Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Oracle Buys Ksplice

oracle.com

Post reply on HN
Print view View original

31–38 of 38 posts

Re: Oracle Buys Ksplice

#31

Post by jamespo »

nodata wrote: post #27
koenigdavidmj wrote: post #15

Earlier quoted context omitted.

> A few minutes planned downtime now in much better than hours of unplanned downtime at some indeterminate future time. This reminds me of something that OpenSolaris did that I really liked. When you updated packages, it would take a ZFS snapshot. It only took up space for the changed bits, and happened almost instantly. If things stopped working after the reboot, then you could select a GRUB menu option to boot into…

So how does it work in practice? 1. You upgrade postgres, and the new on-disk format isn't compatible with the old one. 2. You upgrade mysql, and the new on-disk format IS compatible with the old one, but the new version has a bug that you keep hitting. You want to downgrade without losing all the transactions that you've accrued since the upgrade.

You use backups and do proper testing in a staging environment

Re: Oracle Buys Ksplice

#33

Post by nodata »

jamespo wrote: post #31
nodata wrote: post #27

Earlier quoted context omitted.

So how does it work in practice? 1. You upgrade postgres, and the new on-disk format isn't compatible with the old one. 2. You upgrade mysql, and the new on-disk format IS compatible with the old one, but the new version has a bug that you keep hitting. You want to downgrade without losing all the transactions that you've accrued since the upgrade.

You use backups and do proper testing in a staging environment

Which kind of negates the benefits of this feature.

Re: Oracle Buys Ksplice

#34

Post by rwmj »

soult wrote: post #5
rwmj wrote: post #3

If your servers can't ever be rebooted, I think you have problems.

From what I have heard, ksplice is pretty popular with VPS (virtual private server) providers. When there is an exploitable bug, they have to announce the planned upgrade to their customers and wait for a few days until they can run the upgrade. Even worse, they basically announce that there is a hole in their system to the very people that already have elevated access to that system. (Even non-remotly-exploitable bu…

They should be able to migrate the VMs off to another box, allowing the first one to be upgraded and rebooted. The additional benefit is that this tests a migration path that ought to exist in case of hardware failures.

Re: Oracle Buys Ksplice

#36

Post by niczar »

rwmj wrote: post #3

If your servers can't ever be rebooted, I think you have problems.

There are situations where rebooting always has an impact, even if you're in a cluster. For instance take a database cluster in active/passive configuration. Switching over may be fast, but it will always take some time. Now you could design the app to be able to handle this, but this increases the complexity tremendously, esp. when dealing with transactions (in the DB sense of the word). Using ksplice removes this loss of service, however small, in most cases.

Re: Oracle Buys Ksplice

#37

Post by seanp2k »

6ren wrote: post #14
VanL wrote: post #11

Given Oracle's recent activities, I would guess that Ksplice's patent applications are the real prize. There are two that are published: The first is "Method of finding a safe time to modify code of a running computer program": http://bit.ly/ksplice-1 The second is "Method of determining which computer program functions are changed by an arbitrary source code modification": http://bit.ly/ksplice-2 These are just appl…

> Given Oracle's recent activities What do you mean? Sure, Oracle is an aggressive company, but they are clearly doing business, and not a non-practicing patent troll. Even in suing Google over Java in Android, they not only use Java but are the center of its development. Sun itself sued [Microsoft] over Java. This acquisition makes sense to them, since linux is a key part of their vertical offering. If no one else c…

>"prior art."

#$%^@#*&$%&@#^$ I hate that term when referring to code. Code is not art. Code is science. In a given language, there might only be one way that makes sense to do something. Software patents are awful.

Re: Oracle Buys Ksplice

#38

Post by seanp2k »

6ren wrote: post #29
VanL wrote: post #16

Earlier quoted context omitted.

Oracle is extremely aggressive about IP enforcement - and just because you haven't heard about it in court, doesn't mean that it isn't happening. Ninety-nine percent of these issues are settled quietly. I have been on the other side of licensing discussions twice this year alone. It is not just patents - they are actively trying to monetize their IP, including changing license terms (see the evolution of the license…

Thanks. "Recent activities" sounded like you were referring to something public (e.g. there'd been recent stories on HN about it) - but now it sounds private. We may differ in the following, but I think it's reasonable to enforce your IP if you created it and use it. As an uISV, it's important to me that customers pay for it (although I'd find it unpleasant to pursue infringing customers). Changing the license terms…

>"I think they must be getting value for money."

That, or they're just locked into an Oracle product that Oracle can change the price on whenever it wants. Ever try migrating Oracle Database anything else? They have most customers over a barrel.

Post reply on HN