Recent Features
| |||||||||||||
Stability of release kernelsStability of release kernelsPosted Feb 3, 2005 19:51 UTC (Thu) by shane (subscriber, #3335)In reply to: NETIF_F_LLTX and race conditions by jwb Parent article: NETIF_F_LLTX and race conditions
If you need to run the absolute latest kernel, and care about stability, then you need to set up a test environment and test each release before putting it into production. This is true no matter what pre-release testing procedure is built into the kernel release cycle - chances are no kernel developer has your production environment as a desktop machine!
If you're lazy, you wait for a kernel to age, like a fine wine.
Or you can do what most people do, and use the kernel that comes with your distribution.
(Log in to post comments)
Stability of release kernels Posted Feb 4, 2005 0:54 UTC (Fri) by giraffedata (subscriber, #1954) [Link] If you're lazy, you wait for a kernel to age, like a fine wine. Aging doesn't make the bugs go away. And bugs are always there. The kind of aging you're talking about happens to a series of kernels, not a particular one, and is more precisely called "stabilizing." That doesn't happen any more in kernel.org kernel series, but does in some Linux distribution kernel series.
Stability of release kernels Posted Feb 5, 2005 2:45 UTC (Sat) by set (guest, #4788) [Link] Another choice would be either Alan Cox' ac kernel series,
or Andreas Salomon's series:
(from his first 2.6.10-as1 announcement:)
|
|||||||||||||