What's with the kernels that seem to stay stuck in core-testing longer than they used to?
Devs, I've noticed kernels stay stuck in core-testing longer than they used to. In the past (unless there was some minor version change), they usually stayed in testing a day or less. The past couple have lingered nearly a week and linux 6.8.7.arch1-2 x86_64 has been there a few days? That's great to ensure there are no issue with other packages that depend on building drivers from the kernel, but does raise the question for the curios -- "What's up?" I ask for scheduling purposes. I monitor the newsfeeds from the package repos and when I see a new kernel coming, I try and set aside a convenient time for the update/reboot. Past few kernels, I've done that only to find - they are still in testing. Not a complaint, just a curiosity. -- David C. Rankin, J.D.,P.E.
On April 26, 2024 8:16:45 PM EDT, "David C. Rankin" <drankinatty@gmail.com> wrote:
Devs,
I've noticed kernels stay stuck in core-testing longer than they used to. In the past (unless there was some minor version change), they usually stayed in testing a day or less. The past couple have lingered nearly a week and linux 6.8.7.arch1-2 x86_64 has been there a few days?
That's great to ensure there are no issue with other packages that depend on building drivers from the kernel, but does raise the question for the curios -- "What's up?"
I ask for scheduling purposes. I monitor the newsfeeds from the package repos and when I see a new kernel coming, I try and set aside a convenient time for the update/reboot. Past few kernels, I've done that only to find - they are still in testing.
Not a complaint, just a curiosity.
Haven't paid the closest attention, but I do try to update first thing every morning, and I've not noticed this. With LTS kernels, those I've seen linger a bit. ~ Null
On 4/26/24 20:25, Johnathan Null wrote:
I ask for scheduling purposes. I monitor the newsfeeds from the package repos and when I see a new kernel coming, I try and set aside a convenient time for the update/reboot. Past few kernels, I've done that only to find - they are still in testing.
Not a complaint, just a curiosity.
Haven't paid the closest attention, but I do try to update first thing every morning, and I've not noticed this. With LTS kernels, those I've seen linger a bit.
Thank you for the reply. I initially thought it was waiting on something like the Nvidia driver, but those move on through. Maybe systemd? Saw 255.5-1 hit core-testing on 4/24 with the linux 6.8.7.arch1-2 kernel (and the linux-lts 6.6.28-2 kernel), and now we have systemd 255.5-2 that hit earlier today. I too run both linux and linux-lts (in case there is a surprise on linux update), but hadn't really seen a distinction between the hold times on the two. I'll pay closer attention. Oh well, like I said, not complaining, just curious. I've not had an update packaging OOPS in years on core packages, so what the package process is doing is working really well - no need to hurry it. The kernel observation is only a pedantic observation - which inquiring minds just have to know why? :) -- David C. Rankin, J.D.,P.E.
Hello, On 27/04/2024 04:09, David C. Rankin wrote:
Thank you for the reply. I initially thought it was waiting on something like the Nvidia driver, but those move on through. Maybe systemd? Saw 255.5-1 hit core-testing on 4/24 with the linux 6.8.7.arch1-2 kernel (and the linux-lts 6.6.28-2 kernel), and now we have systemd 255.5-2 that hit earlier today.
Both of those you mention above are part of the python 3.12 rebuild due to rebuild of the toolchain. You can actually assist by being a tester and sign off packages in the testing repos if you want :) Cheers, -- Leonidas Spyropoulos PGP: 59E43E106B247368
On 4/27/24 00:08, Leonidas Spyropoulos wrote:
You can actually assist by being a tester and sign off packages in the testing repos if you want :)
Yes, I've followed the python progress, and I would be happy to help test -- but I don't use python :( I mean, I know what it is and how to write short scripts and have all the python dependencies that get pulled in, but I've been pretty much exclusive C/C++ for -- shit over 35 years now (and FORTRAN before that -- thankfully missed punch-cards by about 5 years - PCs didn't exist until I graduated college) So I don't really have any test code or set of programs that move through all the python "import" packages (that have proliferated like weeds in a vacant lot since 2.7) If you need testing of server packages, databases, (but not containers), I'm happy to help test. -- David C. Rankin, J.D.,P.E.
On April 27, 2024 5:42:31 AM EDT, "David C. Rankin" <drankinatty@gmail.com> wrote:
On 4/27/24 00:08, Leonidas Spyropoulos wrote:
You can actually assist by being a tester and sign off packages in the testing repos if you want :)
Yes, I've followed the python progress, and I would be happy to help test -- but I don't use python :(
I mean, I know what it is and how to write short scripts and have all the python dependencies that get pulled in, but I've been pretty much exclusive C/C++ for -- shit over 35 years now (and FORTRAN before that -- thankfully missed punch-cards by about 5 years - PCs didn't exist until I graduated college)
So I don't really have any test code or set of programs that move through all the python "import" packages (that have proliferated like weeds in a vacant lot since 2.7)
If you need testing of server packages, databases, (but not containers), I'm happy to help test.
Always good to have new testers. Hopefully you join. <https://wiki.archlinux.org/title/Arch_Testing_Team_> I'd say it's as simple and painless as it can get.
participants (3)
-
David C. Rankin
-
Johnathan Null
-
Leonidas Spyropoulos