7 ms·

Show HN: Uptrace – open-source APM (alternative to Datadog, NewRelic)

85 points by vmihailenco 3y ago

vmihailenco 3y ago
Uptrace is an open-source APM tool that supports distributed tracing, metrics, and logs. You can use it to monitor applications and set up alerts to receive notifications via email, Slack, Telegram, and more. In v1.4 release Uptrace got built-in alerting capabilities that allow to monitor metrics, manage alerts, and receive notifications without using AlertManager. But AlertManager is still supported. While OpenTelemetry remains to be the main source of telemetry data, Uptrace also provides integrations for Prometheus metrics, Vector logs, FluentBit, Sentry, and CloudWatch. You can quickly try Uptrace by: - Running a Docker example: https://github.com/uptrace/uptrace/tree/master/example/docker https://github.com/uptrace/uptrace/tree/master/example/docke... - Checking Uptrace Cloud demo: https://app.uptrace.dev/play https://app.uptrace.dev/play If you have any feedback or questions, let me know in the comments!
- cocoland2 3y ago
  Hi there, Right now Uptrace seems to be supporting only Clickhouse sink only. Is there any future roadmap to support other Exporters supported on OpenTelemetry ?
bognition 3y ago
I love seeing open source alternatives to these major vendors. IMO the reason these vendors can and do charge so much is not because telemetry software is hard. Its because managing data at large volumes is incredibly challenging. No matter what solution you pick, you're going to hit scaling issues eventually. Sure you can start with a single instance, but eventually you're going to need to shard, backup, replicate, etc... If you're running a saas business with uptime expectations from your customers then your telemetry systems need to be rock solid. This means either paying through the nose for a system that is mostly going to work. Or spinning up a team of engineers to active manage and operate your telemetry systems, and then spending years making your system rock solid. Managing these systems is a challenging problem and its not one that most software engineers want to do or will enjoy. So its difficult to find the right people.
- pdimitar 3y ago
  I find myself surprised because I totally would love a job like that but yep, filter bubble and all, guess I just didn't realize mine up until your comment.
  - bognition 3y ago
    We should talk, i'd love to hire you
    - pdimitar 3y ago
      My email is in my profile and I'm currently free, resting. Ping me or tell me where to find you, would love to talk.
    - endigma 3y ago
      Are you actually planning on building this? We should talk if so, this is a super interesting problem domain IMO.
      - pdimitar 3y ago
        And I would still participate!
    - pdimitar 3y ago
      Shame that you haven't replied. Sounded like a super interesting job.
- tootie 3y ago
  Yeah exactly. I rarely if ever run these kinds of platforms for exactly that reason. I've been at small, resource constrained orgs where could not afford the developer hours to keep open source products up to date so we just buy entry-level licenses. And I've worked at huge orgs where we have the people but they have more urgent tasks so handing over $250k a year to a SaaS vendor is still cheaper and less risky than hiring. I still very much appreciate the open source stuff as it at least keeps vendors honest.
just_mc 3y ago
We are using this and we have been very happy so far.
darkwater 3y ago
A few questions if authors are around: - is the metrics query language 100% compatible with PromQL or people will need to learn something new? - can logs / metrics / traces be visualized by some external app like Grafana? - speaking of which, how does it compare to GrafanaLabs solutions (Loki/Tempo/Mimir)? - how "plug and play" would be in a system using Prometheus+Thanos for metrics and Filebeat+Logsatsh+ElasticSearch for logs? EDIT: scratch that, there is enough doc on the website about this Thanks!
m-o11y 3y ago
If you're in the market for a DataDog/NewRelic alternative that's based on OSS, but a bit more "few-clicks-and-you-are-up-and-running", check out OpsVerse ObserveNow: https://opsverse.io/observenow-observability/ https://opsverse.io/observenow-observability/ .. An O11y platform entirely powered by OSS tools, ingestion-driven pricing, and without the hassle of managing the stack and scaling up. We also recently spoke about how we leverage Clickhouse to build out our APM solution on top on Jaeger traces: https://www.youtube.com/watch?v=19smsahx7Xg https://www.youtube.com/watch?v=19smsahx7Xg -- Disclaimer: I work at OpsVerse, and we're big fans of UpTrace (and other OSS O11y tools)
- juicypt 3y ago
  Interesting! I couldn't find a link to github (or something else) on your website though. Is there a way to get the source to do some tests?
ryanrussell 3y ago
Can we please not call this open source if it's not? The tool may be great, but the title leaves me skeptical of anything else. From: https://github.com/uptrace/uptrace/blob/master/LICENSE https://github.com/uptrace/uptrace/blob/master/LICENSE Business Source License 1.1 Parameters Licensor: Uptrace Licensed Work: Uptrace The Licensed Work is (c) 2021 Uptrace Additional Use Grant: You may make use of the Licensed Work, provided that you may not use the Licensed Work for a Tracing and Metrics Service. A "Tracing and Metrics Service" is a commercial offering that allows third parties (other than your employees and contractors) to access the functionality of the Licensed Work so that such third parties directly benefit from the tracing and metrics of the Licensed Work.
- whalesalad 3y ago
  the source is right there in the open. licensing is a different story.
  - yellowapple 3y ago
    "Open-source" has a very specific definition which the BSL does not satisfy.
- devidark 3y ago
  How then we should call it instead, if the source code is openly available?
ttul 3y ago
The reason we outsource monitoring to Datadog is so we don’t have to worry about monitoring while the rest of our revenue-generating infrastructure is having a bad day. My team absolutely could do this in-house, but it would be a distraction day-to-day and potentially a disaster when dealing with a crisis.
bovermyer 3y ago
If it doesn't have profiling, it's not an alternative to Datadog or New Relic.
dewlinedew2 3y ago
Should have called it updog

85 points by vmihailenco 3y ago

Uptrace is an open-source APM tool that supports distributed tracing, metrics, and logs. You can use it to monitor applications and set up alerts to receive notifications via email, Slack, Telegram, and more. In v1.4 release Uptrace got built-in alerting capabilities that allow to monitor metrics, manage alerts, and receive notifications without using AlertManager. But AlertManager is still supported. While OpenTelemetry remains to be the main source of telemetry data, Uptrace also provides integrations for Prometheus metrics, Vector logs, FluentBit, Sentry, and CloudWatch. You can quickly try Uptrace by: - Running a Docker example: https://github.com/uptrace/uptrace/tree/master/example/docker https://github.com/uptrace/uptrace/tree/master/example/docke... - Checking Uptrace Cloud demo: https://app.uptrace.dev/play https://app.uptrace.dev/play If you have any feedback or questions, let me know in the comments!

cocoland2 3y ago
Hi there, Right now Uptrace seems to be supporting only Clickhouse sink only. Is there any future roadmap to support other Exporters supported on OpenTelemetry ?

bognition 3y ago

I love seeing open source alternatives to these major vendors. IMO the reason these vendors can and do charge so much is not because telemetry software is hard. Its because managing data at large volumes is incredibly challenging. No matter what solution you pick, you're going to hit scaling issues eventually. Sure you can start with a single instance, but eventually you're going to need to shard, backup, replicate, etc... If you're running a saas business with uptime expectations from your customers then your telemetry systems need to be rock solid. This means either paying through the nose for a system that is mostly going to work. Or spinning up a team of engineers to active manage and operate your telemetry systems, and then spending years making your system rock solid. Managing these systems is a challenging problem and its not one that most software engineers want to do or will enjoy. So its difficult to find the right people.

pdimitar 3y ago
I find myself surprised because I totally would love a job like that but yep, filter bubble and all, guess I just didn't realize mine up until your comment.
- bognition 3y ago
  We should talk, i'd love to hire you
  - pdimitar 3y ago
    My email is in my profile and I'm currently free, resting. Ping me or tell me where to find you, would love to talk.
  - endigma 3y ago
    Are you actually planning on building this? We should talk if so, this is a super interesting problem domain IMO.
    - pdimitar 3y ago
      And I would still participate!
  - pdimitar 3y ago
    Shame that you haven't replied. Sounded like a super interesting job.
tootie 3y ago
Yeah exactly. I rarely if ever run these kinds of platforms for exactly that reason. I've been at small, resource constrained orgs where could not afford the developer hours to keep open source products up to date so we just buy entry-level licenses. And I've worked at huge orgs where we have the people but they have more urgent tasks so handing over $250k a year to a SaaS vendor is still cheaper and less risky than hiring. I still very much appreciate the open source stuff as it at least keeps vendors honest.

just_mc 3y ago

We are using this and we have been very happy so far.

darkwater 3y ago

A few questions if authors are around: - is the metrics query language 100% compatible with PromQL or people will need to learn something new? - can logs / metrics / traces be visualized by some external app like Grafana? - speaking of which, how does it compare to GrafanaLabs solutions (Loki/Tempo/Mimir)? - how "plug and play" would be in a system using Prometheus+Thanos for metrics and Filebeat+Logsatsh+ElasticSearch for logs? EDIT: scratch that, there is enough doc on the website about this Thanks!

m-o11y 3y ago

If you're in the market for a DataDog/NewRelic alternative that's based on OSS, but a bit more "few-clicks-and-you-are-up-and-running", check out OpsVerse ObserveNow: https://opsverse.io/observenow-observability/ https://opsverse.io/observenow-observability/ .. An O11y platform entirely powered by OSS tools, ingestion-driven pricing, and without the hassle of managing the stack and scaling up. We also recently spoke about how we leverage Clickhouse to build out our APM solution on top on Jaeger traces: https://www.youtube.com/watch?v=19smsahx7Xg https://www.youtube.com/watch?v=19smsahx7Xg -- Disclaimer: I work at OpsVerse, and we're big fans of UpTrace (and other OSS O11y tools)

juicypt 3y ago
Interesting! I couldn't find a link to github (or something else) on your website though. Is there a way to get the source to do some tests?

ryanrussell 3y ago

Can we please not call this open source if it's not? The tool may be great, but the title leaves me skeptical of anything else. From: https://github.com/uptrace/uptrace/blob/master/LICENSE https://github.com/uptrace/uptrace/blob/master/LICENSE Business Source License 1.1 Parameters Licensor: Uptrace Licensed Work: Uptrace The Licensed Work is (c) 2021 Uptrace Additional Use Grant: You may make use of the Licensed Work, provided that you may not use the Licensed Work for a Tracing and Metrics Service. A "Tracing and Metrics Service" is a commercial offering that allows third parties (other than your employees and contractors) to access the functionality of the Licensed Work so that such third parties directly benefit from the tracing and metrics of the Licensed Work.

whalesalad 3y ago
the source is right there in the open. licensing is a different story.
- yellowapple 3y ago
  "Open-source" has a very specific definition which the BSL does not satisfy.
devidark 3y ago
How then we should call it instead, if the source code is openly available?

ttul 3y ago

The reason we outsource monitoring to Datadog is so we don’t have to worry about monitoring while the rest of our revenue-generating infrastructure is having a bad day. My team absolutely could do this in-house, but it would be a distraction day-to-day and potentially a disaster when dealing with a crisis.

bovermyer 3y ago

If it doesn't have profiling, it's not an alternative to Datadog or New Relic.

dewlinedew2 3y ago

Should have called it updog