i not ? ... Frans, Fabio, Anthony, i think this is symptomatic on how this whole issue degenerated over all this time. Please, all tree look at my original bug report. I noticed a discrepancy between the rootskel and the finish-install list, suggested a solution which would allow in the future to fix this issue with having to change only one place (a most laudable goal for anyone knowledgeable in programming), and explicitly mentioned the 'unless i miss something', which was an invitation to comment from others of the team. The reply from Frans was "has this patch been tested", and saying it should not be applied before etch. This kind of reply from Frans has been of the kind which has set my teeths on edge since a loong time, one tries to work, to do best, to ask for comments, and mostly gets ignored. Now i am serene enough to not snap, but as i was in personal distress in spring, you can imagine how it affected me. Also, continuous repetitions of similar issues are not prone to appeasing such situations. Please think next time before you do such, and learn that team work is also about letting everyone speak, and discuss problems and possible solutions in the open, in order to find the best possible technical solution to a given problem. And with that, happy christmas to you all, and let's hope that next year will bring better understanding patience and goodwill to everyone concerned. Friendly, Sven Luther -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
