/srv/irclogs.ubuntu.com/2020/08/11/#ubuntu-server.txt

=== tds9 is now known as tds
sarnoldpymagic: what are you seeing?03:06
pymagicsarnold, boot repair gives me this: http://paste.ubuntu.com/p/d6qyFx9mY2/ -- I am trying to boot from nvme0n1p203:07
pymagicBooting just boots me into bios03:07
sarnoldbummer :( this probably would have been a few seconds to fix before the reboot but at this point will require booting to rescue media to fix03:09
amurraypymagic: I think you need to set the boot order to specify Ubuntu03:10
pymagicsarnold, I've rescue media03:10
pymagicI booted into kubunt and can mount the drive03:10
pymagicI tried boot-repair - no use03:10
sarnoldpymagic: start with these "physical machine" instructions to get into a rescue environment, chrooted to your install: https://wiki.ubuntu.com/SecurityTeam/KnowledgeBase/GRUB2SecureBootBypass#Physical_machine03:10
pymagicI can try those commands03:11
sarnoldpymagic: yeah, I don't think anyone's taught boot-repair how to fix incorrectly configured debconf settings -- once you're in the rescue environment, chroot'ed to your installation, sudo dpkg-reconfigure grub-efi-amd6403:12
pymagick - trying03:12
pymagicbooting into kubuntu - will take some time, thanks03:12
pymagicsarnold, what will be ROOT_PARTITION value?03:13
sarnoldpymagic: it'll be the largest partition on your nvme drive; that might be nvme0n1p2 but check gdisk -l /dev/nvme0n1 to double-check03:14
pymagicyes, it is that03:14
pymagicsarnold, grub-efi-amd64 is not installed03:21
pymagicI could do those steps in prepare the chroot03:22
sarnoldpymagic: are you sure you're inside the chroot? line 12 of your paste:03:23
sarnoldThe default repair of the Boot-Repair utility will reinstall the grub-efi-amd64-signed of03:23
sarnold.. so I certainly expected it to be installed :)03:23
pymagicThe paste I did came out of boot-repair utility - maybe they installed it? uninstalled it?03:23
pymagicnot sure03:23
pymagicAll the steps in chroot - went through cleanly for me03:23
pymagiclast command was mount /boot/efi03:24
sarnoldtry an apt install grub-efi-amd64   perhaps? see what that does?03:24
pymagicsarnold, if i do dpkg --list | grep nvidia - I do see driver-440 - which is what was installed on my machine03:25
pymagicI also see other packages that I installed that are not on kubuntu03:25
sarnoldpymagic: aha, good idea03:25
pymagicI do have grub-gfxpayload-liss and grub-pc installed03:26
pymagicthey are going to be removed if i install grub-efi-amd64 - ok03:26
sarnoldgrub-pc? that's for BIOS machines, not EFI machines03:26
pymagicI've not clue how it was working so far03:26
sarnoldme neither03:26
pymagicI had selected UEFI in the bios and could boot03:27
pymagicit says EFI Variables are not supported on this system.../sys/firmware/efi/efivars not found, aborting03:27
pymagicgrub-probe : error: cannot find a GRUB drive for /dev/sdc1 - check your device.map03:27
pymagicnot sure why sdc1 is even a candidate  for grub03:28
sarnoldcurious and curiouser.. try getting another terminal, and check the /sys/firmware/efi/efivars directory from outside the chroot03:28
pymagicnothing else03:29
sarnoldman this is confusing03:29
sarnoldefivarfs on /sys/firmware/efi/efivars type efivarfs (rw,nosuid,nodev,noexec,relatime)03:29
sarnoldthat's what I've got on mine..03:29
pymagicI dont03:30
pymagicI have /sys/firmware/ -> acpi dmi and memmap03:30
pymagicThis is a new Ryzen 3900X X570 machine. AORUS bios03:30
sarnoldmaybe try a mount -tefivars efivars /sys/firmware/efi/efivars/    ... I don't know if the --bind means it's got to be done from outside the chroot, or if it can go from inside the chroot ..03:30
sarnoldand if that works, then apt install -f  ?03:31
pymagiccould it be that i booted using "Legacy" instead of UEFI into kubuntu03:31
pymagiccould that do it?03:31
sarnoldyeah, that'll cause chaos :)03:31
pymagicreboot and try again?03:32
sarnoldif you *want* to keep the legacy thing, then reinstall that grub-pc! if you want to switch to uefi, then, yeah, reboot, flip the switch..03:32
pymagicLet me uefi - rebootin03:32
pymagicbios is uefi :( csm support = on, storage boot = uefi , other pci device = uefi03:33
pymagicIn kubuntu, there is an option in grub which says "UEFI firmware settings" - when i clicked it, it took me to bios...where uefi settings were03:35
sarnoldpymagic: aha, turn off csm and try again :)03:38
sarnoldcsm == legacy03:38
pymagicI thought csm = uefi option?03:38
sarnoldcsm is some cursed thing :)03:38
pymagick - disabling it and trying again03:38
pymagicsarnold, Intramfs -unpacking failed - decoding...its trying to boot now03:46
pymagicsarnold, but /sys/firmware now has efi :)03:46
pymagicsarnold, now it says grub-efi-amd64-signed is already newest version03:47
pymagicupdate-grub?03:48
pymagicor perhaps update-grub2? not sure03:49
pymagicI reinstalled that package using --reinstall03:51
pymagicwill try to reboot and see if that wrks03:51
sarnoldpymagic: so, next up is dpkg-reconfigure grub-efi-amd64  and make sure it thinks to install it in the correct place03:53
pymagicah - k - i'll have to get back into chroot again. I rebooted after i --reinstalled it03:54
pymagicits now in a black screen - nothing on there03:55
pymagicrebooting into kubuntu03:56
pymagicsarnold, the reconfigure is askin me for Linux Command Line04:01
sarnoldpymagic: whatever it's got is probably fine04:02
pymagicI think it had a legacy line or it was empty - it did not show me anything04:02
sarnoldyou pastebin shows empty04:02
pymagicleave it empty?04:02
sarnoldyeah04:03
pymagiclinux default command line: quiet splash -ok04:03
pymagicupdate nvram variables to automatically boot into debian? !04:04
pymagicyes or no04:04
pymagicsarnold, ?04:05
sarnoldpymagic: default sounds good, and yes, if you want to keep booting into ubuntu by default04:06
pymagicsarnold, its now asking grub efi partition04:06
pymagicand the only option it gives me is /dev/nvme0n1p1 - and not p2?04:06
pymagic536 Mb /boot/efi04:07
pymagicsay ok and select that - since i dont have a choice?04:07
sarnoldpymagic: yeah04:07
sarnoldpymagic: this is the EFI boot partiion, a smallish fat32 filesystem shared by all operating systems and uefi firmware on the system04:08
pymagicnow - reboot?04:08
sarnoldpymagic: if you're all done, yeah, give it a shot and lets see how it goes :)04:09
pymagicsarnold, back to the bios screen04:12
pymagicsarnold, :(04:12
sarnoldpymagic: oh no :( I had really hoped that dpkg-reconfigure was going to solve it all04:13
pymagicsarnold, in my bios it only shows ubuntu 4.0 - 2TB - so it does not show the 512MB partition to boot from04:13
pymagicisnt that an issue?04:13
sarnoldpymagic: that's probably fine04:13
pymagicis there a way to chroot and see dmesg perhaps? - I am not even sure if the bios even tried to boot grub - it just hops back into the bios04:15
sarnoldpymagic: probably not.. try the rescue media again, and try efibootmgr -v -- here's my output https://paste.ubuntu.com/p/7bCc8qqWHr/04:15
sarnoldpymagic: note that I've got windows listed here even though wiping windows was the first thing I did. the bootorder line describes which entries to try in which order.. and it takes the first 'working' one..04:16
pymagicsarnold, Trying04:17
pymagicsarnold, with or without chroot?04:19
sarnoldpymagic: shouldn't matter, this one reads from the firmware settings04:20
pymagicthanks - trying to boot into kubuntu04:20
pymagicsarnold, http://paste.ubuntu.com/p/2KxWpWY6Ft/04:24
pymagicsarnold, not sure how i got ubuntu/grub to work with legacy mode so far - but it did work. The only thing that broke the system was - in some grub setting i changed GRUB_TIMEOUT=2 from 0, and ran update-grub - and that broke the system. Not sure if that helps04:27
sarnoldpymagic: well, I'm pretty well confused :( the order suggests using 1, 0, 204:29
sarnoldpymagic: 1 is labeled "flash drive" but has both "cdrom" and "usb" in the string04:30
sarnoldpymagic: 0 looks like the thing you want to boot..04:30
pymagicwhen i try 0 it goes back to bios after trying to boot - no grub screen04:30
sarnoldpymagic: try: efibootmgr -o 0,1,2  ?04:31
pymagick it set that04:31
pymagic0 = ubuntu04:31
pymagicthe problem is "ubuntu" = 0 - when i set it up from bios - tries to boot - then comes back to bios04:31
pymagicI can reboot and try agian04:32
pymagicefibootmgr -o 0,1,2 worked fine04:32
sarnoldpymagic: alright, I'm running very out of ideas, and it's time to run :(04:32
pymagicI fired a reboot04:32
pymagicThanks for the help sarnold04:32
sarnoldpymagic: my last thought -- try getting back in to the rescue chroot, and from there, apt install --reinstall grub-efi-amd64 grub-efi-amd64-bin grub-efi-amd64-signed grub-common04:33
pymagick - will give it a try- thanks04:33
sarnoldhttps://paste.ubuntu.com/p/DhYyvj2gQ6/04:34
sarnoldturns out I've also got a grub2-common package installed. no idea why there's several.04:34
pymagicsarnold, 0 = ubuntu - did not boot - same problem04:34
sarnold:(04:34
pymagicubuntu 20.04lts?04:34
sarnoldpymagic: good luck, I hope you can sort this thing out :) I'm cautiously optimistic with the --reinstall approach.. I hope it's not too much trouble ;)04:36
pymagicthanks04:38
pymagicsarnold, got it :)04:46
pymagicThanks for all the help.04:46
=== Wryhder is now known as Lucas_Gray
=== pedro__ is now known as DodgeThis
kre10hey guys, how are you? Can you please help me with something. By my mistake, I added 2 eth0 address on my ubuntu 64 server(raspberry pi 4). And now I have 2 eth0 addresses which is very confusing.13:54
sdezielkre10: how did you add them?13:59
kre1073sdeziel hey there13:59
kre1073 sudo dhclient -r14:00
kre1073and sudo dhclient14:00
=== kre1073 is now known as kre10
sdezielkre10: a simple trick would probably be to reboot, if you can14:03
kre10yep, I rebooted a few times14:04
sdezielkre10: after rebooting, if you don't run those manual dhclient commands, do you still have the extraneous IPs?14:04
kre10yes14:04
=== kedar is now known as kedar_apte
sdezielcould you pastebin the output of 'ip a'14:05
kedar_aptekre10: xenial or bionic or focal?14:05
kre10focal14:05
kedar_aptecan you see anything in the netplan?14:05
kedar_aptein the folder /etc/netplan/14:06
kedar_aptethere will be a yaml file14:06
kre10just a second14:06
kedar_apteare u able to see two interfaces in the file?14:06
kre10eth0:14:06
kedar_apteonly one interface?14:06
kre10no, actually14:06
kre10yes14:07
kedar_aptecan you paste bin what u see?14:07
kre10Uploaded file: https://uploads.kiwiirc.com/files/e001cd861f42b0206e066e3601058174/pasted.txt14:07
kedar_aptewhat is shown when u type .......... ip r14:07
kedar_aptecan u screen shot that and imgur it14:08
kre10OK14:08
kre10https://www.screenpresso.com/=qv43f14:09
kedar_apteare you running bridged networks?14:10
kedar_aptelike for containers?14:10
kedar_aptelxc / lxd / docker?14:10
kre10yes, I'm using docker14:11
kedar_apteare u on the physicall server or ssh into the server?14:11
kedar_aptecan u exit and ssh again.... on the top you will see the disk usage and network details14:12
kedar_aptecan u screenshot that14:12
kre10ssh14:12
kre10ok14:12
kre10https://www.screenpresso.com/=NI5Zf14:13
kedar_aptehow many yaml files are you seeing in /etc/netplan/ ?14:15
kre10114:16
kre10-rw-r--r-- 1 root root 416 Apr  1 20:23 50-cloud-init.yaml14:16
kedar_aptehave you tried doing this? ....... ip addr flush eth014:17
kedar_aptealso try doing this: ip link set dev eth0 down14:18
kedar_apteu might loose connection with the server14:19
kedar_aptedo only if you have the ability to reboot or be on the server console14:19
kre10so after this command, I won't be able to login in the Pi?14:20
kedar_aptenot sure... wait a second.....just finding something usefl14:20
kre10ok14:21
kedar_aptetry this.... sudo ip addr del <ip address you want to remove> /24 dev eth014:22
kre10sudo ip addr del 192.168.0.109 /24 dev eth014:22
kre10?14:22
kedar_aptedo not delete the ip adress with which you have ssh'd in the server14:22
kre10OK14:22
kre10now ?14:23
kedar_aptehave u run the command?14:23
kedar_aptewhat was the output?14:23
kre10no output, still 2 IP's14:24
kre10should I reboot the system?14:24
kedar_apteI doubt that will help... but no harm trying14:24
kre10I doubt too :(14:25
kedar_apteno harm in trying14:26
kre10ok14:26
kedar_apteif it has restaryed.... can u run this and tell me what it is ....... ps aux | grep dhc14:29
kre10zzlatev@playstation:~$ ps aux | grep dhc14:32
kre10root         243  0.0  0.0      0     0 ?        I<   17:26   0:00 [sdhci]14:32
kre10root        1271  0.0  0.0   3172  2376 ?        Ss   17:26   0:00 /usr/sbin/dhcpcd14:32
kre10zzlatev     3898  0.0  0.0   7684   660 pts/0    S+   17:31   0:00 grep --color=auto dhc14:32
kre10sorry about that14:32
kre10kedar_apte still 2 IP's :(14:32
kedar_aptesystemctl disable dhcpcd.service14:33
kedar_aptesystemctl stop dhcpcd.service14:33
kedar_apteand reboot14:33
kre10I think that 2 users are using eth0? because I ran this command with the new user - zzlatev which I created14:33
kre10OK14:33
kedar_aptewait14:33
kedar_aptewait14:33
kre10OK14:33
kedar_apteboth of the ip addresses are dhcp?14:33
kre10I think so, yes14:34
kre10both are from the router14:34
kedar_apteae both the ips reachable?14:34
kre10but I have to say that I have pihole which IP is the second one 192.168.0.10914:34
kre10192.168.0.111 is registred as reserved dhcp client in the router14:35
kedar_aptealso is the mac ip same for both or it is shwoing different? you can check that mac ip in the router dhcp list14:35
kre10ok14:35
kre10there's isn't a client with 192.168.0.10914:36
kedar_aptehave you made any chage to /etc/network/interfaces?14:36
kre10nope14:38
kedar_apteare all your containers using the same NIC?14:38
kre10sorry, NIC?14:38
kedar_aptewhat is the default gateway for all containers?14:38
kedar_apteNIC - network card14:38
kre10in /etc/network I have 4 folders - if-down.d , if-post-down.d , if pre-up.d, if-up.d14:39
kedar_apteis there just one nic card?14:39
kre10I have one network card14:39
kedar_aptewhat is the default gateway for all all containers?14:39
kedar_aptewhat ip is being used as default gateway for all containers?>14:39
kedar_apte109 or 11114:39
kre1011114:40
kedar_apteand which IP is reserved on the router?14:41
kre10but I have access to the docker portainer with both IP's14:41
kre10111 is reserved on the router14:41
kedar_apteHonestly...I dont know... I have used up all my KB 🤔️14:42
kre10here's all networks in docker14:42
kre10https://www.screenpresso.com/=reNlc14:42
kedar_aptesomeone else could help14:42
kre10OK, thank you for trying! I appreciate it!14:42
kedar_aptesure14:42
kre10hm, ifconfig -a shows only one eth014:47
RoyKkre10: just learn to use the ip command and forget about ifconfig14:50
kre10Can you help me with it?14:51
kre10ip -s -s a f to 192.168.0.109/2414:51
kre10should I try this?14:51
=== tds8 is now known as tds
Slashmanhello, is there a way to skip the media integrity check for ubuntu server?16:56
Slashman(20.04.1)16:56
lotuspsychjeSlashman: press the key combo it suggests16:57
lotuspsychjectrl + C it was?16:57
Slashmandidn't see that on ubuntu server media16:57
lotuspsychjedidnt test myself on server, but i assume the integrity checks are on all ubuntu 20.04 iso's16:58
Slashmanlotuspsychje: https://ibb.co/5rX9mkD17:07
Slashmannothing I can do17:07
Slashmansuper fun when installing from idrac where any read is from the network and ultra slow on a media that was already verified17:07
lotuspsychjeSlashman: you need to early press the combo17:07
lotuspsychjeSlashman: there used to be a bug on that, but it should be resolved by now17:08
Slashmanlotuspsychje: I should do a video, but there is no time17:09
Slashmanno message, nothing17:09
SlashmanI select expert mode, then it immediately goes to check integrity17:10
lotuspsychjeUbuntu now defaults to checking the integrity of the medium in use when booting into live sessions. This can be skipped by hitting Ctrl-C, but due to a bug the message that tells you to hit this key is not shown in some flavours17:13
lotuspsychjehttps://bugs.launchpad.net/ubuntu/+source/casper/+bug/187001817:13
ubottuLaunchpad bug 1870018 in xubuntu-artwork (Ubuntu Focal) "Option (Ctrl-C) not shown to disable ISO verification" [Undecided,New]17:13
lotuspsychjeSlashman: wich iso did you use please?17:14
Slashmanlotuspsychje: http://www.releases.ubuntu.com/20.04/ubuntu-20.04.1-live-server-amd64.iso17:18
lotuspsychjeok tnx Slashman lemme poke some ppl17:18
Slashmanlotuspsychje: thanks17:19
lotuspsychjeSlashman: did you wait long enough till the test is skipped, it might take some time on older machines/media17:23
=== StathisA- is now known as StathisA
fretegihowdy guys21:19
fretegiquick question, have a running ubuntu server, bought a replacement.. both intel chipsets, dif gen tho.  will the new board in theory boot the old drive?21:20
fretegito the point of being able to SSH in, from there i can obviously do what i need to do21:20
fretegiserver is running 16.04 if that helps21:21
RoyKfretegi: should work unless the chipset is too new for the kernel to support it21:23
RoyKfretegi: usually that'll be ethernet or wifi drivers21:23
RoyKfretegi: you might want to try a usb live boot with the same distro version on the new motherboard first, just in case21:24
fretegiroyk, i was hoping for that!  should be good i would think, current server is skylake, new one is haswell, intel nics in both cases21:24
RoyKfretegi: or - well - just replace it and if the nics don't get online, make sure you have a console handy21:25
fretegiwell catch is.... i dont have a vga input monitor ha21:25
RoyKheh21:25
fretegiexactly lol21:25
fretegiso hoping this thing boots my current install as i can ssh in and do what i need21:25
fretegijust need a running environment21:25
RoyKwhat sort of boot disk or disks do you have on the old system?21:26
fretegiso i have 1 drive that the complete OS and then a raid array of data drives21:26
fretegidata already backed up etc.  so gonna build a new array on the new machine21:26
RoyKok, so perhaps put the new mobo on some non-conducting surface (a table cloth?) and connect the bootdisk and let it start up and check for results21:27
RoyKshould be a quick test21:27
fretegithats a thought, just never tried to boot one OS drive in another machine before21:28
RoyKit usually works fine21:29
fretegiUbuntu installs pretty much all kernel modules right? loading whats needed at boot?21:29
RoyKit's not like old windoze shite where everything is hardwired to whatever chipset you have21:29
shibbolethit has separate module packages, -modules and -modules-extra21:29
fretegiso unless u have hardware so new that proper support doesnt exist on  your installed kernel, should be good21:30
RoyKfretegi: the kernel will autoload whatever it finds needed according to the pci IDs. Those change sometimes with newer hardware, so it usually works, but not always21:30
shibboleththe latter typically don't get installed if you choose a virt/minimal baseline21:30
fretegiRoyK, fortunately for me, actually in a sense downgrading haha.  going to older, but yet more capable gear21:30
RoyKthen it will probably work just fine :)21:31
fretegishibboleth, honestly dont recall, pretty sure i did a standard ubuntu server install21:31
shibbolethdpkg -l | grep linux-modules21:32
fretegimay even be the same nic chips, checking now and for modules21:32
fretegishibboleth, yup exactly, remoting in now21:32
RoyKfretegi: really, if going to an older intel board, I don't see a problem here21:33
fretegishibboleth, yea got both linux-modules && linux-modules-extra should be solid!21:33
fretegiRoyK, yea me neither.  Perfect well thanks for confirming im not totally nuts guys21:33
fretegiwell at least with this anyhow21:33
shibbolethalso: later kernels (hwe) might enable support for more hw21:34
fretegishibboleth, yea good point21:38
fretegithanks guys, wish me luck, gonna fiddle with it this weekend if not sooner21:45
RoyKshibboleth: if it's older hardware, I doubt that'll be a problem21:59
kre10hey guys, how are you? Can you please help me with something. By my mistake, I added 2 eth0 address on my ubuntu 64 server(raspberry pi 4). And now I have 2 eth0 addresses which is very confusing.22:17
=== kre1028 is now known as kre10

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!