Skip to main content

A Guide to OpenTelemetry - Part 5: The Challenges

Pete Goldin
APMdigest

While OpenTelemetry offers many advantages, the experts point out several challenges as well.

Start with: A Guide to OpenTelemetry - Part 1

Start with: A Guide to OpenTelemetry - Part 2: When Will OTel Be Ready?

Start with: A Guide to OpenTelemetry - Part 3: The Advantages

Start with: A Guide to OpenTelemetry - Part 4: The Results

The Project is Not Mature

Maybe the greatest challenge for OpenTelemetry is that the project is not mature. While the tracing component is fairly well advanced, the metrics and logging parts are still being formed.

"Currently, the project is not mature enough to support every stack, language, and signal," says Michael Haberman, CTO and Co-Founder of Aspecto. "While we believe it'll get there, the road to full stability is long."

"OpenTelemetry remains a young project in many ways, and many components are still in alpha or beta," explains Austin Parker, Head of Developer Relations at Lightstep by ServiceNow. "There is still work being done to bring in new signals such as profiling, logging, or real user monitoring (RUM), and breaking changes in those signals can be frequent."

"Logs and metrics are slowly catching up, but the API is still unstable which can be an issue," adds Vladimir Mihailenco, Co-Founder of Uptrace. "Logs and metrics are not stable yet so using them is more bumpy and requires some involvement with OpenTelemetry development and reading various changelogs."

Varying Quality

"While the existing OpenTelemetry libraries are already viable for manual and automatic instrumentation, which is a key part of any observability solution, it varies in quality," says Daniel Khan, Director of Product Management (Telemetry) at Sentry. "With new versions of libraries being released almost daily, it is up to the Open Source community to reverse engineer and adapt the instrumentation for every new version. This is not sustainable. Now is the time when library and framework maintainers have to start adding OpenTelemetry to their code. If this doesn't happen, the production use-case for OpenTelemetry will stay rather limited."

"The maturity of documentation, specification, libraries, and collector varies. One's experience might be very different depending on what they want to achieve," Marcin "Perk" Stożek, Software Engineering Manager of Open Source Collection, Sumo Logic, adds.

Bugs

"OpenTelemetry is developing at a fast rate, and the instrumentation is rapidly changing. This can lead to frustrating bugs at times. But the community is actively taking steps to address concerns around instrumentation stability," says Pranay Prateek, Co-Founder of SigNoz.

Does Not Provide Backend Storage, Analysis or Visualization

Another important challenge to be aware of: OpenTelemetry does not provide any backend storage, analysis or visualization, so to gain full value of the project you need to implement these components on your own or with the help of a service provider.

"Without the proper tools and integrations, it can be challenging to make sense of what the data OpenTelemetry uncovers, and what was meant to provide visibility into IT performance and availability could actually end up creating more data noise instead," says Joe Byrne, VP of Technology Strategy and Executive CTO at Cisco AppDynamics.

Requires Large-Scale Initiative

To truly adopt OpenTelemetry requires a large-scale effort by an organization.

"Users will have to replace some of their existing toolchains of telemetry collection (especially for logs and metrics)," Haberman of Aspecto points out. "It will require quite a lot of effort, and usually, companies are not excited to make this large-scale shift."

"Migrating from current proprietary collection technologies to OpenTelemtry is non trivial for any large customer," adds Nitin Navare, CTO of LogicMonitor.

Difficult to Manage at Scale

"As OpenTelemetry evolves, it will become more complex and challenging to configure and manage at scale," warns Jonah Kowall, CTO of Logz.io.

Alois Reitbauer, Chief Product Officer at Dynatrace, agrees: "The challenge will be managing large-scale OpenTelemetry rollouts and monitoring their health."

Hard to Learn

"OpenTelemetry can be challenging for new developers to learn, as documentation gaps still exist due to the rapid pace of development," says Parker of Lightstep.

"Implementing OpenTelemetry across every part of the system requires deep knowledge and has a high entry point effort," adds Haberman of Aspecto. "This forces users to fully understand how OTel works and get involved in the project's updates. Though, as the project matures and the amount and quality of resources grows, adoption will get easier."

Martin Thwaites, Developer Advocate at Honeycomb, explains further: "At first glance, OpenTelemetry can be challenging to get started with, especially for certain languages that don't provide documentation for orchestration. However, even with this, early adopters were more than willing to dig deep and make it work. Therefore as focus on documentation becomes a priority, this barrier will quickly be eliminated. And, as adoption grows into the early majority and beyond, this will be more important to new users looking to come on board."

"Furthermore, looking deeper at the various language and framework SDKs are doing, it becomes harder to understand," he continues. "Providing more 'easy mode' integrations like the Agents and Kubernetes Operators will be essential to broader adoption. This will ease the issue of sampling or managing high data volume."

Developer Priorities

"It's important to consider that observability is not the fundamental goal when developing new software," explains Sajai Krishnan, General Manager, Observability, Elastic. "A software developer's primary goal may be to make the application or library they are building meet key business requirements or reduce the risk of impacting the performance of their code. Implementing observability may not be a primary goal, as developers carry on with familiar logging as has been done for decades, which could hurt the broad adoption of OpenTelemetry."

Download the 2022 Gartner Magic Quadrant for APM and Observability

Lack of Commercial Support

As with any open source solution, tech support and upgrades could be an issue for users of OpenTelemetry because it is not backed by a commercial vendor.

Vendor Enhancements

"There is also the risk that the standardization and vendor neutrality benefits of OpenTelemetry are lost by vendor enhancements beyond the standard features in their downstream distribution," says Krishnan at Elastic.

Go to: A Guide to OpenTelemetry — Part 6: OTel and APM

Pete Goldin is Editor and Publisher of APMdigest

Hot Topics

The Latest

I've spent a lot of time in the channel, and one thing I keep coming back to is this: a partner program is only as good as what it looks like in the field. Many programs look great on paper, but when a partner is in front of a customer navigating a complex hybrid environment or trying to make the case for AI-powered observability, the gap between what a vendor promises and what it actually delivers becomes very clear, very fast ...

Enterprises today operate in a real-time environment where uninterrupted access to trusted data has become a baseline expectation for users, applications and automated systems. Traditional DataOps models, built on manual effort and human triage, cannot keep pace with this always active demand. AI agents are emerging as the operational backbone, ensuring consistent data availability, reinforcing trustworthiness and enabling a level of scale that manual processes cannot achieve ...

For decades, trust in the digital workplace rested on familiar signals. We trusted faces on video calls, voices on the phone, and emails that appeared to come from people we knew. These cues felt human and intuitive. They anchored how decisions were made, approvals were granted, and access was authorized. AI-powered deepfakes have quietly broken that model ...

Cloud migration was supposed to be a one-way door. For most enterprises, it turns out it isn't. Cloud data repatriation is a real and growing trend. A new survey ... finds that 89% of organizations plan to expand their on-premises infrastructure footprint over the next two years — and 75% have already moved at least some workloads back from public cloud in the past 24 months. The findings point to a broad rethinking of where data belongs ...

Over the past few years, large language models (LLMs) have revolutionized the software industry. Given their ability to excel at multi-step reasoning, LLMs have helped enterprises streamline workflows and adapt to the unknown. However, employing such models comes with sky-high costs, latency issues, and limited flexibility. In the realm of IT operations, it is generally wiser to employ smaller, domain-specific models instead ...

For years, DevOps teams operated under a simple assumption: collect enough telemetry, and you can find and fix any problem. That assumption is breaking down. Modern enterprises now operate across microservices, hybrid cloud environments, APIs, Kubernetes, and highly automated delivery pipelines. Releases happen continuously, dependencies shift constantly, and failures spread faster than teams can diagnose them ...

New Relic surveyed IT and engineering leaders from the media and entertainment (M&E) sector to understand what's working — and where challenges persist with their observability practices. The findings reveal how M&E organizations are navigating rising platform complexity, audience expectations, and AI-driven change. Below are five takeaways that stand out ...

Let me start with something I've seen play out more times than I can count. A team hits a wall with the cloud. Costs creep up, then spike. Performance starts to feel inconsistent. Someone in finance asks a simple question like "why did this double?" and nobody has a clean answer ... Maybe this isn't the right place for everything. That realization feels like a breakthrough, like you've identified the problem. In reality, you've just identified the starting line ...

In MEAN TIME TO INSIGHT Episode 24, Shamus McGillicuddy, VP of Research, Network Infrastructure and Operations, at EMA discusses network observability tool sprawl ... 

In cloud-native systems, scaling is often as simple as moving a slider. For on-premise databases, the stakes are different. Over-provisioning hardware is expensive. Under-provisioning leads to performance bottlenecks that are difficult to fix once the equipment is in the rack ...

A Guide to OpenTelemetry - Part 5: The Challenges

Pete Goldin
APMdigest

While OpenTelemetry offers many advantages, the experts point out several challenges as well.

Start with: A Guide to OpenTelemetry - Part 1

Start with: A Guide to OpenTelemetry - Part 2: When Will OTel Be Ready?

Start with: A Guide to OpenTelemetry - Part 3: The Advantages

Start with: A Guide to OpenTelemetry - Part 4: The Results

The Project is Not Mature

Maybe the greatest challenge for OpenTelemetry is that the project is not mature. While the tracing component is fairly well advanced, the metrics and logging parts are still being formed.

"Currently, the project is not mature enough to support every stack, language, and signal," says Michael Haberman, CTO and Co-Founder of Aspecto. "While we believe it'll get there, the road to full stability is long."

"OpenTelemetry remains a young project in many ways, and many components are still in alpha or beta," explains Austin Parker, Head of Developer Relations at Lightstep by ServiceNow. "There is still work being done to bring in new signals such as profiling, logging, or real user monitoring (RUM), and breaking changes in those signals can be frequent."

"Logs and metrics are slowly catching up, but the API is still unstable which can be an issue," adds Vladimir Mihailenco, Co-Founder of Uptrace. "Logs and metrics are not stable yet so using them is more bumpy and requires some involvement with OpenTelemetry development and reading various changelogs."

Varying Quality

"While the existing OpenTelemetry libraries are already viable for manual and automatic instrumentation, which is a key part of any observability solution, it varies in quality," says Daniel Khan, Director of Product Management (Telemetry) at Sentry. "With new versions of libraries being released almost daily, it is up to the Open Source community to reverse engineer and adapt the instrumentation for every new version. This is not sustainable. Now is the time when library and framework maintainers have to start adding OpenTelemetry to their code. If this doesn't happen, the production use-case for OpenTelemetry will stay rather limited."

"The maturity of documentation, specification, libraries, and collector varies. One's experience might be very different depending on what they want to achieve," Marcin "Perk" Stożek, Software Engineering Manager of Open Source Collection, Sumo Logic, adds.

Bugs

"OpenTelemetry is developing at a fast rate, and the instrumentation is rapidly changing. This can lead to frustrating bugs at times. But the community is actively taking steps to address concerns around instrumentation stability," says Pranay Prateek, Co-Founder of SigNoz.

Does Not Provide Backend Storage, Analysis or Visualization

Another important challenge to be aware of: OpenTelemetry does not provide any backend storage, analysis or visualization, so to gain full value of the project you need to implement these components on your own or with the help of a service provider.

"Without the proper tools and integrations, it can be challenging to make sense of what the data OpenTelemetry uncovers, and what was meant to provide visibility into IT performance and availability could actually end up creating more data noise instead," says Joe Byrne, VP of Technology Strategy and Executive CTO at Cisco AppDynamics.

Requires Large-Scale Initiative

To truly adopt OpenTelemetry requires a large-scale effort by an organization.

"Users will have to replace some of their existing toolchains of telemetry collection (especially for logs and metrics)," Haberman of Aspecto points out. "It will require quite a lot of effort, and usually, companies are not excited to make this large-scale shift."

"Migrating from current proprietary collection technologies to OpenTelemtry is non trivial for any large customer," adds Nitin Navare, CTO of LogicMonitor.

Difficult to Manage at Scale

"As OpenTelemetry evolves, it will become more complex and challenging to configure and manage at scale," warns Jonah Kowall, CTO of Logz.io.

Alois Reitbauer, Chief Product Officer at Dynatrace, agrees: "The challenge will be managing large-scale OpenTelemetry rollouts and monitoring their health."

Hard to Learn

"OpenTelemetry can be challenging for new developers to learn, as documentation gaps still exist due to the rapid pace of development," says Parker of Lightstep.

"Implementing OpenTelemetry across every part of the system requires deep knowledge and has a high entry point effort," adds Haberman of Aspecto. "This forces users to fully understand how OTel works and get involved in the project's updates. Though, as the project matures and the amount and quality of resources grows, adoption will get easier."

Martin Thwaites, Developer Advocate at Honeycomb, explains further: "At first glance, OpenTelemetry can be challenging to get started with, especially for certain languages that don't provide documentation for orchestration. However, even with this, early adopters were more than willing to dig deep and make it work. Therefore as focus on documentation becomes a priority, this barrier will quickly be eliminated. And, as adoption grows into the early majority and beyond, this will be more important to new users looking to come on board."

"Furthermore, looking deeper at the various language and framework SDKs are doing, it becomes harder to understand," he continues. "Providing more 'easy mode' integrations like the Agents and Kubernetes Operators will be essential to broader adoption. This will ease the issue of sampling or managing high data volume."

Developer Priorities

"It's important to consider that observability is not the fundamental goal when developing new software," explains Sajai Krishnan, General Manager, Observability, Elastic. "A software developer's primary goal may be to make the application or library they are building meet key business requirements or reduce the risk of impacting the performance of their code. Implementing observability may not be a primary goal, as developers carry on with familiar logging as has been done for decades, which could hurt the broad adoption of OpenTelemetry."

Download the 2022 Gartner Magic Quadrant for APM and Observability

Lack of Commercial Support

As with any open source solution, tech support and upgrades could be an issue for users of OpenTelemetry because it is not backed by a commercial vendor.

Vendor Enhancements

"There is also the risk that the standardization and vendor neutrality benefits of OpenTelemetry are lost by vendor enhancements beyond the standard features in their downstream distribution," says Krishnan at Elastic.

Go to: A Guide to OpenTelemetry — Part 6: OTel and APM

Pete Goldin is Editor and Publisher of APMdigest

Hot Topics

The Latest

I've spent a lot of time in the channel, and one thing I keep coming back to is this: a partner program is only as good as what it looks like in the field. Many programs look great on paper, but when a partner is in front of a customer navigating a complex hybrid environment or trying to make the case for AI-powered observability, the gap between what a vendor promises and what it actually delivers becomes very clear, very fast ...

Enterprises today operate in a real-time environment where uninterrupted access to trusted data has become a baseline expectation for users, applications and automated systems. Traditional DataOps models, built on manual effort and human triage, cannot keep pace with this always active demand. AI agents are emerging as the operational backbone, ensuring consistent data availability, reinforcing trustworthiness and enabling a level of scale that manual processes cannot achieve ...

For decades, trust in the digital workplace rested on familiar signals. We trusted faces on video calls, voices on the phone, and emails that appeared to come from people we knew. These cues felt human and intuitive. They anchored how decisions were made, approvals were granted, and access was authorized. AI-powered deepfakes have quietly broken that model ...

Cloud migration was supposed to be a one-way door. For most enterprises, it turns out it isn't. Cloud data repatriation is a real and growing trend. A new survey ... finds that 89% of organizations plan to expand their on-premises infrastructure footprint over the next two years — and 75% have already moved at least some workloads back from public cloud in the past 24 months. The findings point to a broad rethinking of where data belongs ...

Over the past few years, large language models (LLMs) have revolutionized the software industry. Given their ability to excel at multi-step reasoning, LLMs have helped enterprises streamline workflows and adapt to the unknown. However, employing such models comes with sky-high costs, latency issues, and limited flexibility. In the realm of IT operations, it is generally wiser to employ smaller, domain-specific models instead ...

For years, DevOps teams operated under a simple assumption: collect enough telemetry, and you can find and fix any problem. That assumption is breaking down. Modern enterprises now operate across microservices, hybrid cloud environments, APIs, Kubernetes, and highly automated delivery pipelines. Releases happen continuously, dependencies shift constantly, and failures spread faster than teams can diagnose them ...

New Relic surveyed IT and engineering leaders from the media and entertainment (M&E) sector to understand what's working — and where challenges persist with their observability practices. The findings reveal how M&E organizations are navigating rising platform complexity, audience expectations, and AI-driven change. Below are five takeaways that stand out ...

Let me start with something I've seen play out more times than I can count. A team hits a wall with the cloud. Costs creep up, then spike. Performance starts to feel inconsistent. Someone in finance asks a simple question like "why did this double?" and nobody has a clean answer ... Maybe this isn't the right place for everything. That realization feels like a breakthrough, like you've identified the problem. In reality, you've just identified the starting line ...

In MEAN TIME TO INSIGHT Episode 24, Shamus McGillicuddy, VP of Research, Network Infrastructure and Operations, at EMA discusses network observability tool sprawl ... 

In cloud-native systems, scaling is often as simple as moving a slider. For on-premise databases, the stakes are different. Over-provisioning hardware is expensive. Under-provisioning leads to performance bottlenecks that are difficult to fix once the equipment is in the rack ...