Hacker News Reader

Comment by trevoranderson | Hacker News Reader

If you haven't tried Gitlab yet, I would give it a serious look.

You can self host it which is cool and means that if the company's funds dry up and they stops development (hopefully never), you can continue using the product. With Google Code shut down and Github losing 66 million in 9 months in 2016, it's an important question to ask about where you keep your source code.

Featurewise, I haven't noticed anything that I can't use Gitlab for that I wanted to do. The interface is still less familiar to me than Github, but that seems to be a function of it being changed relatively often as well as the time spent looking at it.

I also like how much Gitlab embraces open source. They recently acquired Gitter and will open source it. Even though I mainly just use it as an alternative to the still more popular Github, I'm considering picking up one of the Enterprise plans to help ensure the development pace continues like this!

View on HN

> Featurewise, I haven't noticed anything that I can't use Gitlab for that I wanted to do.

I've only noticed one: the ability to do less. Along with its excellent improvements over the years, it also gained some staggering RAM usage.

It used to be really nice for self-hosting your own private GitHub alternative on a little VPS or RasPi, but those days are apparently behind us.

Yeah, they’ve added more features which end up using more memory. For the small version you’re talking about you might look at gogs.io. Haven’t used it, but I hear good things.

Just for the record, I believe most of the community has moved from Gogs to Gitea (a fork) because of management issues in the original project.

Huh?

Gogs is fine.

People, for whatever reason, think devs have to continue adding more and more features otherwise they see the project as "dead". It's not.

In my opinion the Gitea folks are creating a bloated version of Gogs. YMMV.

My understanding was that the maintainer of Gogs had stopped replying to any issues/PRs for a period of a few weeks and refused to add any other maintainers to the project, so people wanted to fork and ensure that it'd always have an active maintainer.

https://blog.gitea.io/2016/12/welcome-to-gitea/

He did have some periods of inactivity. He didn't want to add other maintainers because he wanted to keep the code quality as good he wanted as far as I know and that's also why he didn't accept pr's, because the code quality wasn't good enough for him.

He was also totally ok with the fork to gitea.

Gitea exists so that more features can be added faster, not necessarily keeping the code quality ideal.

I prefer gogs. Better documentation. Original author.

For many more features I use gitlab. With gogs I'm looking for simplicity.

GitLab's RAM requirements have increased considerably and we now recommend 4GB. It still runs on a RasPi2 but it is not a great experience. We would love help running GitLab with Puma so that it is still fast if you have little memory https://gitlab.com/gitlab-org/gitlab-ce/issues/3592

Out of curiosity, how much of a difference would Puma make in terms of memory usage once available (specifically at the low end, i.e. for teams of less than 5 members)? And would you happen to have a rough idea of which features are the biggest contributors to the high memory usage of GitLab?

Being able to comfortably host a limited version of GitLab for a small team on the lowest tier of Digital Ocean or EC2 instances (<512MB) would be a game changer in terms of the accessibility of GitLab as a product. It would give you access to the long tail of potential privacy-conscious users who would like to avoid hosting their code on a SaaS service (and thus wouldn't be already captured by the market leader, GitHub), but might be put off by the high operational costs involved with self-hosting a GitLab instance compared to some of the lighter weight alternatives.

There are people at GitLab that know more, but my take is that it would make it more practical to run with 1GB.

As listed in https://docs.gitlab.com/ce/install/requirements.html the minimum is 1GB "1GB RAM + 3GB of swap is the absolute minimum but we strongly advise against this amount of memory. See the unicorn worker section below for more advice."

Most of the memory for the configuration of 4GB we recommend is to run extra unicorn workers https://docs.gitlab.com/ce/install/requirements.html#unicorn... You would save this memory because you can multithread one worker.

This depends on the amount of memory a Puma setup would use vs an equivalent Unicorn setup. Say you have determined you need e.g 10 Unicorn workers, each taking 300 MB. If you could handle the same traffic using 1 Puma process with 10 threads that means you just saved yourself ~2.7 GB of RAM.

It's entirely possible that a single Puma process may end up using more memory than a single Unicorn process, so the only way to know for certain is to measure the difference.

It may or may not help with the requirements, but when I was using it 6mo - 1yr ago you could definitely disable "extras" like wiki, issue management, etc. It was nice to simplify things for users and keep them from splitting info between that and your company's solutions for things like issue management.

Yes, you can disable what you don't need. For example you can even disable the repository and have just an issue tracker.

You should look at gitea ! It's much simpler than gitlab, but is also much more lightweight in terms of resource usage. Used it succesfully to host my startup's git.

If all you need is a server to host a bunch of git repos, there is also gitolite. http://gitolite.com/gitolite/

It doesn't have a pretty interface, but it works, and its requirements are minimal. I've had good success using it to host student repos in my classes for about 4 years.

Just wanted to add to this and point out that Gitter is already completely open-source and available at https://gitlab.com/gitlab-org/gitter/

To me, Gitlab's UI is a hybrid of Github and Bitbucket. Best of both worlds. IMO it's just a matter of time until it surpasses Github in terms of usability. Now, you raise a good point noting that Github's been operating on losses. How will Github make enough money to sustain growth as it becomes THE go-to place for developers of all kinds and levels?

We have an open core model where some features are only available in the proprietary version. We make 90% of our money selling subscriptions for https://about.gitlab.com/products/

GitLab.com subscriptions https://about.gitlab.com/gitlab-com/ are more recent but this is growing rapidly.

Any plans to go back to selling subscriptions for managed instances?

I'm leeching on gitlab.com today, but not enthused about it, if only because I've become disillusioned with GitLab itself.

I pay for email, and I'd be willing to pay for code hosting, too, on the following conditions:

- it's running something like Gogs/Gitea instead of GitLab

- whatever software's on the backend, my subscription needs to allow me to choose the community edition, not some enterprise version

- priced at XX/yr instead of X/mo

> priced at XX/yr instead of X/mo

I understand I risk looking like an idiot but did you mean X per user per month? If not, what's the difference?

No.

The difference between XX/yr and X/mo is that the latter is a tactic that relies on perceptual psychology to downplay the cost, same as when you see something priced at $3.99 instead of $4. It might be important to clarify that when I say XX/yr what I mean by XX is ~50USD. That's a reasonable price for a small instance (especially since the current going rate for something comparable is $0). Something priced at $8–9/mo, however, costs twice as much, even though the monthly pricing scheme helps to hide it. If you're buying hosting for your code, you don't really need the flexibility of monthly pricing. (What're you gonna do, spin up some short-lived instances for as long as you need them and then shut them down when you're "done" with them? No. The expected user model for code hosting is one that's cumulative.)

Side note: charging per user per month for a hosted instance would be pretty arbitrary, anyway. Put a cap on instance size, workload, number of simultaneous requests, and daily or hourly data in and data out, not on number of users. Charging by the user for a hosted instance is very nearly unashamed rent-seeking. The only reason it makes sense for GitLab EE is because it's on-prem.

Thank you for explaining it to me. It totally makes sense.

I can't help but think of the old "what are you trying to optimize for?" question though. Gitlab.com is OK for me because I don't make money writing code so I'm OK just getting free private repositories and any ci stuff I get is just gravy. I used to think "how will they ever make money?" and this makes sense.

I read this other story about how a DNS service was getting rid of its free tier because the free tier users can't convert. I think gitlab and github have found a good opportunity by giving away free service to free and open source software which means people will have experience using the service and might just keep using it at work (haven't checked but I'm sure github enterprise is not cheap).

GitLab is priced at XX per user per month. It can be paid annually. See https://about.gitlab.com/products/

I am a GitLab reseller and offer a 10% discount to HN users.

We don't have plans to sell subscriptions for managed instances. We found that it was not a great experience for customers unless we would do lots of development to make it work better. And then still you always have spend time making modifications that a customers request.

You're asking GitLab people for a subscription to a code hosting service on the condition it's not GitLab. Did I miss something?

You may have missed the fact that GitLab had an employee working on Gogs/Gitea full time.

Even ignoring that, telling someone their product isn't what you want and to please offer a different product isn't an outlandish thing to say.

We no longer have someone working on Gogs nor Gitea. Kim is now on the Gitaly team.

Hey, Sid. This is what I meant by my message that said you'd "had an employee working on Gogs/Gitea". Sorry if it was confusing.

Ah, I didn't note the past tense. Sorry for my oversight.

I use their community edition and barring few hiccups while deleting auto-merged branches, the whole product is beyond terrific. Highly recommended to host it yourself and have your source on it

Thanks for the kind words. BTW we fixed the hiccup you speak about. When you now create a merge request we added an option to set the state of 'delete branch on merge' on the merge request. You can set this to give a suggestion to the person merging, and it is off by default.

Given that git has such a distributed nature, couldn't you simply move over your code to Gitlab if and when Github actually is on the verge of shutting down?

I'm glad that Gitlab exists, however the lazy person in me sees no immediate need to use it. Even if all Github datacenters are destroyed by an asteroid, copies of repos I care about will still be around in enough people's computers to be able to recover.

What about issues, wiki and other non-git data?

(Nit: The GitHub wikis are also git repositories. https://help.github.com/articles/adding-and-editing-wiki-pag...)

But anyway, the same argument could be made of an asteroid hitting your own datacenter with self-hosted GitLab instead.

IIRC GitLab can import GitHub issues straight from the UI.

Also PRs will be imported, milestones, labels and more: https://docs.gitlab.com/ce/user/project/import/github.html#o...

Those can be extracted. Way more relevant imo: Links to your repo. For a mildly successful project a key part are external links pointing to your project. Those will point to a wrong location and you won't have control.

> You can self host it which is cool and means that if the company's funds dry up and they stops development (hopefully never), you can continue using the product.

You still have a git repo on both github and bitbucket, even if they stop development you will be able to download your code and host it anywhere you want. No?

Code yes. Issue tracker and other features? Not so easy.

It's web UI used to be way slower than GitHub. Has this improved?

We're really committed to improving the overall performance of GitLab.

You can take a look at some ongoing performance-related issues at https://gitlab.com/gitlab-org/gitlab-ce/issues?scope=all&utf... as well as some already completed improvements at https://gitlab.com/gitlab-org/gitlab-ce/issues?scope=all&utf...

There's also a plethora of ongoing efforts on making our hosted instance GitLab.com even faster. Lots of issues to look over at https://gitlab.com/gitlab-com/infrastructure/issues?scope=al...

Rights now our issue speed is much improved. Finally GitLab.com http://stats.pingdom.com/81vpf8jyr1h9/1902794 is faster than GitHub.com http://stats.pingdom.com/81vpf8jyr1h9/1902795 in one aspect. Still lots of work to do.

Searching is still slow but otherwise Gitlab is excellent, I'm using it every day.

It will get better when we enable Elasticsearch for the search feature. I don't think we have a canary issue but this may be useful: https://gitlab.com/gitlab-org/gitlab-ee/issues/3492

Yep, I would say so. In the past month or so, merging an MR has become as fast as github some times.

Their real-time updates like the pipeline completion signalling still has some lag. But I think the plan to improve real-time updates is on the roadmap for a big overhaul.

Something that I find missing is the ability to embed code snippets into other websites

I would love to see that too. Issue is at https://gitlab.com/gitlab-org/gitlab-ce/issues/8088

Reply on news.ycombinator.com