Il 2021-04-30 16:26 Johnny Hughes ha scritto:
On 4/30/21 4:32 AM, Gionatan Danti wrote:
Il 2021-04-30 06:55 Gordon Messmer ha scritto:
Why do you think that? Are RHEL (and CentOS) point releases backward compatible or not? If you trust point releases to work, why would you hesitate to trust a distribution that resembles an upcoming point release?
Because it very often break kABI compatibility, with 3rd party module heavily affected.
Don't get me wrong: I understand that Stream is the way forward and that things are not going to change, and this is fine. But trying to ignore the key differences (shorter support, unknown upgrade from Stream-8 to Stream-9, broken kABI, etc) is not useful to anyone.
Stream is a *different* product, because is avoid (for the good or the bad) basically *all* things that make RHEL so special. And lets face it: kABI and long/quality support from RedHat are the only two things which make RHEL special. Stripping them from CentOS will produce a very different product. And, as a side note, things break more often on Stream-8 then CentOS8. Maybe Stream only needs to mature, but it still a different product.
My personal opinion is that RH created Stream to give cloud vendors a place to experiment/repackage *before* adding that to the main RHEL distro. Stream really does not seem targeted aSo, t small sites / "normal" sysadmins, rather at large cloud vendors.
Which, again, is perfectly fine unless trying to disguise it as an "almost-RHEL" distro. Regards.
Sure .. so block kernels and build your own in that situation. Or use something else. There are always edge cases. There are millions of CentOS users. What percentage use 3rd party modules (other than nvidia drivers). There are some, and this would be a problem for those people.
So, IF another downstream distro works for you .. use it. Or use Debian or Ubuntu, or BSD. Use Alma or Rocky Linux. Buy RHEL.
As stated above, if this is the vision of Stream, fine. I am not arguing about the vision: while I don't like it, my opinion is irrelevant.
But disguising Stream as "almost-RH" (a mantra repeated many times both here and in various blog) is plain wrong, and I genuinely don't think it will be good for Stream.
And you know better than me that what you wrote above regarding the kernel is a double-edge sword: as you cope with security patches if the kernel is blocked? How do you cope with HP/DELL/Lenovo kmod needed to configure the RAID subsystem if using a rolling kernel? Did you notice that even RH-sponsored modules as kvdo were broken multiple times on Stream? If you are using VDO to access your storage and it suddenly is not usable anymore, how would you feel? What about ZFS on Linux users? Do you realize this drastically reduces Stream fitness to bare-metal install (one of the main CentOS usage was as hypervisor)?
The correct answer is to buy RH: fine. But do not let Stream touch anything which require a kABI compatible modules. As said above, the Stream move is squarely addresses *cloud* vendor requests and needs. Again, fine. But please leave apart the RH comparison, this is not going to help Stream.
Again, don't let me wrong: I wishes the best to Stream, and I will use it where appropriate. But "where" is much smaller today than yesterday. But this aside, I really thank you all CentOS maintainer for your monumental work, and I really hope Stream will be a success.
Regards.