Sentry risks for error-tracking teams choosing Datadog
Sentry provides specialized error tracking but lacks the native log aggregation and infrastructure visibility found in Datadog. Reliability risks include historical service interruptions, such as a May 2022 incident where 90% of incoming events were lost.
Sentry functional gaps in full-stack observability
Sentry provides specialized error tracking and application performance monitoring, but teams requiring full-stack observability face gaps in log management and infrastructure visibility. Sentry captures uncaught exceptions and provides detailed reports with stack traces and environment context. The platform links errors to specific code releases and identifies the suspect commit that caused a bug. Sentry 9 also connects issues to specific teams and enables Slack alerts based on context. The company recently announced a $16 million series B round of funding to support its 9,000 paying customers and half a million developers. However, Sentry lacks native log aggregation for troubleshooting complex distributed systems. It provides minimal visibility into host or container infrastructure, which creates difficulties for cloud-native environments. Teams must often integrate Sentry with disparate tools for metrics, logs, and tracing, a process that increases workflow complexity and cost. While Sentry excels at code-level issues, it may not satisfy teams needing full APM with infrastructure monitoring, leading many to use Datadog instead. Sentry captures uncaught exceptions and provides detailed reports with stack traces and environment context, but it lacks the native log aggregation and infrastructure monitoring capabilities that full-stack observability platforms like Datadog provide to modern DevOps teams that manage complex cloud environments.
| Plan / Tier | Monthly Price | Annual Price | Key Included Features |
|---|---|---|---|
| Sentry Starter | $9/mo | $7.2/mo | Core Platform Access, Up to 3 seats |
| Sentry Pro / Business | $22.5/mo | $18/mo | Core Platform Access, Up to 15 seats |
Reliability concerns and event loss
Reliability concerns arise from historical service interruptions and the risk of event loss during provider failures. On May 6, 2022, Sentry experienced a large-scale incident that lasted over 6 hours. During this window, approximately 90% of incoming events disappeared between 3:15 AM and 11:15 AM PDT. The root cause involved an issue within the Google Cloud Platform primary compute region, which affected persistent volumes attached to the infrastructure. This failure made the Sentry UI and API unavailable, preventing customers from accessing existing data or using artifact uploads during build and release processes. The SRE team identified abnormal levels of IO wait and latencies at 1:38 AM PDT. By 3:00 AM, much of the core infrastructure in GCP became heavily degraded or inaccessible. The team discovered that event ingestion struggled to process the typical incoming event load because the geographically distributed ingestion infrastructure, or Points of Presence, failed to operate properly. The team diverted traffic directly to a centralized ingestion path to route around the problem. This process involved DNS propagation that allowed the pipeline to recover slowly. After the engineers restored services, they spent 2 hours on a cleanup phase to verify all core services. The investigation and resolution work lasted 11 hours, with the incident being marked resolved at 12:17 PM PDT. Because this incident affected the ability for applications to reliably write to data volumes, many services required careful verification before returning to full operation. Over the last 5 years, Sentry appeared on outage timelines 9 times. The average resolved outage for Sentry lasted 1 hour and 23 minutes.
Operational concentration and the blast radius
Operational concentration creates risk when a single provider outage disrupts critical workflows. A massive Exchange Online outage on November 12, 2026, reduced servers to quivering wrecks and impacted multiple Microsoft 365 services. This outage showed that putting all organizational eggs into one cloud basket creates a large dependency blast radius. Because Exchange Online connects to other Microsoft 365 services, the failure spread across various tools. About 15 hours before the outage, an automatic switch moved Teams traffic onto the gateway already serving Exchange, changing the volume of requests. Separately, nearly 14% of the gateway capacity was unavailable following an unrelated issue. As demand rose, processes on the remaining servers began waiting on one another. At the worst point, more than 80% of the affected gateway fleet was unresponsive. The UK minister Ian Murray sent a test email that bounced back at 2pm on Friday. This uncertainty affected hospitals, government departments, and financial institutions. For the NHS, staff reported difficulty accessing email and exchanging information across hospitals and GP practices. Financial firms like Aviva had to issue alternative dial-in numbers for earnings calls, while Euronext moved inquiries to the phone. The US House of Representatives also lost email and court systems in Singapore experienced disruption. Microsoft recorded active impact from 16:20 GMT on Thursday 12 November to 16:10 on Friday 13 November. Can an error-tracking team maintain visibility if their monitoring provider suffers a similar concentration-based failure?