| [09:44] <Khepra> Hello. Debian recently had a regular kernel security upgrade which was actually a rebase to 5.10.120 which includes the massive changes in the random subsystem. I don't see this change in current ubuntu LTS kernels (regular or HWE). How can I find out whether this change is planned? Other than auditing changelogs before upgrade? | |
| [11:29] <tjaalton> Khepra: it is in 5.15.44, which jammy will eventually get | |
| [13:59] <Khepra> tjaalton: k, thanks. btw, what's ubuntu kernel's team's opinion on such a radical change mid-LTS? | |
| [13:59] <Khepra> (LTS kernel I mean) | |
| [14:04] <tjaalton> I doubt we'll revert upstream decisions | |
| [14:12] <Khepra> then that makes RHEL the only enterprise friendly distro, unfortunately. | |
| [14:24] <tjaalton> because? | |
| [16:49] <xnox> rhel applies humoungous amount of changes to their kernels during its life, well above and beyond what is even acceptable for regular linux-stable trees =/ | |
| [20:13] <Sylwah> Hello, quick question. I believe I am impacted by a kernel bug: encrypted zfs read/write on 5.15.0-37 is about 10x slower than on 5.15.0-27. I can 100% repro across several machines, with the only change being booting a different kernel. This includes writing to a mirror zfs pool backed by a ramdisk (in tmpfs). Is this the right forum to discuss, | |
| [20:13] <Sylwah> or Launchpad? | |
| [20:14] <sarnold> probably launchpad, irc doesn't get nearly the use it once did | |
| [20:15] <Sylwah> I will, thank you. Finally, should I report to zfs, or linux-generic? | |
| [20:17] <sarnold> I think linux is probably better, but I'm not positive on that one | |
| [20:17] <Sylwah> I'll figure it out. Thank you, have a great day/evening/night :) | |
| [20:18] <sarnold> you too :) thanks! | |