I was sucked in by Gitlab's feature list and marketing. But holy shit so many things are half assed and so many features are buggy and broken. Like their wikis would sometimes just straight up delete text and their webhooks would break images. I started a bunch of issues but they really have no chance of getting fixed anytime soon. Just shows how different your impression of something is after actually using it.
Our teams evaluated Gitlab for a year before completely migrating away from it (to Github Enterprise for SCM, Confluence for wiki, YouTrack for issues and TeamCity for CI).
Not a single team (out of 20) was happy with the overall performance (and especially the performance of code search).
As far as wiki is concerned, Confluence's UI has its own share of issues, even after their recent overhaul, but all in all it is much better product. YouTrack and TeamCity are simply much more polished & stable products and are (in our experience) easier to administer than the competing offerings from Gitlab.
> Not a single team (out of 20) was happy with the overall performance (and especially the performance of code search).
Hi! I'm the current PM for our search and as Sid mentioned we've steadily been working to improve that. It's been getting a lot better, but most of the improvements are heavily reliant on also having Elasticsearch enabled. Without that, there's really no way for us to provide a optimal search experience for the amount of data and content there.
If you have any specific search feedback, please feel free to open an issue or reach out to me @phikai on GitLab.
I don't really use gitlab, but I wrote my own git implementation recently (https://github.com/oridb/git9), and I've had bug reports when people tried to use it on gitlab.
Apparently it can deadlock processes on the server, stopping the clone -- and they never get cleaned up. While they don't prevent anyone else from using the machine, they do sit around wasting resources, and potentially causing a denial of service attack.
I figured out bugs for github's issues, but I haven't been able to find one for gitlab's, and they've been pretty silent on the bug that someone else filed about this as far as I can tell.
It didn't feel like a particularly easy to trace system when I tried debugging.
My impression is the opposite. Compared to Github, Gitlab is simply phenomenal, especially the UX. It's honestly one of the few pieces of software that I truly enjoy using every day.
Besides, if you manage to find their issue you'd see carnage this decision caused for some of the customers. Even those with own runners found them swamped in jobs doomed to fail , blocking actually inportant pipelines.
Autodevops as a product feature targets very niche audience of teams running gitlab, but without dedicated DevOps team , it is by definition a very opinionated view on CI/CD and I can not see how it was acceptable to suddenly enable it on all existing projects on whole gitlab.com overnight.
I agree with you, we don't currently cover all the use cases we'd like to, but we're working to expand the feature set and technologies we cover.
If you had a particular use case that was not well covered, I'd love to learn more about it so we can evaluate related improvements. You can reach me at daniel at gitlab dot com. Thanks.
The wiki is GitLab should not delete text, please file a bug if you had dataloss.
The current state of the wiki isn't great but we're also not seeing a lot of people care about it. Many people are switching to static websites. Therefore we're not investing to get the wiki better than the current state of a half circle.
A wiki's not the same usecase as static site, a static site is for marketing and wikis are for project management. I mean maybe people would use it if it weren't crappy? I would've been happier if the wiki didn't exist and I didn't invest time into wrangling it. ATM it's just a frustration.
It's not dataloss but it deletes text in the same line as wiki internal links. And what kind of wiki doesn't use lots of internal links. https://gitlab.com/gitlab-org/gitlab-ce/issues/67132 This basically makes the wiki unusable but it's marked as "backlog".
If you really do feel this way about the wiki then it should be clearly marked in your marketing. I read about your "integrated devops experience" and was onboard, but if your intent is as you say then you need to put parentheses next to the wiki feature item list that says "this is crap and we intend for it to stay crap" and we'll know to value it appropriately when choosing a provider.
I'm surprised that this wiki bug that two people here mention has only one upvote https://gitlab.com/gitlab-org/gitlab-ce/issues/48641 I'm not sure if the complaints are over multiple issues, if the wiki doesn't get a lot of usage, or something else is going on. The bug sounds legitimate.
> A wiki's not the same usecase as static site, a static site is for marketing and wikis are for project management. I mean maybe people would use it if it weren't crappy? I would've been happier if the wiki didn't exist and I didn't invest time into wrangling it. ATM it's just a frustration.
Hi! I'm the current PM for our wiki's and I agree that the experience isn't up to par at the moment. As Sid mentioned, historically wiki's were not a priority as many users weren't coming to GitLab for those features. We're starting to see shifts in that thinking as we've penetrated deeper in to some markets so we're working to adjust accordingly.
> It's not dataloss but it deletes text in the same line as wiki internal links. And what kind of wiki doesn't use lots of internal links. https://gitlab.com/gitlab-org/gitlab-ce/issues/67132 This basically makes the wiki unusable but it's marked as "backlog".
That issue is interesting because it's a link mechanism that relies on the underlying wiki project that we use. (i.e. that's no common markdown syntax). Discoverability of things like that is limited to some power users who know that we're using Gollum underneath and know of some of the supported syntax. What we've seen is that most users don't use that functionality and instead just use absolute links in markdown. This likely explains why the upvotes are so low on the bug report as well.
As an FYI, I've also asked a couple of our engineers to take a look at that specific issue to see if it's something that can be relatively easy to fix. No guarantees on anything here, but we'll try to get a bit more relevant engineering information to assist in understanding scope.
> If you really do feel this way about the wiki then it should be clearly marked in your marketing.
That's a lot of what our maturity pages are trying to do. We're being honest with ourselves and with our users about where we think the functionality of certain features are. In fact, when I started we had the Wiki listed at a `Complete` maturity which I reduced to viable: https://gitlab.com/gitlab-com/www-gitlab-com/merge_requests/...
If you take a look at the wiki strategy (https://about.gitlab.com/direction/create/wiki/) our next focus is going to be on making editing easier and improving navigation. We think these things will start to move the conversation forward for users and we'll see more adoption.
Even more, if you think we're not moving in the right direction with the Wiki's please be vocal on the issue tracker, or open an issue for the Wiki Strategy or even a merge request to change something. We're happy to have the feedback, and don't hesitate to tag me on wiki issues I'm @phikai on GitLab as well.
Thanks for explaining and for taking another look at the issue. Your comment makes sense to me in context of your company's strategy. I'm not saying that's wrong, as surely the breath allows you to gain an niche that is less competitive and less risky than a depth strategy that forces you to compete head to head with great products like notion... but as this thread shows it has enormous downside.
But anyway the double brackets is not a Gollum feature. It's a wiki thing, going back to wikipedia.
Since you're here I think this issue is also misprioritized: https://gitlab.com/gitlab-org/gitlab-ce/issues/66898 You should reconsider as a lot of integrations rely on webhooks and broken images make the whole thing impossible.
> But anyway the double brackets is not a Gollum feature. It's a wiki thing, going back to wikipedia.
That's fair, the point was that it's syntax specific to the wiki system vs. the rest of our markdown filters.
> Since you're here I think this issue is also misprioritized: https://gitlab.com/gitlab-org/gitlab-ce/issues/66898 You should reconsider as a lot of integrations rely on webhooks and broken images make the whole thing impossible.
I'll take a look at this - it popped on to my radar recently but I need to dig in further to understand the use case and functionality here. Thanks for bringing it up.
Perhaps an argument for focusing on doing a few things well?
I work for a GitLab shop, but we don't use the bugtracker (Jira) or wiki (Confluence), we mostly don't use the CI (Jenkins is just more flexible), and we definitely don't use the container stuff. We currently do use the review workflow but at various points have considered moving to a different tool for that as well.
So yeah, we would have been fine with a GitLab that did less— a lot less, and instead prioritized supplying first class integration points and supported plugins for other best-in-class tools.
Hi there! Just wanted to add my two cents here. Just like you choose to use SCM and workflow review, we have a lot of other users and customers utilizing different permutations and combinations of the various stages. Over time, various stages all move toward maturity, some faster than others. I know this is a different model than most software companies and it has its downsides. But the net positive is that we are able to provide an increasingly unified experience for DevOps. Note that at other companies the same thing happens except they market different stages as separate "products". They gain on the marketing but lose on the ultimate user benefit (again, IMHO).
I would much rather have fewer features which work well. For example, until recently there was a restriction that jobs could only run in parallel if they were defined in the same "stage", and no two stages could have running jobs at the same time. This means that a job whose dependencies were met could not necessarily be started because it belonged to a later stage (and jobs cannot depend on other jobs from its own stage, more on that later).
GitLab then introduced another way of specifying dependencies using the keyword "needs" ("dependencies" was already taken by the half-baked implementation). This allows jobs from multiple stages to run concurrently, as you would expect. However, you are still required to shoehorn your jobs into stages, even though they should be completely deprecated by now. Worse, the restriction that jobs cannot depend on other jobs from the same stage still remains. On top of that, the visualization of a pipeline pretends that the new way of specifying dependencies does not exist, so all the arrows between the jobs are meaningless.
I would much rather have had a proper implementation of the simple, more general approach, than having to deal with the legacy of a hacky and half-baked solution.
We released DAG as an MVC, which helped a lot of people out even in its current state. We do release features here iteratively intentionally, with the idea that feedback will help make future iterations better in unexpected ways compared to if we released a big feature all at once.
The items you mention are scheduled for follow-ups in our epic https://gitlab.com/groups/gitlab-org/-/epics/1716. Your feedback on sequencing or how we are approaching the different improvements is more than welcome.
I would have loved for the feature to be useful for you too in the MVC iteration, and I'm sorry it wasn't. We are still working on it, though, and I hope that it does become valuable for you also. In the meantime you should still be able to use GitLab in the same way you always have - let me know if you're having trouble running pipelines without the DAG.
It's not that the DAG feature isn't useful to me, the problem is that it has to coexist with a badly engineered version (the stage based model) which should never have been released in the first place. It is better to have GitLab not support CI runners than it is to push something that is so badly engineered. Better to delay that feature for 6 months than to increase the amount of legacy you push downstream to your users.
Also, this is just a single example that I pointed out to outline what I think is a problem with the whole development culture around GitLab, which puts too much focus on releasing early, and too little focus on quality assurance. Of course, you shouldn't spend years perfecting the next release, but you release too many features too early for my taste.
Ah, I understand. We do have users who prefer the stage model or hybrid DAG mixed with stages, but I get that that isn't your point of view. I do hope that after the next couple releases it becomes something more useful for you - your use case is important too.
Releasing early to get feedback on issues is important to us but we can do better communicating around what's an early preview vs. a mature feature. The maturity page that's linked to in this discussion is actually part of how we are trying to improve our communication around that. This is more at the stage and category level, but features have a maturity level as well and it's worth us reflecting on how that can be made more clear.
Sorry if I am a bit frank, but I simply do not believe you when you say you have users who prefer the "hybrid DAG mixed with stages". That's a bit like Microsoft saying that there are users who enjoy the 255 character path limit. My use case isn't an "improved DAG model", it is a DAG model without the pointless restrictions introduced by broken legacy.
In the DAG model (which really is just the model implemented by every sane build tool ever made), stages give you no extra expressive power, they only add arbitrary restrictions to that model and aren't useful at all.
Whole this discussion is mostly about how feedback is not taken into account and customers and users are not heard. Gitlabbers keep using this "release early and then let feedback to shape future " mantra, but in practice it is rarely happen or at least doesn't happen quick enough.
Hi! GitLab employee, first of all thank you so much for the detailed feedback! It's part of my job to relay feedback and make customers feel heard, so I'll talk to my team on ways to improve (you can see my job description here if you want: https://about.gitlab.com/handbook/marketing/community-relati...)
As per your second point, and this is speaking personally and not really officially, I do think our feature goal has been historically a bit too ambitious (although hopefully transparently so: https://about.gitlab.com/company/strategy/#breadth-over-dept...) seeing as how we're barely out of the startup phase. But we have had a big hiring push this year and have almost tripled our headcount (which did introduce growing pains of it's own lol). Once we've stablilized a bit, we will be able to dedicate more resources solely on maturing features.
I hope this doesn't come across like making excuses, these are just my observations as a user-turned-employee. I will take your feedback about not being heard into consideration though so we can improve on that!
It's also an attitude that pushes the cost of wrong decisions to the users, and really it just sounds like an excuse for not spending the internal resources required to do proper designs and quality assurance.
But the net positive is that we are able to provide an increasingly unified experience for DevOps.
The point of the GP and others where that as of now there is nothing positive about the "unified experience for DevOps", it is very buggy and lacking, maybe someday.
I'd rather have no (eg) wiki feature at all then have a half-assed one.
Because _if_ I feel like the feature might be useful, I'm going to try to use it, and invest time in trying to use it, and only then find that it's not reliable or well-polished after all. The more such features there are, the more chance I'll spend at least some time being frustrated by at least one of them.
One of the joys of really well-polished software is that if the feature wasn't worth doing right (and not every feature is!), it simply isn't there at all, so if a feature is there I can count on it being well-thought-out and well-executed.
They love producing public facing verbose process documents... At first I thought, oh these are some neat ideas, but the more that they produce, the more I wonder how they get anything done with so much (verbose) process.
I work at GitLab and can shed some light here. As a 100% remote company we work in all timezone. Writing things down actually helps make things more efficient. At my last job, to get anything done I needed to call a meeting because state lived in people's heads. When someone left, of if you couldn't get them to a meeting, it was very painful. It took forever to get things done because you needed to physically be present in order to work. At GitLab because so much is documented it's easy to pick up state and collaborate with co-workers in all timezones. What's cool, is because it's public, other random people will jump in and help you out that you'd never expect since they have access to everything you have.
