Atlassian announces end of support for Opsgenie(atlassian.com) |
Atlassian announces end of support for Opsgenie(atlassian.com) |
For months, we’ve heard from customers jumping ship to ilert, citing Opsgenie’s stagnation as Atlassian folded its features into Jira Service Management (JSM). JSM’s a beefy ITSM platform - great if you need the extras, overkill if you just want real-time incident response.
Now there’s Compass: a dev-centric service catalog with basic alerting, on-call, and real-time notifications. Replacement or sidekick? Compass is a standalone Opsgenie successor for devs, yet it complements JSM’s broader IT support scope. Together, they tag-team the incident game, but neither fully mirrors Opsgenie’s features.
ilert’s the alternative: a German-built incident response tool covering alerting, on-call, status pages, and call routing—tightly focused.
Thoughts? Any existing customers following one of the migration paths suggested by Atlassian?
While most companies can operate for some time without JIRA: losing your paging service means you're flying in the dark. And yet, Atlassian did not prioritize restoring OpsGenie.
I covered the details at the time [1]. To this date, this incident is a real head-scratcher and makes me wonder if Atlassian has internalized how much more critical an incident alerting software is, compared to a ticketing software (JIRA) or wiki (Confluent).
[1] https://newsletter.pragmaticengineer.com/i/52148641/what-atl...
https://webrazzi.com/en/2019/08/08/5-reasons-why-atlassian-b...
Damn. As a founder in the incident management space (Rootly) I've had a lot of respect for Opsgenie. They had unique features like heartbeats and ran a lean but mighty team before selling to Atlassian. We saw them the most in Europe by far.
Okay here comes to promotional part (don't hate me). If anyone is looking for a modern alternative to Opsgenie that isn't as expensive as PagerDuty, Rootly is worth checking out (Slack-native, holiday scheduling, request coverage, clean mobile app, etc).
Previous to Rootly I worked at Instacart where I helped us transition from PD to Opsgenie. Afterwards, still not being happy I built Rootly.
Today, we've helped Trivago, Motive, Yahoo, and quite a few others make the switch. Pretty easy with our importer tool, etc.
I championed and led the introduction of Rootly’s incident management product at two tech companies and counting. Rootly scales well - the Slack bot replaced chains of comms we used to have humans to handle. It’s reliable. And the attention to detail in their product feels like a love letter to SRE.
Rootly has the product chops to make a top notch incident management product. The domain modeling they’ve done for Rootly Oncall is solid - it feels like a superset of PagerDuty and Opsgenie.
Anyone moving off Opsgenie would benefit from a close look at it.
There are some pretty low hanging fruit stuff that would make the PD experience palatable (e.g. ability to page teams not just services, allow for partial overrides). These things for us were <1 week projects.
I do think PD is trying to rectify it by acquiring Jeli, giving it away for free in their new pricing plan, etc. Def feeling the squeeze a bit if I had to guess.
To not give too much of our roadmap away in a public forum, some of the basic components of overrides, paging groups and the web UI - keep your eyes peeled this year, one of our big focus areas is looking at how not only do we eliminate some of these gaps around on-call management and UI, but also going beyond just filling these table stakes. As you might imagine, PagerDuty scaled pretty aggressively (while also keeping our high reliability standards in place - customer trust is our #1 priority, after all we need to be up when everyone else is down), and we've had to make some investments to unlock our roadmap for this upcoming year.
Real talk though, happy to continue the conversation/ meet up over a zoom. Send me an email: dgodbout[@]pagerduty.com. There are humans over here at PagerDuty who care a lot about folks like yourself and I consider myself lucky to get build products for people who are responsible for ensuring things just "work".
One small bit: maybe include a link to your website? That would make it easier to clickthrough and looksee
https://rootly.com/ || and Opsgenie comparison (features): https://rootly.com/comparisons/opsgenie-vs-rootly-on-call
I don't get the pricing of PagerDuty and OpsGenie. It seems too expensive for us. We're only a few people on-call and we need something simple, barebones, that just works and works reliably.
You get on-call + incident response + status pages. And a pretty "defaults only" experience ootb.
Is weirdly broken (at least for me) on Safari on desktop. Every mouse movement or key click causes all the tiles to enlarge.
It's fine on Chrome and Firefox.
However, one make-or-break feature - does Rootly have a Terraform provider? We rely heavily on Terraform for building out our Opsgenie setup.
[1]: https://registry.terraform.io/providers/rootlyhq/rootly/late...
Super bizarre. We're a large Opsgenie customer. The Opsgenie website or mobile apps never showed any notice, not even as of now. The "Announcements" section in the mobile and web apps as I'm writing reads a feature announcement: "Coming soon: Simplified integration setup experience!"
Our customer relationship nor billing teams never received any communication.
Atlassian — we don't mind you sunsetting any product, that's fine. But honestly, your paying customers shouldn't get to know this from a blog post in social media, especially for an On-call emergency product.
I wouldn't trust any product from Atlassian after this fiasco.
> Opsgenie end of support – effective April 5th, 2027:
A public post followed up by them reaching out individually (which takes time) along with a 2 year grace period, seems pretty reasonable?
Anyway the product map of Atlassian is unreadable now (Jira Work ≠ Jira Portfolio ≠ Jira Software ≠ Jira Service Management, all in Jira), and they don’t make clear what you’re subscribed to in the administration: Products, addons, same products but other sites, etc. This mess is downright visible in the announcement:
> Starting today, there are two options for Opsgenie customers: move to Jira Service Management for robust end-to-end incident management, or move to Compass for alerting and on-call management alongside an intuitive software component catalog.
I think they botched the project where they unified the login, and failed to make a centralized dashboard. In any case, I always wonder whether that’s intentional or a dedicated effort to make people spend more.
Sep 8, 2022 - I sent a 2 line PR to fix an incorrect spelling of a json field and add a missing one in their Go SDK.
June 23, 2023 (9 months later) - It was reviewed and merged.
<self-promotion>
We built Better Stack (https://betterstack.com/incident-management) after being frustrated with PagerDuty a couple years back. It's a solid place to land if you need to migrate away from Opsgenie.
Let me know at juraj@betterstack.com if you have questions, happy to help! (I'm the founder)
</self-promotion>
We also use the AI features quite a bit, but they can be easily ignored if unwanted
I don't want my incident management system to be AI anything. But then I thought maybe I'm being unreasonable so I checked out https://betterstack.com/docs/uptime/incident-silencing/ which was the first thing I found that was AI related.
Frankly, the thought of some black box AI system deciding when to silence incidents utterly terrifies me. I want to be absolutely 100% certain that I'm receiving all my alerts. If they're too noisy or repetitive, it's on me and my team to improve the alert quality, not leave it to "AI" to figure out what to silence.
The phrase "AI-native" increases your startup's valuation by 100x nowadays
(just kidding)
The confusion around what's actually happening is understandable. While services will continue until 2027, end-of-sale in 2025 means many teams are already evaluating alternatives rather than waiting for the last minute.
Something that really resonated with me from gregdoesit's comment was that reminder of the 2022 outage where OpsGenie was down for two weeks for some customers. That incident highlighted something we think about constantly at Zenduty: paging services are mission-critical infrastructure.
When your alerting system goes down, you're essentially flying blind. I've been working with teams dealing with exactly these transitions, and the migration challenges are real - it's never just "update a webhook URL" as someone mentioned. Your alert routing rules, escalation policies, on-call schedules, and integrations are complex systems that need careful migration.
What I'm curious about: for those of you considering alternatives, what features matter most to you? Is it reliability? Pricing? Integration flexibility? The ability to reduce alert fatigue? Feel free to drop your thoughts here or DM me. Always happy to chat about these challenges even if you're not looking at Zenduty specifically. These transitions are stressful, and having been through them myself, I know how disruptive they can be to engineering teams who just want to focus on building their products.
Here's a detailed comparison of all the opsgenie features and what you need to know before you decide to migrate to a different tool - http://zenduty.com/zenduty-the-best-opsgenie-alternative-com...
Another one that's surprisingly important to my employer: select the country of origin of phone calls.
The company I work for offers (among other things) datacenter colocation in Germany, and one of the selling points to compete against AWS is German ownership, no foreign control. Yes, some of our customers are very conservative. A few of our customers get direct notifications from our alerting tool, and if those phone calls arrive from a US-based number, it leaves a really bad impression.
OpsGenie has instances hosted in EU for that.
And they still haven’t finished making deployments have feature parity with builds in Bamboo.
I made it partway through the interview process at a startup a couple weeks ago and realized I didn’t have a good answer if they asked my advice for what to use for project management, if not Atlassian. Which was fairly likely given the company maturity and the position. I’m still trying to find that answer.
Aside from GitHub I only have one other answer and it starts with, “this is going to sound crazy but hear me out.”
We’ve got a good migration story (import OG schedules and escalation paths, etc) so most customers just migrate to us, but if you didn’t have that option the Atlassian migration is much more painful.
To me it sounds like this API will stop responding after 2027-04, so you need to rework your integration. Not just a pricing/packaging/branding change.
If you have sources that disagree, I'd be happy to read them.
There’s a first wave of incident startups that responded to the market having stagnated about 4 years ago (incident.io, FireHydrant, Rootly) then a slew of extremely recent (<1 year) companies leaning into AI incident response.
It’s weird that Opsgenie is just quitting that race but realistically they weren’t really competing in terms of pace of development. Felt more like Opsgenie was bought under the assumption IR was a ‘solved’ problem that Atlassian could just add to their stack and be done with it, while today it’s increasingly apparently that just paging someone is the smallest part.
Probably part of it but every job I've had has used a different alerting service anyway (sometimes multiple).
https://www.atlassian.com/software/jira/service-management/p...
it's a "no frills, in a good way" type of product. dirt-simple API with straightforward pricing ($5 lifetime subscription) and limits so generous I've never worried about hitting them (10k messages/month).
Actually one company I was at did use Stride and it was truly awful - I don't know how its product manager thought it could compete without such basic features as... the ability to edit and delete a message. But some fool at the company I was at chose it because it was cheaper than Slack.
However, it was not easy to find the announcement about the closure. The title of the announcement was very confusing. That is why my teammate decided to write an article with a clearer title. He did highlight why moving to TaskCall would be easy for current OpsGenie clients without impacting their operations, but here is a link of to the article if anyone is interested: https://medium.com/@riasat.ullah/opsgenie-reaches-end-of-lif...
Most of them can be self-hosted so you won't be locked in in case that happens again.
But today's incident management needs to be a lot more than just paging.
For anyone looking to unlock a lot more value out of their incident management tool, may I suggest https://www.temperstack.com. A number of OpsGenie customers have already switched to Temperstack even before this announcement came out.
I'm one of the cofounders of Temperstack and have personally helped companies make this transition. And I would be happy to get you set up as well. Feel free to drop me a line on amal@temperstack.com
The broader trend here is the shift from unbundling to consolidation. Over the past decade, many “features” were created as standalone SaaS products, but that era is winding down. We’re now seeing the pendulum swing back, with more consolidation across the industry.
In my opinion, OpsGenie should have been a built-in feature of Jira Service Management from day one of the acquisition.
The amount of opportunity costs that racked up nearly killed them.
Or do something sane like move off Atlassian products completely.
Our teams feature is most like OpsGenie compared with others.
We've created All Quiet as an Alternative to Opsgenie and PagerDuty et al. We are building an incident management platform for developers from developers. We devs don't need a "full service platform" that also brews our team's coffee. We simply want to get a critical notification on our phones when sth's broken.
Check out how we compare to Opsgenie: https://allquiet.app/opsgenie-alternative
Kinda like that this one is cheaper on the SSO plan than the other advertised here that starts at $20/user and proudly advertises no SSO tax.
Kinda easy if you make your plan as expensive as github enterprise.
I know Mercurial isn't overly popular but I thought it was wise to have a point of difference to the other offerings.
The writing has been on the wall for a while with Opsgenie though. I work at incident.io and loads of our customers have had deprecation/migration warnings for a year or so now, pushing them away from Opsgenie and into different Atlassian products (with no migration plan, sadly).
It’s been great for us as they tend to move to us (we have good migration support for Opsgenie) but confusing to watch from the outside as Atlassian never stopped selling the product even while actively shutting accounts down.
For us it was a seamless migration for us to Jira Service Management from OpsGenie. There were many emails, and many in app banner notifications for MONTHS. All our rules transferred over. I only had to setup our notification integrations so all links into email/slack goto Jira Service Mgmt instead of OpsGeneie. Our OpsGeneie instance is currently read only.
We take extensibility pretty seriously. Outside the basics, we just released our own agent ready API (https://rootly.com/blog/introducing-rootlys-api-ai-agent-fir...), MCP, docs are now LLM ready (https://docs.rootly.com/llms-full.txt).
I can't think of any company off the top of my head that has acquired a company and successfully integrated their tech stack into their own product offering. I work for a huge health care company whose stated goals are to grow through acquisition and in the 5 years I've been here, none of the painfully long process of integrating another companies tech has gone well - at all.
If anybody has any good examples of this happening, I would love to read about them.
Once you start to see Jira as 'central repository for realizing work streams' the options become easier to assess.
The place where Jira does sem to excel is automations, but I think thats simply due to maturity and alot of platforms cover most common ground
Do tell...
I thought it was going to be a shitshow but it turned out to hit the Pareto Front for project management, especially if you were doing Kanban. It’s one of the first tools that auto-linked between wiki, tickets, commits and commit messages. You could put ticket numbers and Wikiwords into your commit messages and it just worked.
I’m trying to design a Trac but with CI now, but I’m working on a personal time management app instead, because I don’t build panipticons, so those features need to live where management cannot see.
Distributed scaling is a little difficult, you have to set up a webcam and hire an intern to move things around, but it's probably still cheaper than a lot of enterprise solutions.
People LOVED it. Worked amazingly well, was easy to hand off if I was out, etc.