lol wut?
I’d expect if command1 or command2 fails, command3 is attempted. Am I reading that incorrectly?
lol wut?
I’d expect if command1 or command2 fails, command3 is attempted. Am I reading that incorrectly?
It’s behaving like:
command1 | command3 || command2 | command3
I expected that if command1 succeeded but command3 failed, it would just fail.
Which shell?
But that actually makes sense, due to shell expansion. Kinda.
zsh. didn’t test it in bash
Not sure that bash would have different results.
It’s a zsh thing:
bash:
$ set -o pipefail
$ { ls || fakenews; } | fakernews
bash: fakernews: command not found...
zsh:
% set -o pipefail
% { ls || fakenews; } | fakernews
zsh: command not found: fakernews
zsh: command not found: fakenews
Bash would try to pipe the output from the curly bracket stuff into fakernews
Right
Yeah, it’s the zsh behavior that’s surprising.
New Mac M1 stuff is really interesting. That along with the GPUs this year and 5000-series Ryzen, I can’t remember a better year for PC tech… Thanks, 2020. Suck it, Intel.
2020 can still fuck itself
Gave the website some love after a long period of neglect. Still have to migrate some knowledge base stuff over, but the format is much better… the top left logo is a placeholder, although it’s kind of interesting to use all unicode characters instead of a graphic.
I’m converting a 32GB zvol to a vmdk and it’s going to take 8-12 hours at the rate its going. Is that normal?
Same zpool?
I haven’t done it. But that doesn’t sound right to me. I usually sustain 120 MBps sustained write on my SMR drives. I think it is a 3x disk zpool.
No, it’s going onto an exfat USB drive… maybe I should have mentioned that.
I don’t think it’s I/O constrained, the conversion process is the bottleneck.
Gave up on Artix. Working on Arch install for my Linux desktop. It is a journey of discovery…
https://wiki.archlinux.org/index.php/EFI_system_partition#ESP_on_software_RAID1
It is possible to make the ESP part of a RAID1 array, but doing so brings the risk of data corruption, and further considerations need to be taken when creating the ESP
Hmm, ok, cool let’s try it out.
https://outflux.net/blog/archives/2018/04/19/uefi-booting-and-raid1/
Alright, kind of hacky, but seems like it’ll work.
https://nwildner.com/posts/2020-07-04-secure-your-boot-process/
You can’t use grub here cause it does not support luks2 volumes yet (maybe 2.06 will do).
No problem, since I’m installing gnu+linux÷systemd
, I wanted to try systemd-boot
anyway. IIRC, grub was kind of shimmed into using EFI anyway and since I will soon be a belligerent Arch user, I am allergic to the idea of using v1 of something when v2 exists.
[root@archiso /]# bootctl --esp-path=/efi install
Failed to probe partition scheme of "/dev/block/9:127": Input/output error
Uh oh…
it’s certainly nothing we should officially support because it cannot work in the general case. Sorry.
Poettering strikes again. Time for some stupid rsync shenanigans.
Ok we get it you use arch XD
I don’t even yet, unless the live ISO counts… making good progress though.
NAME FSTYPE FSVER LABEL UUID FSAVAIL FSUSE% MOUNTPOINT
loop0 squashfs 4.0
sr0 iso9660 Joliet Extension ARCH_202011 2020-11-01-06-38-10-00 0 100% /run/archiso/bootmnt
vda
├─vda1 linux_raid_member 1.0 esp_mirror 5ee49e75-7f37-ec2e-b469-da751fffda05
│ └─md127 vfat FAT32 ESP A65A-9809 548.8M 0% /mnt/efi
├─vda2 linux_raid_member 1.0 boot_mirror 00065d3d-4f58-9ad5-caa0-a69a3b0c3e74
│ └─md126 crypto_LUKS 2 f3a042c0-b5bb-4dc5-b756-692555175ea9
│ └─boot_luks ext4 1.0 boot 30413d69-d103-472c-85f5-8769c3fa306f 3.6G 2% /mnt/boot
└─vda3 linux_raid_member 1.2 root_mirror bc0d7efe-beab-6931-3a06-351295c4e587
└─md125 crypto_LUKS 2 34c66b0e-0ec5-46e1-adb9-a3514575dce7
└─root_luks LVM2_member LVM2 001 buzMeQ-elxd-4bv3-lTeL-Ik2p-SWel-etxBHR
├─root_vg-root_lv ext4 1.0 root 293b5364-88ae-473f-b260-508314898528 42.8G 4% /mnt
├─root_vg-var_lv ext4 1.0 var 9318b045-ebda-409a-9607-b19e5b1766a6 29.2G 2% /mnt/var
├─root_vg-var_log_lv ext4 1.0 var_log 768d71f9-5d43-4511-8ff4-185b1025ff3f 7.4G 0% /mnt/var/log
├─root_vg-var_log_audit_lv ext4 1.0 var_log_audit ff91f1ce-13bc-40ed-8e8e-c9a079dcd8d1 7.4G 0% /mnt/var/log/audit
└─root_vg-swap_lv swap 1 de85bf6a-c4fd-4375-a850-76a320b3d229 [SWAP]
vdb
├─vdb1 linux_raid_member 1.0 esp_mirror 5ee49e75-7f37-ec2e-b469-da751fffda05
│ └─md127 vfat FAT32 ESP A65A-9809 548.8M 0% /mnt/efi
├─vdb2 linux_raid_member 1.0 boot_mirror 00065d3d-4f58-9ad5-caa0-a69a3b0c3e74
│ └─md126 crypto_LUKS 2 f3a042c0-b5bb-4dc5-b756-692555175ea9
│ └─boot_luks ext4 1.0 boot 30413d69-d103-472c-85f5-8769c3fa306f 3.6G 2% /mnt/boot
└─vdb3 linux_raid_member 1.2 root_mirror bc0d7efe-beab-6931-3a06-351295c4e587
└─md125 crypto_LUKS 2 34c66b0e-0ec5-46e1-adb9-a3514575dce7
└─root_luks LVM2_member LVM2 001 buzMeQ-elxd-4bv3-lTeL-Ik2p-SWel-etxBHR
├─root_vg-root_lv ext4 1.0 root 293b5364-88ae-473f-b260-508314898528 42.8G 4% /mnt
├─root_vg-var_lv ext4 1.0 var 9318b045-ebda-409a-9607-b19e5b1766a6 29.2G 2% /mnt/var
├─root_vg-var_log_lv ext4 1.0 var_log 768d71f9-5d43-4511-8ff4-185b1025ff3f 7.4G 0% /mnt/var/log
├─root_vg-var_log_audit_lv ext4 1.0 var_log_audit ff91f1ce-13bc-40ed-8e8e-c9a079dcd8d1 7.4G 0% /mnt/var/log/audit
└─root_vg-swap_lv swap 1 de85bf6a-c4fd-4375-a850-76a320b3d229 [SWAP]
vdc crypto_LUKS 2 7b187f20-c4f4-4161-954c-e8c01f6866d4
└─home_luks LVM2_member LVM2 001 SzlCcZ-zTEX-4VR4-iakF-rMns-c2Nf-MJmexS
└─home_vg-home_lv ext4 1.0 home a951f588-e73e-4cad-b678-35a8f1ac53ad 79.5G 0% /mnt/home
Selfhost for pride, or use a service for ease?
Squarespace. The self-hosting isn’t the issue so much as the building/designing. That said, I could see myself migrating to Ghost as some point. Looks like they’ve added a lot of features since I initially checked it out years ago.
Ghost was nice but I can’t stand the ability to not tweak stuff, especially if I plan to do documentation.