On 10/3/18 11:13 AM, Fred Gleason wrote:
> On Oct 3, 2018, at 12:08, Alan Smith <[email protected]> wrote:
> 
>> To be completely honest, what scares me most is updates.  While rare I'm 
>> sure, a failed update on a production machine could be disastrous.  Yet at 
>> the same time if you don't update at all [don't fix something that isn't 
>> broke] yet station A has features that station B doesn't [because they are 
>> on different versions], then they complain.
> 
> What many groups or larger sites do is maintain an independent ‘sandbox’ 
> system which is reserved strictly for testing. Any updates are tested there 
> *first*, before being applied to production. This tends to prevent nasty 
> surprises.
> 

I concur with and encourage testing updates before applying them to
production systems.

Having said that, I've been working with Red Hat and CentOS systems for
more than 15 years as a system administrator and I have yet to
experience a "yum update" (or "upgrade", and "rpm -Uvh" before yum
existed) to cause a problem with a production system. Not that updates
have always worked flawlessly, but they've not interrupted running
applications. Sigh. Now that I've stated that publicly, I will be jinxed.

Test. Test. Test. And test again.

  ~David

Attachment: signature.asc
Description: OpenPGP digital signature

_______________________________________________
Rivendell-dev mailing list
[email protected]
http://caspian.paravelsystems.com/mailman/listinfo/rivendell-dev

Reply via email to