Bug#705961: installation-reports: debian-installer does not create an EFI partition by default
Hi Steve, I am afraid I can not repeat the whole installation. I have copied already all teh files I had in my old computer to the new one, so it would be a mess to repeat the whole process again. I have made an exercise, but I don't think it is very useful. I just repeated the first steps of the installation (with cdnetinst of 20130423) until the step of the partition: - I started the Linux CD through the UEFI loader - I got the weird error message of the prefix while loading the installer grub (I remember to see this one also during a former installation) - In the partitions step, I deleted my current Linux partitions and asked the guided partition to propose new partitions. It made just one partition for Linux and one for swap. Then I did not want to commit the changes proposed, so I just left. But I think to remember that in my former installation the program issues a warning after that saying that I have no EFI partitions defined (while actually there is one of ~300 MB in fat32, the Windows one). I am sorry I can not go beyond this point. I have the suspicion that the problem may have to do with an HP loader that is what actually use to boot Windows if I do not pass through Linux GRUB. Anyway, if nobody else complains it mens that this was an effect of the failed installation I made with Squeeze before. You can close the bug, at least from my side. Thanks a lot for your help. Cheers, Jorge I believe so. I certainly had that intention, and I don't think the d-i would have seen the former partitions as EFI if booting in legacy mode, right?I remember to see that weird message prefix not found, but I do not remember in which boot. I am sorry I could nto send this report before, I would have this information more fresh in my mind. I hope it helps, but if you do not receive any similar report it may have to do with the two trials of installation before the succesful one. Yes, I think so. If you're prepared to delete your existing Debian installation and try again from scratch using Wheezy, that's more likely to work. But I understand if you don't want to spend the time on that now...!
Bug#705961: installation-reports: debian-installer does not create an EFI partition by default
Hi Steve Thanks for your help. Yes, I was surprised of this problem because I did not see anything in the release notes, despite of having some information on the UEFI issues. Let me extend a bit more on the process followed: The PC came with Windows 8 pre-installed. That implies also a recovery partition, so grub now offers two Windows starts: one for normal Windows, one for the recovery partition (I guess this is standard). Anyway, I tried to boot from the installation CD of squeeze. I had to do that through the legacy mode and deactivated the secure booting just in case. d-i did not notice that there were EFI partitions, and installed linux in the space I freed from the disk. No need to mention that I could not boot the linux partition even from the legacy mode. Then I decided to move on to Wheezy. I formatted the linux partition to prepare the new installation (see below) El 22/04/13 22:05, Steve McIntyre escribió: On Mon, Apr 22, 2013 at 09:29:57PM +0200, Jorge Sanz Forcdada wrote: Package: installation-reports Severity: normal Hi Jorge, I must admit that I'm very surprised to see this bug report - I've written a lot of the amd64 UEFI support code in debian-installer, and it's been working just fine for me in testing. So, if you could answer a few questions for me that would help enormously in working out what's gone wrong here. 1. You say that you want to install on a disc with Windows 8 - is Windows 8 installed there already? If so, then the installer code *should* pick up on the existing EFI system partition that Windows will have created, and use it accordingly I'm guessing you didn't already have Windows 8 installed, from the information further down. If it doesn't find an exiting EFI system partition, d-i should create one itself automatically. Yes, Windows 8 was installed before. When I tried the first installation of Wheezy the bios only showed the UEFI Windows boot (and the linux CD too), so I could not boot the new linux installation and decided to repeat the installation preparing a new EFI partition. The debian installer saw the Windows EFI partition already present in the disk before installation. Besides, I think it recognized (?!) the former linux squeeze partition as an EFI partition (not 100% sure of this). I must say that during the whole process I kept a doubt about a partition of 1 MB that I believe was created by squeeze, but I prefered not to delete it just in case it was a bios thing (it still there, it does not bother :-) ). 2. Are you *100%* sure that you booted the installer in UEFI mode? You can check this by looking at startup messages as the machine boots. If it's booting via UEFI, you'll get a cosmetic complaint from grub at early boot: prefix not found. I believe so. I certainly had that intention, and I don't think the d-i would have seen the former partitions as EFI if booting in legacy mode, right?I remember to see that weird message prefix not found, but I do not remember in which boot. I am sorry I could nto send this report before, I would have this information more fresh in my mind. I hope it helps, but if you do not receive any similar report it may have to do with the two trials of installation before the succesful one. Thanks for pointing about the other bug on the grub not booting the Windows 8. I will fix it now. I agree with you that the graphics issue is not related, I just mentioned it just in case somebody wants to insist in the installation docs (I think it is written already) about the need of using non-free packages to make some graphic cards to work. Newbies use to be discouraged of using Debian because of this kind of things Thanks again. Please let me know if I can help further. Jorge Boot method: CD netinst Image version: debian-testing-amd64-netinst.iso (version 20130417) Date: 2013-04-17, 18:00 UT (20:00 CET) Machine: HP Pavillion p6-2306es, Intel core i5, 6 GB RAM Partitions: rootfs rootfs653954576 156762724 463972884 26% / udev devtmpfs 10240 0 10240 0% /dev tmpfs tmpfs608444 664 607780 1% /run /dev/disk/by-uuid/23716695-21dc-4f05-8429-291f7621f862 ext4 653954576 156762724 463972884 26% / tmpfs tmpfs 5120 05120 0% /run/lock tmpfs tmpfs 2466160 292 2465868 1% /run/shm /dev/sda7 vfat 34260 117 34144 1% /boot/efi Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[O] Configure network: [O] Detect CD:
Bug#407998: ds9 breaks with Iraf in etch
Hi Justin, You ask me about the problem with DS9. When the ds9 is started (for instance from an xterm) it opens normally. Then if a display command is executed in Iraf, an error message starts to show up in the xterm and goes on until a ^C is done. The message was something like Error: IIS iisIO problems There is a thread of something like this in iraf.net http://iraf.net/phpBB2/viewtopic.php?t=85566highlight=0b8 In my case I installed iraf from the linux binaries v2.12.2 (including noao binaries too) provided in the Iraf page since the old iraf packages for Debian would likely give problems with these libraries, I guess. Iraf works well in general, but I found this problem with the display. I can tell you about other two problems just in case you plan to resurrect the iraf debian package :-) - First is about ecl, the new cl package, It does not work because it looks for the old termcap library file. In a former version of debian there was a termcap-compat package but it is not available anymore. - Second is a problem with the x11iraf package. In general it works fine with the binaries installation (always using the linux binaries, not Red Hat or anything similar), but ximtool doesn't even open. When I try to build from source then at some point it gives me a message that I don't know how to solve. It was something about a variable that is no longer used in the modern version of the libraries. And then the problem with ds9, but as I said it works fine with the 4.0b10 version. Thanks a lot, Jorge -- Dr. Jorge Sanz Forcada LAEFF, European Space Astronomy Centre Apartado 50727, E-28080 Madrid (Spain) Phone: +34-918-131-267, Fax: +34-918-131-160 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#407998: saods9: update needed for Debian etch
Package: saods9 Version: 4.0b7-1.4 Severity: important Hi Jorge, On Mon, Jan 22, 2007 at 08:10:26PM +0100, Jorge Sanz Forcada wrote: Hi Justin, I am a Debian user who has installed Debian Etch recently, and I have found a problem that breaks the DS9 deb package when used with Iraf (I haven't tested it with other situations). What precisely is the problem? Also, how did you install iraf? Did you use NOAO binaries, my old iraf package, compile from source, ...? I write directly to you because it will take much more time to file a bug (I -haven't ever tried before) It's pretty easy (if not intuitive) to file bugs; just mail the submit address with (minimally) a Package: pseudoheader. The subject will become the bug title; it is conventionally prefixed with the package name and a colon. I'm doing so now, so I don't forget the bug. Cheers Justin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
