DevOps pilgrim Progress forks out $220m for automation crew Chef

Boston-based development veteran Progress – beloved by greybeards for its Progress 4GL – has just spent nearly a quarter of a billion greenbacks on DevOps darling Chef. The deal, worth $220m in cash, is schedued to close in October and the transaction "is expected to be accretive beginning in Q1 FY2021", according to Progress …

[ovation1357]

Does anybody actually still use Chef?

Everyone I know in this sphere is using Puppet or Ansible.

Last time I looked at Chef it seemed to be quite neglected in terms of up to date 3rd party modules (not that puppet is a shining example either)..

I can't criticise Chef itself as I've never used it but neither, it seems, has anyone else I know. So does it still have a niche somewhere? What makes it stand out?

I'm getting a bit blasé about the main contenders for automated configuration management. It's not that I don't appreciate their plus points but it starts to feel to me like they're all minor variations on the same theme, which is some form of horrible YAML and/or a DSL to come with a declarative state for a system, except that declarative just doesn't handle everything so you've then got all kinds of different kludges to handle procedural elements and conditional cases.

I've spent so many hours fighting Puppet to make it do things which should be simple but aren't (here's looking at you: all-powerful yet rather evil augeas!) I keep looking at the competition but nothing has convinced me that it's any better.

[Anonymous Coward]

oh bother...

Welcome to hell.

[Anonymous Coward]

I don't use Chef, nor am I familiar with the segment it's in. But I'm wondering why a software company with a $70m ARR would be acquired for only $220m?

That's a horrible multiplier. Were they running out of money? Not growing fast enough? Investors tired of waiting and just wanted an exit and return of any kind so decided to take the first offer that came along?

Anybody know the inside story or have some good theories why they were acquired for so cheap compared to their ARR?

[Rainer]

Not really surprised

The problem with Chef was and is (AFAIK) that integration into 3rd-party things was always quite complicated.

This is maybe rooted in the idea that chef would be *the* source of truth of an enterprise and not some cog-in-the-wheel.

You can see this when you look how e.g. The Foreman tries to integrate chef in comparison to ansible.

I have done some chef in the past (now everything is ansible) and the learning-curve was comparatively steep.

So, chef's "mind-share" was certainly shrinking.

If you do some google-searches, you also get to postings on reddit where people claim that large chef-environments claim that they became unmanageable over time...

Chef's hosted chef-server (which we never used) was also often criticized for having less-than-stellar availability-figures.

Some aspects of chef I really liked, e.g. the fact that it ran continuously and thus there was little doubt about the state of the node.