On Tue, 20 Apr 2010, Edward Ned Harvey wrote:

>> From: [email protected] [mailto:[email protected]] On Behalf
>> Of Matt Lawrence
>>
>>> What are the lists thoughts on using something like KSplice to keep
>> the
>>> kernel up to date?
>>
>> If I have systems that really need to be up all the time, then it
>> sounds
>> like an excellent product.  If I cn afford scheduled downtime, I would
>> prefer to stick with the more traditional method of making updates.
>
> Better yet, some form of high availability.  The ability to have multiple
> machines all providing the same services.  Take one down for service, and
> when it's back up again, take the other down for service.

That's not exactly traditional high availability, that more of the current 
horizontal scaling that so many apps are being built around today.  It's a 
wonderful way to do things, just take systems out of the pool, update them 
and put them back in.

In traditional HA, the servers have to be aware of each other and one 
takes over functionality a short time after the other has failed.  Given 
the possible failure modes, I recommend against trying to implement it 
unless it is really necessary.  When it screws up, everything goes down 
and it's a bigger job to figure out how to fix it with the extra HA stuff 
added on.  Can you tell that I recently had an unpleasent experience?

> Of course you can't do that for *every* thing, but whenever possible, that's
> the preference.

I so agree.

-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
_______________________________________________
Tech mailing list
[email protected]
http://lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to