Re: USB Install Fails, Complains about CD-ROM
Kent West wrote: > I have a Dell Latitude E7250 laptop. I'm trying to install Debian to it using > a USB stick. [...] > The real problem is that after going through the first three or four > screens, the install halts, complaining about not being able to read > the CD-ROM. Yes, I ran into the same problem with my E7250, using a netinst image and a live image. When trying to boot the live image I noticed these messages on the console: usb 1-2: device descriptor read/64, error -110 ... usb 1-2: reset high-speed USB device number 2 using xhci_hcd I tried booting from a different USB port; that worked once, but on a subsequent attempt it failed with the same issue. I did eventually get Debian installed, using a convoluted process that I won't describe here because I suspect it's not relevant (see below). I'm afraid that once I got Debian installed, I never went back to figure out what was causing the earlier failures. After reading your posting, I tried again to create a live USB stick. I used cp to copy the image to /dev/sdb, and I ran sync afterwards. Same failure. I then tried using dd instead of cp, and I used a different USB port on my desktop to write to the USB stick. This time I could boot the live image. I checked kern.log, and it had usb 2-2: device descriptor read/64, error -110 but did not have the "reset" message. (And I notice that this time it's "usb 2-2" rather than "usb 1-2".) I tried to boot off the USB stick a second time, and this time I got the failure again. So I suspect the problem has to do with peculiarities of the E7250, rather than issues writing the USB stick. I have a copy of kern.log from the first (successful) boot, and I have the dmesg output from the second (failed) boot. If anyone wants to look at them, let me know. mike
Re: rsync - newbie question
On Saturday, May 12, 2018 02:54:53 PM Richard Owlett wrote: > But you started me thinking about how my use case differs from any > normal user. Yup, it appears so. ;-)
Re: minimal installation
Dan Norton composed on 2018-05-12 17:47 (UTC-0400): > Now, I want to update and upgrade and start to slowly add packages, but > my network ignorance is getting in the way. > Tried to run "/sbin/ifconfig -a" but ifconfig is not found. Deprecated over a decade ago, and finally dropped from basic installation. Use its replacement ip instead. > Please point me to how to get ethernet up. Thanks. https://wiki.debian.org/NetworkConfiguration should do it. -- "Wisdom is supreme; therefore get wisdom. Whatever else you get, get wisdom." Proverbs 4:7 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/
Re: minimal installation
On Tue, 8 May 2018 00:10:50 +0100 Brian wrote: > On Mon 07 May 2018 at 18:37:16 -0400, Dan Norton wrote: > > > On Mon, 7 May 2018 23:22:46 +0100 > > Brian wrote: > > > > > On Mon 07 May 2018 at 18:04:47 -0400, Dan Norton wrote: > > > > > > > On Mon, 7 May 2018 10:42:17 -0400 > > > > Felix Miata wrote: > > > > > > > > > jpff composed on 2018-05-07 12:34 (UTC+0100): > > > > > > > > > > > Felix Miata wrote: > > > > > > > > > > >> My Debian installations are all net installs that > > > > > >> include > > > > > > > > > > >>tasks=standard > > > > > >> base-installer/install-recommends=false > > > > > > > > > > >> on the kernel cmdline. I get nothing I don't need installed > > > > > >> that way. Xorg and whatever else I need I get with apt* > > > > > >> once booted normally. > > > > > > > > > > > That looks interesting; it attemts to answer my deep problem > > > > > > about no X, xdm, xterm etc. > > > > > > > > > > > My problem nowis I do not know where/how to apply this. I > > > > > > have not seen any mention of a kernel command line in the > > > > > > net install. More please! > > > > > > > > > > I'm not up to speed on the conventional HOWTO for answering > > > > > this. I rarely use conventional installation boot media. > > > > > Virtually all my installs are in multiboot environments. This > > > > > enables installation booting by using a bootloader already > > > > > present on the system, by loading an installation kernel and > > > > > initrd, complete with the parameters mentioned, plus several > > > > > others, such as network configuration, and leaving off quiet > > > > > and splash=silent. > > > > > > > > > > IME, virtually any distro's installation media when its > > > > > presence first appears on screen allows for some method of > > > > > appending parameters to the kernel cmdline. It may be an "e" > > > > > key, or an ESC key, or an up or down arrow key, or a function > > > > > key, and likely will suggest how when its screen first > > > > > paints. > > > > > > > > That "kernel cmdline" phrase is a point of confusion. By editing > > > > the "Install" item in the netinst menu, I can change: > > > > > > > > linux /install.amd/vmlinuz vga=788 --- quiet > > > > > > > > ...to this, all on one line: > > > > > > > > linux /install.amd/vmlinuz vga=788 --- tasks=standard > > > > base-installer/install-recommends=false > > > > > > > > Is that going to result in a minimum installation? > > > > > > Leave off tasks=standard for more minimalness. > > > > > > > This is getting exciting. Will I still get a command line? > > If you mean at first boot; yes. > > I preseed in a file with "tasksel tasksel/first multiselect", which > means no task is selected for installation. I've never done it from > a prompt. Perhaps "tasks="? > With the ethernet unplugged, installing with a netinst iso, I do get a minimal installation. With ethernet plugged in, the installed kernel can be booted repeatedly, presenting a login prompt each time. adduser, mount, cat, lp... just work. Now, I want to update and upgrade and start to slowly add packages, but my network ignorance is getting in the way. Tried to run "/sbin/ifconfig -a" but ifconfig is not found. Anyway, ran: dhclient lo dhclient eth0 ...with no effect (could not ping google) Comparing files on this install with the same on this one (where I'm writing this post): /etc/network # same contents on both after I added a link: run -> /network /etc/network/interfaces.d # empty on both /run/network # same on both /var/log/syslog # doesn't show anything obvious to me root@debm:/etc/network# cat interfaces # This file describes... # and how to ... source /etc/network/interfaces.d/* # The loopback network interface auto lo iface lo inet loopback root@debm:/# /etc/init.d/networking stop Stopping networking (via systemctl): networking.service. root@debm:/# /etc/init.d/networking start Starting networking (via systemctl): networking.service. root@debm:/# ping -c 3 http://www.google.com ping: http://www.google.com: Temporary failure in name resolution Please point me to how to get ethernet up. Thanks. - Dan
Re: rsync - newbie question
Yes, rsync has a “-x" option, which does the same thing as for cp: it keeps it from crossing filesystem boundaries. If you are using rsync to back up whole filesystems, it’s indispensable. Rick On May 12, 2018, at 10:50 AM, Tixy wrote: > Some commands have options to stop them looking at other filsytsems (cp > has -x and find has -xdev) I don't know about rsync (I don't have it > installed).
Re: rsync - newbie question
On Sat, 2018-05-12 at 13:58 -0500, Richard Owlett wrote: > On 05/12/2018 12:50 PM, Tixy wrote: > > On Sat, 2018-05-12 at 13:28 -0400, Jude DaShiell wrote: > >> Another hierarchy in Linux not to sync is /system for the same reason > >> you don't sync /proc. > > > > Presumably you meant /sys ? > > > > Basically, the OP probably don't want to try and sync mount points for > > things that aren't ordinary filesystems and that's quite an extensive > > and variable list. > > I exclude /media for similar reasons. Or would exclude it anyway? The command options I was thinking about would stop a command crossing over into another filesystem. E.g. if my desktop automounts a USB stick as /media/tixy/disks-label, that directory path would be part of the root filesystem, but the actual contents of the USB stick that you see under that path wouldn't be included in the command. So you'd just get an empty directory with that path (/media/tixy/disks-label). I googled for rsync man page [1] and it does seem to have the same option as the cp command for this: -x, --one-file-system This tells rsync to avoid crossing a filesystem boundary when recursing. This does not limit the user's ability to specify items to copy from multiple filesystems, just rsync's recursion through the hierarchy of each directory that the user specified, and also the analogous recursion on the receiving side during deletion. I don't know what it is you're trying to achieve. If your intent is to create an identical copy of a system as some kind of backup, you'll also need to make sure you copy file permissions, owners and special attributes, duplicate hardlinks and possibly other things I haven't thought of. That will require root privileges to execute and the correct set of commandline options. Looking at that rsync man page I found, that would be options like -p -X -H. (Note, I haven't used rsync before so don't rely on my advice too much). [1] https://linux.die.net/man/1/rsync -- Tixy
Re: USB Install Fails, Complains about CD-ROM
Hi, Pascal Hambourg wrote: > Why then does parted complain about a block size discrepancy ? Because the Apple Partition Map announces to count blocks with size 2048 whereas the Linux device file announces 512 (via ioctl(BLKSSZGET) ?)). It is quite a poor choice of parted to hop on the Apple Partition Map rather than onto the MBR partition map, which will be of interest for much more firmwares and operating systems. (I should really pester debian-cd more about that inappropriate option -isohybrid-apm-hfsplus in the xorriso run.) I wrote: > > > But if only a few MB were copied (for what strange reason ever) there > > remains the riddle why it booted and probably loaded the initrd. > > Why does it fail when it looks for the ISO filesystem ? > Probably because the MBR and the boot files are located at the beginning of > the image and were part of the data which were actually written. Well, Kent obviously experienced a running Linux kernel and the attempt to execute software from the initrd. The message "Load installer components from CD" can be found in initrd.gz as package Description which belongs to "load-cdrom" https://packages.debian.org/stretch/load-cdrom This package is not to see as file in the ISO. It could be that there is no problem with finding the "CD-ROM" but rather with installing or activating that package. So maybe it could not even begin its work. About its role in debian-installer https://d-i.alioth.debian.org/doc/talks/debconf6/paper/ says "The first three stages are where the fundamental difference between installation methods can be seen. All components (udebs) used there need to be included in the initrd[2] with which the installer is booted." and lists "load-cdrom (anna)[...]Retrieve and unpack additional components" as part of stage 2. --- So the problem obviously occurs while the installer is still using the initrd. The filename in the ISO is /install.amd/gtk/initrd.gz or /install.amd/initrd.gz. In debian-9.4.0-amd64-xfce-CD-1.iso the initrds and the kernel are stored in the lower blocks indeed. ("Startlba" is the logical block address with 2048 bytes per block. "Blocks" is the number of occupied blocks.): Startlba , Blocks , Filesize , ISO image path === ... 5642 ,19182 , 39282964 , '/install.amd/gtk/initrd.gz 24824 ,1 , 66 , '/install.amd/gtk/install.b 24825 , 2063 , 4224800 , '/install.amd/gtk/vmlinuz' 24825 , 2063 , 4224800 , '/install.amd/vmlinuz' 26888 , 7667 , 15700220 , '/install.amd/initrd.gz' 34555 ,1 , 45 , '/install.amd/install.bat' ... In debian-buster-DI-alpha2-amd64-xfce-CD-1.iso : Startlba , Blocks , Filesize , ISO image path === ... 5694 ,19570 , 40077490 , '/install.amd/gtk/initrd.gz' 25264 ,1 , 66 , '/install.amd/gtk/install.bat' 25265 , 2181 , 4466448 , '/install.amd/gtk/vmlinuz' 25265 , 2181 , 4466448 , '/install.amd/vmlinuz' 27446 , 7811 , 15995172 , '/install.amd/initrd.gz' 35257 ,1 , 45 , '/install.amd/install.bat' ... The messages Load installer components from CD There was a problem reading data from the CD-ROM. are well known to the web. But occurences show no pattern about particularly bad releases or spin-off distros. The reported remedies are not much convincing as technical improvements, but they all involve re-copying the ISO onto USB stick. I.e. they all could simply have helped by trying again and having more success with copying. Kent could try to damage the USB stick content for experiments. First: # Damage ISO after the end of initrds and kernel dd if=/dev/zero bs=2048 count=35 seek=5 of=/dev/sdc And after checking out the effects of that damage: # Damage the GTK initrd itself dd if=/dev/zero bs=2048 count=4 seek=1 of=/dev/sdc (This does not damage the USB stick. Only its content data.) --- Have a nice day :) Thomas
Re: rsync - newbie question
On Sat 12 May 2018 at 13:54:53 -0500, Richard Owlett wrote: > I'll keep that in mind. > But you started me thinking about how my use case differs from any normal > user. You would have to put yourself in the thought mode of a normal user (whatever he is) instead of seeing youself as having special needs or requirements. -- Brian.
Re: rsync - newbie question
On 05/12/2018 12:50 PM, Tixy wrote: On Sat, 2018-05-12 at 13:28 -0400, Jude DaShiell wrote: Another hierarchy in Linux not to sync is /system for the same reason you don't sync /proc. Presumably you meant /sys ? Basically, the OP probably don't want to try and sync mount points for things that aren't ordinary filesystems and that's quite an extensive and variable list. I exclude /media for similar reasons. Or would exclude it anyway? Some commands have options to stop them looking at other filsytsems (cp has -x and find has -xdev) I don't know about rsync (I don't have it installed).
Re: rsync - newbie question
On 05/12/2018 12:48 PM, Hans wrote: Am Samstag, 12. Mai 2018, 19:37:40 CEST schrieb Richard Owlett: Please note, the directory is NOT /system, it is /sys. Juda got a little typo. :) I won't complain too much. Otherwise peuple will start talking about mine ;/ However, I would avoid /proc, /sys, /tmp and /lost+found Hint: If you might put /home on another partition, you can easily install or sync a new system, but leave the user settings. In that case, you should of course also exclude /home. I'll keep that in mind. But you started me thinking about how my use case differs from any normal user. Best Hans On 05/12/2018 12:28 PM, Jude DaShiell wrote: Another hierarchy in Linux not to sync is /system for the same reason you don't sync /proc. thank you.
Re: e-mail addresses not being redone
On 2018-05-12 09:18 AM, Jude DaShiell wrote: On Fri, 11 May 2018, Gary Dale wrote: Date: Fri, 11 May 2018 17:30:26 From: Gary Dale Reply-To: g...@extremeground.com To: debian users Subject: e-mail addresses not being redone Resent-Date: Fri, 11 May 2018 21:30:56 + (UTC) Resent-From: debian-user@lists.debian.org I'm running Debian/Buster on an AMD64 system. When I use "mail" to send e-mail from the command line, it doesn't use /etc/email-addresses to rewrite the "from" header but s-nail does. Apparently on my system, mail is provided by the Gnu Mailutils. Is the the normal behaviour for Mailutils or do I have to configure it to respect /etc/email-addresses? You may have had exim installed and in that case will have to adjust things in exim. Yes. I have exim4 installed. I even told it to rewrite the from header to use the ISP's domain but it also seems to ignore that.
Re: rsync - newbie question
Am Samstag, 12. Mai 2018, 19:37:40 CEST schrieb Richard Owlett: Please note, the directory is NOT /system, it is /sys. Juda got a little typo. :) However, I would avoid /proc, /sys, /tmp and /lost+found Hint: If you might put /home on another partition, you can easily install or sync a new system, but leave the user settings. In that case, you should of course also exclude /home. Best Hans > On 05/12/2018 12:28 PM, Jude DaShiell wrote: > > Another hierarchy in Linux not to sync is /system for the same reason > > you don't sync /proc. > > thank you.
Re: rsync - newbie question
On Sat, 2018-05-12 at 13:28 -0400, Jude DaShiell wrote: > Another hierarchy in Linux not to sync is /system for the same reason > you don't sync /proc. Presumably you meant /sys ? Basically, the OP probably don't want to try and sync mount points for things that aren't ordinary filesystems and that's quite an extensive and variable list. Some commands have options to stop them looking at other filsytsems (cp has -x and find has -xdev) I don't know about rsync (I don't have it installed). -- Tixy
Re: USB Install Fails, Complains about CD-ROM
Le 11/05/2018 à 22:38, Thomas Schmitt a écrit : Pascal Hambourg wrote: It also has Apple and GPT partition tables, but they are bogus The GPT is not valid because there is a non-"protective" MBR partition table. The APM is valid, but should be of no interest for any firmware that does not expect a HFS or HFS+ filesystem. (And there is no such filesystem in the ISO image.) Why then does parted complain about a block size discrepancy ? But if only a few MB were copied (for what strange reason ever) there remains the riddle why it booted and probably loaded the initrd. Why does it fail when it looks for the ISO filesystem ? Probably because the MBR and the boot files are located at the beginning of the image and were part of the data which were actually written.
Re: rsync - newbie question
On 05/12/2018 12:28 PM, Jude DaShiell wrote: Another hierarchy in Linux not to sync is /system for the same reason you don't sync /proc. thank you.
Re: rsync - newbie question
On Sat, 12 May 2018, Richard Owlett wrote: Date: Sat, 12 May 2018 11:54:13 From: Richard Owlett To: debian-user Subject: Re: rsync - newbie question Resent-Date: Sat, 12 May 2018 15:54:40 + (UTC) Resent-From: debian-user@lists.debian.org On 05/12/2018 10:47 AM, Eero Volotinen wrote: You should not sync /proc. it's not normal directory Eero Thank you. -- Another hierarchy in Linux not to sync is /system for the same reason you don't sync /proc.
Re: USB Install Fails, Complains about CD-ROM
Le 12/05/2018 à 01:04, Rick Thomas a écrit : After doing the cp or dd to write the .iso to the USB, do you do a “sync” before you eject it? I don't, because I don't feel the need to. According to its man page description, sync "flush file system buffers", but the destination is a raw device, not a filesystem, so filesystem buffers should not be involved. Writing to a USB stick can seem to go quite fast Not in my experience when writing to the raw USB device. The USB drives I use have an activity light. It stops blinking as soon as the dd or cp command terminates. I also monitor block I/O with vmstat and block writes drop as soon as dd or cp terminates. If I run sync, I do not see any more activity from the light or vmstat.
Re: rsync - newbie question
Am Samstag, 12. Mai 2018, 17:54:13 CEST schrieb Richard Owlett: As Eero said, do not sync /proc, you can use the --exclude option. There is a good description here, how to exclude things: https://www.thegeekstuff.com/2011/01/rsync-exclude-files-and-folders/? utm_source=feedburner Have fun! Hans > On 05/12/2018 10:47 AM, Eero Volotinen wrote: > > You should not sync /proc. it's not normal directory > > > > Eero > > Thank you.
Re: rsync - newbie question
You should not sync /proc. it's not normal directory Eero la 12. toukok. 2018 klo 18.37 Richard Owlett kirjoitti: > In another thread it was suggested that I use:> rsync -avzh --delete > -n > > I tried it and got ~200 error messages of form: > > > file has vanished: "/proc/10/exe" > > > file has vanished: "/proc/10/task/10/exe" > > > file has vanished: "/proc/101/exe" > > > file has vanished: "/proc/101/task/101/exe" > > > file has vanished: "/proc/10146/exe" > > > file has vanished: "/proc/10146/task/10146/exe" > > > file has vanished: "/proc/10300/cwd" > > > file has vanished: "/proc/10300/root" > > > file has vanished: "/proc/10300/exe" > > Spot checking properties of those showed that although they had multiple > sub-directories the content was "0 bytes". > > Browsing [https://manpages.debian.org/stretch/manpages/proc.5.en.html] > implied that excluding everything under /proc would be reasonable. > > Am I correct? > TIA > > >
Re: rsync - newbie question
On 05/12/2018 10:47 AM, Eero Volotinen wrote: You should not sync /proc. it's not normal directory Eero Thank you.
rsync - newbie question
In another thread it was suggested that I use:> rsync -avzh --delete -n I tried it and got ~200 error messages of form: file has vanished: "/proc/10/exe" file has vanished: "/proc/10/task/10/exe" file has vanished: "/proc/101/exe" file has vanished: "/proc/101/task/101/exe" file has vanished: "/proc/10146/exe" file has vanished: "/proc/10146/task/10146/exe" file has vanished: "/proc/10300/cwd" file has vanished: "/proc/10300/root" file has vanished: "/proc/10300/exe" Spot checking properties of those showed that although they had multiple sub-directories the content was "0 bytes". Browsing [https://manpages.debian.org/stretch/manpages/proc.5.en.html] implied that excluding everything under /proc would be reasonable. Am I correct? TIA
Re: LVM setup with snapshots
On 11-05-2018 21:46, Forest Dean Feighner wrote: > I really didn't prepare for lvm. I never used lvm before this so had > no idea of lvm before. > > Snapshots sound like an awesome idea. > > I would like to do a configured base install, create a snapshot, and > modify (fork), the base for different things. > > With 20/20 hindsight. The default doesn't seem to have room. What are > different solutions other debian/lvm users have used? You can shrink most kinds of partitions (including ext4), and then shrink the logical volumes (in this order). It's not as convenient as growing the LVs since the partitions must not be mounted and shrinking can be somewhat slow if data needs to be shuffled around, but it's possible. As always, there's a small risk of data loss, so better have a backup of important data. -- She is descended from a long line that her mother listened to. -- Gypsy Rose Lee Eduardo M KALINOWSKI edua...@kalinowski.com.br
Re: USB Install Fails, Complains about CD-ROM
On Fri 11 May 2018 at 15:49:59 -0500, Kent West wrote: > I no longer have a failing setup, but this Ubuntu user was seeing the exact > same thing I was seeing, except in Debian words/colors: > > > https://askubuntu.com/questions/127398/usb-drive-install-of-ubuntu-12-04-server-fails-cant-find-components-from-cd-r > > In case of link-rot, here's what he writes: > > >1. The computer boots up the installation process ok. >2. It gets through the Ubuntu language, locale and keyboard selection. >3. Then starts loading additional components. At this point it gets >about a quarter of the way through then throws big error message saying: > > *[!!] Load installer components from CD* > > There was a problem reading data from the CD-ROM. Please make sure it is in > the drive. If retrying does not work., you should check the integrity of > your CD-ROM. > > Failed to copy file from CD-ROM. Retry? This user used Universal-USB-Installer-1.8.9.4 to write a Ubuntu 12.04 Server (32bit) image to a USB stick - a somewhat different technique from yours. The tool used to create the USB image quite possibly modified and damaged the image when writing it out. AfAIK, there is no well-documented account of dd, cat or cp reproducibly failing to write a Debian installation image properly to a USB device. -- Brian.