Christopher et al.,

Might it make sense to have some middle ground (maybe super-sparse) image and then have blessed "recipes" by some means (not implying Chef) and after install get them to their pre-install state?

I imagine one of the appeals of the images in question is that (theoretically) it reduces the barrier to use of SmartOS.

To some an ideal case might be when a VM appliance consumer never even knew that it was running in an SmartOS zone instead knowing only that there was a hosted "InsertNameHere" Appliance somewhere.

In another context but very germane I had a bit of trouble installing some software on older RHEL install due to some dependency issues (and some questionable documentation) and ended up just using a `Bit -nami' pkg/image on a some other vendor's cloud VM. I was impressed with what they had done, but I was very impressed with what I didn't have to do, or rather how little I had to do. It installed all it's own stuff, e.g. MySQL, so it was relatively independent from the OS (maybe I could get it to work in a LX Zone?)

The short shrift is: folks are usually going to use the tools they think get them the most value and even though SmartOS is a superior OS/VM offering in every way (IMHO), it may not still be the best business decision.
(Isn't CoreOS just a Linux homage to Sol/Ill/SmartOS)?

So, maybe the lack of usage of the images may not be a sign of their lack of a place in the market but of another short-coming.

From my experience managers and other deciders just want tools to make their problems go away, in a way that they aren't going to come back in a worse way later.

I'd expect that if Joyent were to provide recipes in way that someone could through a form say: I want Webserver type-A to use DB type-B which push logging into Elastisearch; all which feed DashBoard type-D; and, all the gritty details be hidden from them, I am pretty sure you'll find a lot of love in the market.

Thank You,
Will



On 3/3/17 12:51, Christopher Horrell wrote:
It's mostly driven by image usage and the amount of effort required to maintain a given image. We're currently going through an EOL exercise this year to pick and choose which images will stay and which will go. Generally speaking most of the database related images (like percona, postgresql, mongodb etc) will stay. The WordPress image is something we might drop in the future.

For image updates we generally try to update the SmartOS (non-base) images once a year for the pkgsrc Q4 LTS release, with occasional interim updates where warranted.

We recently release a percona 16.4.1 image. If you had some issue with that image, please let me know off-list and we'll take a look.

On Fri, Mar 3, 2017 at 1:34 PM Matthew Parsons <[email protected] <mailto:[email protected]>> wrote:

    What is the driving factor(s) for if an image line is kept up or
    abandoned? The couple time I've tried anything besides a basic
    image, it either outright was broken (percona) or been woefully
    outdated (wordpress).

    I expect it's mostly a manpower vs. interest/usage issue, just
    would like to know what is worth bumping/filing requests/bugs for,
    and what to ignore. Thanks.

--
Christopher Horrell
Joyent Inc.
http://www.joyent.com/
*smartos-discuss* | Archives <https://www.listbox.com/member/archive/184463/=now> <https://www.listbox.com/member/archive/rss/184463/26368559-ba231e1c> | Modify <https://www.listbox.com/member/?&;> Your Subscription [Powered by Listbox] <http://www.listbox.com>





-------------------------------------------
smartos-discuss
Archives: https://www.listbox.com/member/archive/184463/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb
Powered by Listbox: http://www.listbox.com

Reply via email to