Okta's NextJS-0auth troubles
joshua.hu199 points by ramimac 3 days ago
199 points by ramimac 3 days ago
Okta requiring to create a video for a pretty obvious vulnerability shows that Okta does not take security seriously, contrary to what they say at their earnings calls. Sounds like deceiving their investors.
That’s funny. I spotted a similar issue in their Go SDK[1] a few years back. I was pretty appalled to see such a basic mistake from a security company, but then again it is Okta. [1]: https://github.com/okta/okta-sdk-golang/issues/306
Kind of funny that stalebots are the new "won't fix" methodology to ignore security issues with plausible deniability.
Yeah I got a kick out of that. "We might have fixed your issue, if we didn't, open a new one because we took so long acknowledging this one".
> I was pretty appalled to see such a basic mistake from a security company, but then again it is Okta.
Oh. Em. Gee.
Is this a common take on Okta? The article and comments suggest...maybe? That is frightening considering how many customers depend on Okta and Auth0.
We evaluated them a while ago but concluded it was amateur-hour all the way down. They seem to be one of those classic tech companies where 90% of resources go to sales/marketing, and engineering remains "minimum viable" hoping they get an exit before anyone notices.
I'm convinced Okta's entire business model is undercutting everyone with a worse product with worse engineering that checks more boxes on the feature page, knowing IT procurement people aren't technical and think more checkboxes means it's better.
Okta sucks balls. That's from my perspective as a poor sod who's responsible for some sliver of security at this S&P listed megacorp that makes its purchasing decisions based on golf partners.
Among the reasons to leave my last job was a CISO and his minion who insisted spending $50k+ on Okta for their b2b customer and employee authentication was a bulletproof move.
When I brought it up, they said they didn't have anyone smart enough to host an identity solution.
They didn't have anyone smart enough to use Okta either. I had caught multiple dealbreakers-for-me such dubious / conflicting config settings resulting in exposures, actual outages caused by forced upgrades, not to mention their lackluster responses to bona fide incidents over the years.
I use Authentik for SSO in my homelab, fwiw.
Yep. They're an Enterprise™ company. That means they prioritize features purchasing departments want, not functionality.
Yeah, I have the misfortune of inheriting a SaaS that built on auth0, and the whole stack is rather clownish. But they tick all the regulatory boxes, so we're probably stuck with them (until they suffer a newsworthy breach, at any rate...)
okta is the worst. Their support is the worst (we always got someone overseas who only seemed to understand anything, probably they were trained on some corpus) and would take forever to loop in anyone that could actually help.
We've recently moved to Auth0. I'm no security expert. Whats the recommended alternative that provides the same features and price, but without the risks suggested here?
Heya, I work for FusionAuth. We have a comparable product for many use cases.
Happy to chat (email in profile), or you can visit our comparison page[0] or detailed technical migration guide[1].
0: https://fusionauth.io/compare/fusionauth-vs-auth0
1: https://fusionauth.io/docs/lifecycle/migrate-users/provider-...
If you’re looking for b2b identity, I’m the founder of WorkOS and we power this for a bunch of apps. Feel free to email me, mg@workos.com
It's not difficult to implement OAuth2. There are good libraries, and even the spec is not complicated. Or use AWS Cognito.
I used Auth0's Nextjs package for a previous application I built. Auth0 seemed really tempting to use at first glance – support for lots of languages, lots of marketing, and lots of features. It reality, it was a nightmare to use because its documentation is bad and there are a whole host of edge case bugs that never seem to make its way out of auth0 forums and into an actual PR.
Okta is, if you may excuse my French, straight garbage.
I think GitHub should allow disabling PRs. I don't believe most big corporations are interested in dealing with fly-by contributions because it might make them look bad or be riddled with quality issues.
Also some projects like the Linux kernel are just mirrors and would be better off with that functionality disabled.
While that is true, I feel like it is irrelevant here since it seems like Okta definitely wants (and perhaps needs) the fixes. God only knows why GitHub still forces it on though. Early on it might've been some mechanism to encourage people to accept contributions to push the social coding aspect, but at this point I have no idea who this benefits, it mostly confuses people when a project doesn't accept PRs.
> Okta definitely wants (and perhaps needs) the fixes
They definitely don't want them if their process requires signed commits and their solution is 1) open another PR with the authors info then sign it for them, and 2) add AI into the mix because git is too hard I guess?
No matter how you slice it, it doesn't seem like there are Okta employees who want to be taking changes from third parties.
GitHub actually can natively mark a repo as a mirror (or could? I can’t find an example now, but they have always been rare). The book-with-bookmark icon before “user / repo” in the page header is replaced by a mirror-and-reflection-ish–looking thing, and the badge after it changes from “Public” to “Public mirror”. Unfortunately, forcing you into “social coding” (wait, is that no longer on the homepage?) takes priority, so that mark can only be given out by GitHub staff through manual intervention, and it doesn’t often happen.
Maybe the community should use less of github if github doesn't provide features the community finds useful.
I think it is distasteful and disrespectful to call out an employee by name in this way, regardless of the merit of the rest of the OP's post.
well, it was distasteful of to them to close op's pr and apply the same patch with improper attribution, and then use ai to respond when they were asked about it
I agree with the parent post that it's distasteful.
There's no value in naming the employee. Whatever that employee did, if the company needed to figure out who it was, they can from the commit hashes, etc. But there's no value in the public knowing the employee's name.
Remember that if someone Googles this person for a newer job, it might show up. This is the sort of stuff that can disproportionately harm that person's ability to get a job in the future, even if they made a small mistake (they even apologized for it and was open about what caused it).
So no, it's completely unnecessary and irrelevant to the post.
> Remember that if someone Googles this person for a newer job, it might show up.
Not to sound too harsh, but this is a person who rudely let AI perform a task badly which should have been handled by just… merging/rebasing the PR after confirming it does what it should do, then couldn't be bothered to reply and instead let the robot handle it, and then refused to fix the mess they made (making the apology void).
That's three strikes.
What if it's some junior given a job beyond their abilities, and struggling manfully using whatever tools they have to hand. Is it worth publicly trashing their name? What does their name really add to this article?
I agree what occurred is quite egregious. But "use ai to talk to customers" and "play games with signed commits" sound much more like corporate policy than one employees mistake.
> This is the sort of stuff that can disproportionately harm that person's ability to get a job in the future.
Isn't that beneficial in this case?
> Remember that if someone Googles this person for a newer job, it might show up.
That's the whole point; I sincerely hope it does. Why would anyone want to hire someone that delegates their core job to a slop generator?
I don't think it is distasteful or disrespectful, he's just explaining what happened and why, and he's obviously unhappy with the whole ordeal.
You couldn't pay me a billion dollars to use Okta.
Sadly many people will spend a million dollars to use Okta for their 10,000 logins/day (read: <1 tps) instead of running their own Keycloak or Authentik or whatever.
OIDC is not scary, and advanced central authorization features (beyond group memberships) are a big ole YAGNI / complexity trap.
Running your own local AuthN/AuthZ is more than just 'install it on a box in the closet'. I don't blame anyone for letting one of the giants do this on their behalf -- they have the expertise, though I agree I wouldn't touch Okta.
For your average enterprise it really is that simple. Register some IDPs. Connect a backend. Add some clients over time.
Yes, you need someone to wear the IAM admin hat. But once you get it configured and running it requires 0.1 FTE or less (likely identical to whatever your Okta admin would be). Not worth 6+ figures a year and exposure to Okta breach risk.
No, it isn't "simple". Protecting your IdP is critical and not easy.
Yes, creating a SAML integration is easy, but that's only one piece of the puzzle.
The workload to run Authentik locally is about identical to the workload to set up and configure Okta. (Or you could just fine someone who will host Authentik for you, if deploying a container is too hard for you.)
I'm currently building on the Auth0 SaaStarter because it seemed to be the only option in the market for something with all the core features enterprises are looking for. Is there an alternative that doesn't require building from scratch?
I find it funny that this seemingly fictitious person Simen A. W. Olsen my@simen.io will forever be engraved as a co-author of a one-line change in the nextjs-auth0 repo.
Honestly when I saw Okta in the headline, I had assumed the article was going to say they were breached again.
This one is amusing, and as another comment mentioned below, large companies are awful at accepting patches on github. Most use one-way sync tools to push from their internal repositories to github.
I've been quite happy with FusionAuth so far. Free to run on your own server, easy to understand and set up, easy to program against, reliable.
We're another happy FusionAuth customer. We started with self-hosted but just moved to their hosted option this year.
IANAL but unfortunately, I think the fix itself shown here might be too simple to actually clear the bar for copyright eligibility. (And in fairness to copyright law, it is basically the only sane way to fix this.) That means that there's probably not much you can really do, but I will say this looks fucking pathetic, Okta.
I'm more confused by the fact that the OP freely submits a PR into an open source repo but then wants to use "copyright" because the code he submitted ended up being used under the wrong name, which was then corrected.
Why is it confusing to you to expect attribution?
thats not the confusing part, its rather confusing to threaten to sue for copyright because of mistaken attirbution
He even asked them to force-push a new history because they got the name wrong!
Mistakes happen, I guess this hurts his 'commits in a public repo' cv score.
That maintainer seems clueless
FWIW, the employee reply (who the author is putting on blast) seems like it was written by a human, not an AI.
"You're absolutely right!" is the Claude cliche (not a ChatGPT one) - "You are absolutely correct." is not that.
Directly from the employee (tusharpandey13) in the github PR:
> Yeah, i had to manually stop it and delete the ai-generated comment.
I LOVE LLMs as a learning tool. I HATE LLMs as a communication tool. I know, there are people with serious handicaps who benefit from LLMs in this area. If only I could talk to those people and not wade through all this other garbage.
Especially when the AI is being represented as a person, this to me is dishonest. Not to mention annoying, almost more-so than the number of different apps that think they are important enough to send me push notifications to fill out a survey (don’t even get me started).
LLMs have definitely helped me reduce my social anxiety when writing, especially in a technical work setting. I don’t use it like the respondent in the article though, I would feel really embarassed to not edit an llm’s output to be in my own voice. But I feel it helps provide me with some structure in whatever I’m trying to write when I don’t have the mental energy or wherewithal to provide it myself.
You're either free OSS that gets flooded with AI slop PRs to overwhelm maintainers or you're a corporate OSS that uses AI slop to frustrate contributors. Are there any positive stories I've not seen?
Is there any non shite managed oAuth solution with a free tier available?
Auth0 really is super easy and comfortable to integrate and I don‘t want to run my own keycloak or whatever.
Security companies that prioritize bugs being sold rather than be reported will eventually blow up. Good luck Okta shareholders.
[dead]
[flagged]
Is this ai generated
More than likely. Look at the users most recent comment with the random “ at the end too.
[flagged]
WTF is Okta?
Basically an enterprise single sign on solution. We use it to allow staff to sign into pretty much any external service using Gsuite credentials.
An auth integrator, a pretty notable one, mostly (originally?) OAuth I think. Multiple people calling it a trash fire here came as a surprise to me, but I defer to their experience.