Bug#857063: debian-installer: Strips off partition from /dev/nvme* device names
Package: debian-installer Severity: normal Tags: d-i -BEGIN PGP SIGNED MESSAGE- Hash: SHA256 Hi, when using the Stretch RC2 installer on a ThinkPad X1 Carbon V5 with a built-in NVMe disk, I can’t resize an existing NTFS partition: > ntfsresize: ERROR(22): Opening '/dev/nvme0n1' as NTFS failed: Invalid argument According to lsblk, the partition is actually called /dev/nvme0n1p1, so d-i strips off the partition here. Sebastian - -- System Information: Debian Release: 9.0 APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.10.1+ (SMP w/4 CPU cores) Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: sysvinit (via /sbin/init) -BEGIN PGP SIGNATURE- iQIyBAEBCAAdFiEEdExETVib0Dz3caHc+HHcS2EGWqMFAli+48cACgkQ+HHcS2EG WqNRSg/3WAwBp8a6wrrEbSqAjjO1Ijz2j0kCFlT9MXEYo7S9xRu7RZZDy5T6WtAC zKPlpM3IEacXQbGOntPCE1Xpi5GyFKUC4S7ZYMoTKooxKP0mu3cpac4Z/rzHAWe0 /nv0+inS7qjgBdYmhgtt5sQAzlK6OsbgIH7GrQEYMCCTBBcOdT51ajjL7SHT7uQS MoN6BMcbosBkKsrtY+/YvKSQMPnjyUZc++is/O5FHMA/nqeopDcx1NTlWSlF4Xq0 NKwhKkHj2uclQopZVlfhcYiev3l3ZxO4YrAEEmPU6r7bfEv8alrFAF8Phy6dmFfm DGHjrbN+Dfm/K72BbTwlBOiuQKKnvnpTRN4HRxTbZR8kGNUzRS9pxI8sAM+N97IP cMR5sl1iV0KukaXVDyfxZ2NnyXzERX34g+AmiU6weEXpFY0y/EMZyTNLpKpDp6Gf NTKal//nldcZjeq7jr8PPMZzw0AenxgN093+BITrbcdis3btHtwoirrYmEw6cPX+ xWC9zv7eMDLG8J6Esbz9D9rlBxm+wb7H/rlg4qWhK/Q4kVK0xf0hJfmJ3EcvSk96 Shv6H5JO/WVEnLjA6oeMILWt3a9xNQIBpeCXj5bzUZBI4BHWkcsEf8qDr4QrCIk4 VoINSglnRacGhpsR4w8Zd7JnRT0fbdfeEAVYA0wZoNh3uZCFHw== =fhPC -END PGP SIGNATURE-
Bug#857064: debian-installer: Can't list partition types in fdisk
Package: debian-installer Severity: normal -BEGIN PGP SIGNED MESSAGE- Hash: SHA256 Hi, when using fdisk from within a rescue console in the Stretch RC2 installer, I can’t list available partition types: | Command (m for help): t | Hex code (type L to list all codes): L | sh: less: not found Linking /bin/more to /bin/less isn’t sufficient since busybox checks the process’ argv[0] and bails with “less: applet not found”. Creating a wrapper script with ‘more "$@"’ works, though. Sebastian - -- System Information: Debian Release: 9.0 APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.10.1+ (SMP w/4 CPU cores) Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: sysvinit (via /sbin/init) -BEGIN PGP SIGNATURE- iQIzBAEBCAAdFiEEdExETVib0Dz3caHc+HHcS2EGWqMFAli+5yIACgkQ+HHcS2EG WqMY/Q/7BOC4ccgs98PNwu6pR/wQYHVM83PCdh8hWCGx919SozW8SUSNh8tT+O3h 4wvuN4Z9rZro+lfh3l/bqwJEZtxVHLKn9o+7pRTvRiWNhVg/mingvmSHGjlmJX/6 XBN3uWU3zJThO52PWB2ZS7NukHIxTUAKLTWydBJVWy06le2dt/I4Qc79MjmI7oKP KAEPvqcEklmVOb2zzmgdV7xjoN/7riWVdAMbbFA0XQr1NwOgVxbGlIX8p6ZGNtY6 e38c9F4Djae9C3uyk5xozsXY+SuudwnTYV3nCHmk1MhuXe8HVpth/JPuYM05JsAz BoIV8fc9hgYWLNBy5KSknqTWH0KLk4DrKXH6TodXE9DELtu2Q3NS82yuXYsRNAjw UesiBPyPnBOlxdHK5DHEDbebtxftgjcXuOx+qptmKStQ5tmZlPPegJDlH0pyhqrY oDY63kQTpw+XGx+Fsk/pENEVk6Pu5XVRrXcdybCUxH6jKZD9UEcqCTqQ3DIXETie M5EH1qhD8mO6VoTJfEs6wQTsHtt23c463A+qQPj2igQpPM6je2VHUz+RvlYoCS7K pLVUs8Skzi4FPdEoHn7RvTiTU3Wma5crUoSSo6yjHvZXAZxrurJjI06QtlxStRTG w94N2h94RQzt0RwlXRl15XqP6Ku50vf1qIFzmZLPW5dS/Pl8F68= =5t+M -END PGP SIGNATURE-
Bug#857064: debian-installer: Can't list partition types in fdisk
On Thu, Mar 09, 2017 at 12:18:23PM +0100, Cyril Brulebois wrote:
> Hi,
>
> Sebastian Schmidt (2017-03-07):
> > when using fdisk from within a rescue console in the Stretch RC2
> > installer, I can’t list available partition types:
> > | Command (m for help): t
> > | Hex code (type L to list all codes): L
> > | sh: less: not found
> >
> > Linking /bin/more to /bin/less isn’t sufficient since busybox checks the
> > process’ argv[0] and bails with “less: applet not found”. Creating a
> > wrapper script with ‘more "$@"’ works, though.
>
> I'm not sure where this comes from, since list_types doesn't seem to be
> calling any external pager. But there's maybe something wrapping around
> the output more globally. Only looked briefly at util-linux/fdisk.c, and
> you can pretend I don't know anything about busybox anyway…
It’s in disk-utils/fdisk.c in void list_partition_types():
| } else {
| /*
| * Prints 1 column in format
| */
| size_t i;
|
| pager_open();
|
| for (i = 0; i < ntypes; i++) {
| const struct fdisk_parttype *t = fdisk_label_get_parttype(lb, i);
| printf("%3zu %-30s %s\n", i + 1,
| fdisk_parttype_get_name(t),
| fdisk_parttype_get_string(t));
| }
|
| pager_close();
| }
The setup is done in lib/pager.c in void __setup_pager():
| static void __setup_pager(void)
| {
| const char *pager = getenv("PAGER");
| struct sigaction sa;
|
| if (!isatty(STDOUT_FILENO))
| return;
|
| if (!pager)
| pager = "less";
| else if (!*pager || !strcmp(pager, "cat"))
| return;
|
| [...]
So setting $PAGER to more, cat or nothing in the shell should fix it.
signature.asc
Description: PGP signature
Bug#857063: debian-installer: Strips off partition from /dev/nvme* device names
merge 857063 820818 thanks On Thu, Mar 09, 2017 at 12:12:36PM +0100, Cyril Brulebois wrote: > > when using the Stretch RC2 installer on a ThinkPad X1 Carbon V5 with a > > built-in NVMe disk, I can’t resize an existing NTFS partition: > > > ntfsresize: ERROR(22): Opening '/dev/nvme0n1' as NTFS failed: Invalid > > > argument > > > > According to lsblk, the partition is actually called /dev/nvme0n1p1, so > > d-i strips off the partition here. > > I think this is a duplicate of: > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=820818#90 Oh, you are right. In fact, I tested the ISO image you linked there and can confirm it fixes this issue. Thanks! Sebastian signature.asc Description: PGP signature
Bug#857063: debian-installer: Strips off partition from /dev/nvme* device names
On Thu, Mar 09, 2017 at 07:45:34PM +0100, Cyril Brulebois wrote: > Sebastian Schmidt (2017-03-09): > > On Thu, Mar 09, 2017 at 12:12:36PM +0100, Cyril Brulebois wrote: > > > > when using the Stretch RC2 installer on a ThinkPad X1 Carbon V5 with a > > > > built-in NVMe disk, I can’t resize an existing NTFS partition: > > > > > ntfsresize: ERROR(22): Opening '/dev/nvme0n1' as NTFS failed: Invalid > > > > > argument > > > > > > > > According to lsblk, the partition is actually called /dev/nvme0n1p1, so > > > > d-i strips off the partition here. > > > > > > I think this is a duplicate of: > > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=820818#90 > > > > Oh, you are right. In fact, I tested the ISO image you linked there and > > can confirm it fixes this issue. > > Great, thanks. I'm uploading the fixed package right now, and will push > to git once alioth is back from fsck'ing. Ah, I thought my connection was faulty. :-) SGTM, thanks a lot! Sebastian signature.asc Description: PGP signature
Bug#396412: installation-reports: Installation report on AS/400 / iSeries / System i / i5
Package: installation-reports Severity: normal -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Boot method: Virtual CD-ROM (LODIMGCLG IMGCLG(MYCLG) DEV(OPTVRT01) OPTION(*LOAD)) Image version: etch with d-i beta3 (as of August 2006) - http://cdimage.debian.org/cdimage/etch_di_beta3/powerpc/iso-dvd Date: 27 Oct 2006 Machine: IBM iSeries / i5 / AS/400 / System i (whatever IBM calls it now) Model 570 Processor: POWER5+, 2.2GHz, dual-core Memory: 1275696 kB Partitions: sda1 PPC PReP Partition, sda2 /, sda5 swap Output of lspci -nn and lspci -vnn: Machine has no PCI Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [E] Detect network card:[O] Configure network: [ ] Detect CD: [O] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [E] Install base system:[O] Clock/timezone setup: [O] User/password setup:[O] Install tasks: [O] Install boot loader:[E] Overall install:[O] Comments/Problems: I tried to install Debian as a "guest" within an i5/OS (or OS/400) partition via the CRTNWSD (Create Network Server Description) command. Unfortunately, i5/OS refused to recognize neither the yaird image nor the kernel itself as a valid IPL source. However, turning the partition on via the HMC and booting into SMS worked. Note that the the NWS has still to be varied on via WRKCFGSTS *NWS. Else, virtual I/O devices (and most notably NWSSTG) are missing. After booting into SMS, I was able to choose the virtual CD-ROM drive as IPL source. Booting and installing was straightforward (in case, one chose the "expert64" or "install64" yaboot entry - i5 refuses to boot from 32 bit kernels and crashes into ofdbg displaying a nice register dump). Graphics were a bit garbled[1] on the console (via telnet 2301), but d-i was usable. D-i allowed me to make my root partition managed by LVM, what I chose on my first try. However, yaboot was unable to read its config from this partition, so I reinstalled (it was friday afternoon anyway) it without LVM on my /. yaboot was installed to a PReP partition (nothing notably until then) and the system rebooted. Unfortunately, I didn't start installation in expert mode, so I didn't notice that d-i only installed a 32 bit kernel. As expected, yaboot booted right into the open firmware debugger. After re-restarting the installation in expert mode, I hoped to find a 64 bit kernel in the installer's kernel selection list. However, only 32 bit kernels where listed, so I had to install a 64 bit version via a chroot environment in the command line. After rebooting, yaboot loaded the Elf64 kernel and all services started. However, a getty did not start on /dev/console (what made me a bit nervous first). I don't know if this is considered a bug or not... Anyway - nice job! I didn't expect the installation to go on that smoothly ;-) Footnotes = 1: http://kitt.yath.de/~yath/2006-10-31-d-i-1.png -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.2 (GNU/Linux) iD8DBQFFR3oavBVic1oTsEgRAq6UAJ9m8Tb8J7NSDJUqk2/fIuIGybYp9QCfe0a4 bwsmyTFGQ3ocvbbb8HvJu+M= =6CHk -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#396412: installation-reports: Installation report on AS/400 / iSeries / System i / i5
On Wed, Nov 01, 2006 at 07:23:22AM +0100, Frans Pop wrote: > Hmm. Nice to see a report for this. I didn't even know we supported AS/400 > boxes :-) Neither I did, but it was worth a try. :-) > The installer uses UTF-8 encoding. You may get a better display if you're > able to configure the remote console to use that. Oups. I'll try that out. > This probably could be made to work by using a separate /boot partition. Doesn't place yaboot its config in /etc? Anyways, a warning like "yaboot may be unable to read its configuration file" or something would have been nice. > I'm not quite sure how the kernel selection for powerpc works, but we > should be able to fix this. > Could you send us the output of 'cat /proc/cpuinfo'? # cat /proc/cpuinfo processor : 0 cpu : POWER5+ (gs) clock : 2197.80MHz revision: 2.1 (pvr 003b 0201) processor : 1 cpu : POWER5+ (gs) clock : 2197.80MHz revision: 2.1 (pvr 003b 0201) timebase: 27506 machine : CHRP IBM,9406-570 # > > After rebooting, yaboot loaded the Elf64 kernel and all services > > started. However, a getty did not start on /dev/console (what made me a > > bit nervous first). I don't know if this is considered a bug or not... > > What did you change in /etc/inittab to fix this. just started a getty on /dev/console by putting this into the inittab: 0:2345:respawn:/sbin/getty 38400 console > If you used a serial console, what is its name? Is it /dev/ttyS? or > something fancy? I really have no idea. I have found out two ways to get a console: a) via "Open a terminal window" in the partitions's context menu in the HMC or b) connecting via telnet to the i5/OS partition and selecting a "remote panel" for the guest partition. I don't know what type of console are either of them. Maybe this is some kind of serial console (I have configured virtual serial adapters from the linux to the i5/OS partition). I will investigate on that console thing further next week. Greetings, Sebastian -- signature intentionally left blank pgpk8NjI3ftyK.pgp Description: PGP signature
