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

Decommissioning Otto

hashicorp.com

Post reply on HN
Print view View original

61–70 of 102 posts

Re: Decommissioning Otto

#61

Post by gameofdrones »

erichmond wrote: post #38
Mizza wrote: post #8

Does anybody _actually_ use the HashiCorp stack, besides Vagrant, for serious work? I tried and honestly found their products sorely, sorely lacking. Very shiny documentation, very incomplete, un-battled-tested tools, no examples given, little response from their devs other than the PR team. For a small example, I _still_ get +1 notifications on this critical issue nearly every day: https://github.com/mitchellh/packe…

Packer, Vagrant, Terraform, and we are working internally with Vault and Consul. Best devops tools (along with ansible) I've ever worked with.

What other deveops tools have you used/want to use?

Re: Decommissioning Otto

#62

Post by gaius »

gameofdrones wrote: post #60
bsuh wrote: post #28

Earlier quoted context omitted.

Heck I found the same to be true of vagrant. Big promises, short on delivery. While mostly usable, it's definitely not as seamless across various guests, providers, and provisioners as their docs might have you think. Stray off the beaten path even the tiniest bit and expect to spend hours looking at github issues for workarounds (submitted by other users not the dev team). My experience. Windows guests: buggy FreeBS…

Abstractions-on-top-of-abstractions which promise to reduce complexity by adding more levels of indirection, when the UX of the underlying tools need to be better. (Having wrote a multihypervisor disk-cloner in Ruby.)

Ah well, that's because people didn't pay enough attention to Wheeler.

Re: Decommissioning Otto

#63

Post by aleksi »

bsuh wrote: post #28
Mizza wrote: post #8

Does anybody _actually_ use the HashiCorp stack, besides Vagrant, for serious work? I tried and honestly found their products sorely, sorely lacking. Very shiny documentation, very incomplete, un-battled-tested tools, no examples given, little response from their devs other than the PR team. For a small example, I _still_ get +1 notifications on this critical issue nearly every day: https://github.com/mitchellh/packe…

Heck I found the same to be true of vagrant. Big promises, short on delivery. While mostly usable, it's definitely not as seamless across various guests, providers, and provisioners as their docs might have you think. Stray off the beaten path even the tiniest bit and expect to spend hours looking at github issues for workarounds (submitted by other users not the dev team). My experience. Windows guests: buggy FreeBS…

Never used Windows or FreeBSD guests, but what's wrong with Ansible provisioning?

Re: Decommissioning Otto

#64

Post by csomar »

Mizza wrote: post #8

Does anybody _actually_ use the HashiCorp stack, besides Vagrant, for serious work? I tried and honestly found their products sorely, sorely lacking. Very shiny documentation, very incomplete, un-battled-tested tools, no examples given, little response from their devs other than the PR team. For a small example, I _still_ get +1 notifications on this critical issue nearly every day: https://github.com/mitchellh/packe…

I'm having the same miserable experience with Terraform. Lacking documentation, and really un-batted-tested product.

The concept is nice, and the concept of execution is nice too. The execution itself is a failure.

But, hey, they know how to make shiny websites.

Re: Decommissioning Otto

#65

Post by eropple »

Mizza wrote: post #8

Does anybody _actually_ use the HashiCorp stack, besides Vagrant, for serious work? I tried and honestly found their products sorely, sorely lacking. Very shiny documentation, very incomplete, un-battled-tested tools, no examples given, little response from their devs other than the PR team. For a small example, I _still_ get +1 notifications on this critical issue nearly every day: https://github.com/mitchellh/packe…

Speaking for myself, I've put enough tooling around Packer to make it tolerable in my build pipeline and I haven't found a better tool that does what it does (suggestions welcome!). I would be using Vault, except it's just on the far edge of usable for me (particularly around the aws-ec2 provider) and I am repelled sufficiently by Go that I'm not driven to hack it to suit my needs until I exhaust my other alternatives, and I haven't yet.

Consul is...fine, I guess? It's not particularly useful for me, but that's because I have a Zookeeper background and am comfortable with its primitives. I'd use it if I had a need for it. OTOH, I've had extremely bad experiences with Terraform, both in terms of trying to build tooling around it and in having it hose states when resources fail to create correctly (I've mentioned previously around here that multiple clients independently invented the term "terrafucked" for the results of their Terraform state) and I don't feel like I can trust the tool with something as critical as my cloud infrastructural resources. I might bite the bullet and use it if I were working on Google Cloud or Azure, but in AWS, CloudFormation (with tooling on top) is sufficient.

Re: Decommissioning Otto

#66

Post by 0xmohit »

Mizza wrote: post #8

Does anybody _actually_ use the HashiCorp stack, besides Vagrant, for serious work? I tried and honestly found their products sorely, sorely lacking. Very shiny documentation, very incomplete, un-battled-tested tools, no examples given, little response from their devs other than the PR team. For a small example, I _still_ get +1 notifications on this critical issue nearly every day: https://github.com/mitchellh/packe…

It is unfortunate that all people can do is to complain.

One should probably think that about sending a PR for a feature request or a bug report if it really impacts them. An attempt, even if in the wrong direction, would probably push the priority of the underlying issue higher for the core devs working on the project.

If you can't do that, at least attempt to be constructive.

Re: Decommissioning Otto

#67

Post by axelfontaine »

eropple wrote: post #65
Mizza wrote: post #8

Does anybody _actually_ use the HashiCorp stack, besides Vagrant, for serious work? I tried and honestly found their products sorely, sorely lacking. Very shiny documentation, very incomplete, un-battled-tested tools, no examples given, little response from their devs other than the PR team. For a small example, I _still_ get +1 notifications on this critical issue nearly every day: https://github.com/mitchellh/packe…

Speaking for myself, I've put enough tooling around Packer to make it tolerable in my build pipeline and I haven't found a better tool that does what it does (suggestions welcome!). I would be using Vault, except it's just on the far edge of usable for me (particularly around the aws-ec2 provider) and I am repelled sufficiently by Go that I'm not driven to hack it to suit my needs until I exhaust my other alternative…

If what you are after is generating AMIs for JVM, Node.js or Go apps, we at Boxfuse offer a great alternative to Packer with very fast creation of very small and secure AMIs https://boxfuse.com/blog/amis-in-30-seconds

Disclaimer: I am the founder and CEO of Boxfuse

Re: Decommissioning Otto

#68

Post by unixhero »

aleksi wrote: post #63
bsuh wrote: post #28

Earlier quoted context omitted.

Heck I found the same to be true of vagrant. Big promises, short on delivery. While mostly usable, it's definitely not as seamless across various guests, providers, and provisioners as their docs might have you think. Stray off the beaten path even the tiniest bit and expect to spend hours looking at github issues for workarounds (submitted by other users not the dev team). My experience. Windows guests: buggy FreeBS…

Never used Windows or FreeBSD guests, but what's wrong with Ansible provisioning?

It's difficult

Re: Decommissioning Otto

#69

Post by superswordfish »

axelfontaine wrote: post #67
eropple wrote: post #65

Earlier quoted context omitted.

Speaking for myself, I've put enough tooling around Packer to make it tolerable in my build pipeline and I haven't found a better tool that does what it does (suggestions welcome!). I would be using Vault, except it's just on the far edge of usable for me (particularly around the aws-ec2 provider) and I am repelled sufficiently by Go that I'm not driven to hack it to suit my needs until I exhaust my other alternative…

If what you are after is generating AMIs for JVM, Node.js or Go apps, we at Boxfuse offer a great alternative to Packer with very fast creation of very small and secure AMIs https://boxfuse.com/blog/amis-in-30-seconds Disclaimer: I am the founder and CEO of Boxfuse

Your limited, proprietary, paid tool? Yea...

Re: Decommissioning Otto

#70

Post by stephenr »

Mizza wrote: post #8

Does anybody _actually_ use the HashiCorp stack, besides Vagrant, for serious work? I tried and honestly found their products sorely, sorely lacking. Very shiny documentation, very incomplete, un-battled-tested tools, no examples given, little response from their devs other than the PR team. For a small example, I _still_ get +1 notifications on this critical issue nearly every day: https://github.com/mitchellh/packe…

Based on regular use of Vagrant and irregular use of Packer (to build base boxes for vagrant) I would be reluctant to rely on their other tools.

They mostly work, but it's never as seamless as they make it out to be.

I do appreciate the work that's put into these tools but I also believe anything can be criticised. That doesn't mean the author has to fix what everyone says is wrong but an attitude of "well it's free, either PR or stfu" doesn't help either.

Post reply on HN