Oracle Makes Red Hat Kernel Changes Available As Broken-Out Patches
Artefacto writes "The Ksplice team has made available a git repository with the changes Red Hat made to the kernel broken down. They are calling this project RedPatch. This comes in response to a policy change Red Hat had implemented in early 2011, with the goal of undercutting Oracle and other vendors' strategy of poaching Red Hat's customers. The Ksplice team says they've been working on these individual patches since then. They claim to be now making it public because they 'feel everyone in the Linux community can benefit from the work.' 'For Ksplice, we build individual updates for each change and rely on source patches that are broken-out, not a giant tarball. Otherwise, we wouldn't be able to take the right patches to create individual updates for each fix, and to skip over the noise — like a change that speeds up bootup — which is unnecessary for an already-running system.'"
Yes, and I'm sure Oracle-owned K-Splice has NO alterior motive for doing this, esp considering the RH change was purportedly made in response to oracles so-called 'unbreakable linux' (Aka oracles for-$ RHEL builds)
If you want a real enterprise class O.S. ditch RHEL and go with Solaris 11.
First Post
Red Hat wouldn't need to start obfuscating their patches in the first place. You'd think with all the billions of dollars Oracle and its consultants mooches off of companies that they would at least be able to develop their own Linux distribution instead of relying on something else.
release patches that upgrades Oracle 9 to 11.
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
Better cover that RedPatch with an iPad
Would be nice if Oracle would break out their MySQL patches.
No, but it's damn good evidence that it's not being run inefficiently.
You're looking at *one* specific release. What if Oracle only once sent in code and it made it into 2.6.33? You need a larger dataset in order to come up with anything significant.
What sort of code was committed? If it were some hardware drivers for SUN hardware they made themselves, it's not that much benefit to other companies, only to a few end users that buy very expensive SUN hardware to run Linux on it, that will run just as well on "generic" hardware that's in a lower price class. I'm not saying that's what happened, but you need to factor this in before you come up with any conclusions.
The RedHat patches that are released as a big bunch, are the patches they backported to the "old" kernel they base their Enterprise distribution on. These are not *new* patches that are sent upstream to be merged in new kernels, to fix unfixed bugs or support new hardware or features. RedHat backports security, stability and in some cases new hardware to the old kernel. These are merely existing patches that are being applied to an old kernel. Only maintainers/users of clones of RedHats' Enterprise Linux benefit from this. Anyone that wants to use a new kernel has no use for these, since they are already in the new kernel by default.
RedHat also contributes to a lot of the *new* features and drivers in the kernel. They don't make any hardware themselves, but they fix other vendors hardware drivers if they are buggy. They are large contributors in several filesystems, SElinux and many other parts of the kernel.
No, I don't work for RedHat, nor hold any of their certifications. I in fact do have certifications for Oracle Solaris products. I think both products by themselves are pretty good, but I despise the business practices or Oracle and the way they continuously rip their employees, the open source community and their customers another one at any opportunity they can find or create.
I was promised a flying car. Where is my flying car?