Is Baselime now still the way to go? Is it pure worker logs and open telemetry / traces will follow?

Is Baselime now still the way to go? Is it pure worker logs and open telemetry / traces will follow? Slightly confused by the current state

SSly Is Baselime now still the way to go? Is it pure worker logs and open telemetry /...

James•1/4/25, 7:16 PM

Probably not. There’s lots of outstanding issues and questions with baselime that have yet to be answered or resolved - it’s essentially abandoned as far as we can tell.

Workers observability is slowly getting up to feature parity but it’s not there yet, and doesn’t support anything external to Cloudflare.

JJames Probably not. There’s lots of outstanding issues and questions with baselime tha...

SlyOP•1/4/25, 7:28 PM

Not worried about not having something outside of cf, but afaik it doesn’t have traces and on par dx yet, does it?

SSly Not worried about not having something outside of cf, but afaik it doesn’t have ...

James•1/4/25, 7:36 PM

It does not, no

BlackRain•1/4/25, 9:06 PM

Anyone know how to grant read only access to just the Worker Logs section?

sharqwy•1/6/25, 2:00 PM

I'm having trouble integrating Sentry with my Worker. Basically I connected Sentry as instructed but no seeing any of my projects even they're already out there in my Sentry dash.

Walshy•1/7/25, 1:49 AM

I'm not too sure what more info you'd expect

Walshy•1/7/25, 1:50 AM

This error was thrown by discord.js from Discord, it's saying it can't find the webhook

Michał Wróblewski•1/7/25, 11:34 PM

are there any plans on adding alerts/notifications for Logs or Metrics? I would really like to receive alert when my worker throws an error

SSly Is Baselime now still the way to go? Is it pure worker logs and open telemetry /...

TKR👑•1/14/25, 12:21 PM

We basically have no visibility when using Baselime for last few months... so hoping for some clarity from CF on this

SSly Is Baselime now still the way to go? Is it pure worker logs and open telemetry /...

rohin•1/15/25, 5:50 AM

Our efforts are focused on native observability within the Cloudflare Developer Platform. We'll be rolling out an improved Workers Metrics dashboard and new features for Workers Logs soon. We have also started working on longer-term projects -- including distributed traces. I don't have any dates to share just yet, but I will update this channel once I do.

DDeclan Just want to drop a BIG THANK YOU to @Walshy | Workers/Pages and @Erwin for Obs...

Erwin•1/16/25, 1:25 AM

Thanks for the shout out. Means a lot. I never quite have enough time to really get the library to where I want it to be.

But it good to be reminded every now and then that is already really useful for a lot of people

SlyOP•1/16/25, 11:22 AM

I appreciate that one, but that I can’t use it in conjunction with testing via the suggested vitest approach is a bummer

sam•1/16/25, 3:19 PM

is this the best channel to ask specifics about behaviour we're seeing on our tail workers? or should that be a #workers-help forum post or something else?

cheers

Ssam is this the best channel to ask specifics about behaviour we're seeing on our ta...

rohin•1/16/25, 5:38 PM

this channel works!

Mozzy•1/21/25, 3:17 PM

Something broke with the time selection. It registers mouse clicks as a hold

MMozzy Something broke with the time selection. It registers mouse clicks as a hold

rohin•1/21/25, 5:32 PM

thanks for the report @Mozzy! we're on it

Mozzy•1/21/25, 5:57 PM

cheers

Mozzy•1/23/25, 2:26 PM

Are notifications/emails/triggers on error thresholds currently planned for the future? Or real-time error alerts as mentioned before

matt•1/25/25, 10:02 PM

will otel traces be viewable in the cloudflare dashboard at some stage? currently i'm using baselime (was using honeycomb) - but also looking at other options if anyone has recommendations

matt•1/26/25, 1:05 AM

in baselime how do you customize how spans appear? some have cloudflare/http/nodejs icons etc

matt•1/26/25, 1:06 AM

ah i think it's because im using the newer semconvention attrs, need to go back to the old ones ie.

http.request.method

http.request.method

http.method

http.method

Original message was deleted

Thomas Ankcorn•1/27/25, 1:23 PM

Currently there is a limit of 128kb of data limit for each workers invocation.

The sample rate works on an invocation by invocation basis.

i.e. a sample rate of 30% means you will get all the logs for 30% of your invocations and 0% of the logs from the rest.

To fix this you want to look at consolidating your logging for each invocation and make sure you are not logging unecessary data.

Hope this helps

Mmatt will otel traces be viewable in the cloudflare dashboard at some stage? currentl...

rohin•1/27/25, 6:52 PM

do you want to use cloudflare as an observability provider, similar to baselime? ie send traces from third parties and render it in the cf dashboard?

MMozzy Are notifications/emails/triggers on error thresholds currently planned for the ...

rohin•1/27/25, 6:53 PM

we're going to start with real-time errors, first

Rrohin do you want to use cloudflare as an observability provider, similar to baselime?...

matt•1/27/25, 7:05 PM

It would be nice to have everything in one place, I thought with the addition of worker logs that CF might be aiming to support the other 2 OTEL signals (metrics and trace spans)

Rrohin do you want to use cloudflare as an observability provider, similar to baselime?...

kchro3•1/27/25, 10:09 PM

... i thought that the goal was for cf to support alerting by early this year?

otherwise what was the point of buying & phasing out baselime?

Mmatt It would be nice to have everything in one place, I thought with the addition of...

rohin•1/28/25, 5:47 AM

got it! yep, we're actively working on this

�

😈 Donkey 💫•1/28/25, 5:55 AM

I can't access Worker logs in past 12hours

dhkl•1/28/25, 6:19 AM

Same here

dhkl•1/28/25, 6:19 AM

dhkl•1/28/25, 6:19 AM

Seems to affect newly deployed workers only

dhkl•1/28/25, 6:20 AM

Wonder if we're going to lose logs over this, or if it's just a display issue

Walshy•1/28/25, 7:16 AM

https://www.cloudflarestatus.com/incidents/jwxdzrmbgyn9

Workers Logs dashboard availability

WWalshy https://www.cloudflarestatus.com/incidents/jwxdzrmbgyn9

dhkl•1/28/25, 9:50 AM

It's back up. Thanks!

�

😈 Donkey 💫•1/28/25, 10:09 AM

Thanks.
Btw, It hangs the browser tab, sometimes when I expand a log. You guys acknowledged?

Ddhkl Wonder if we're going to lose logs over this, or if it's just a display issue

Maksym•1/28/25, 11:12 AM

it's just display issue, the actual data is intact

Aster•1/31/25, 10:00 AM

Hello!

I have a question about tail workers.
We want to balance what is best between good DX and easy integration.
Currently if I understand correctly, each worker should have its own tail worker?
Because we are thinking to create 1 tail worker handling all our workers. For logs it's fine, but for sentry it makes it very complicated as this one tail worker should have knowledge about all the sentry DSNs...

What would be the recommended way? Note: we are using a monorepo

Bboristane your assessment is correct. I've used a single tail worker attached to multiple ...

Aster•1/31/25, 2:54 PM

It feels weird to have the tail worker handling all the SENTRY_DNS...
So would you say it is a better practice to instead have 1 tail worker per worker?

Aster•1/31/25, 2:54 PM

That's at least what seems to suggest the documentation

James•1/31/25, 9:44 PM

feature request: it'd be nice to be able to get a breakdown quickly in the overview of number of obs events per worker to see which ones are doing the most https://i.james.pub/file/2025/01/df95b736-f971-4d8a-bee6-4680466f1b9a.png

James•1/31/25, 9:44 PM

even if I click into a worker, I don't see a clear and easy way to get counts over a period of time without adding up all the events in the graph myself manually

ssandler•2/3/25, 6:22 PM

Any advice for using the https://github.com/evanderkoogh/otel-cf-workers with Durable Objects? The examples given don't seem to work with the current SDK, because they don't

extend DurableObject

extend DurableObject

but rather implement it as an interface. Trying to run that yields a runtime error

The receiving Durable Object does not support RPC, because its class was not declared with `extends DurableObject`

The receiving Durable Object does not support RPC, because its class was not declared with `extends DurableObject`

Calling

instrumentDO()

instrumentDO()

on a class that

extends DurableObject<Env>

extends DurableObject<Env>

results in a different error:

Failed to construct 'DurableObjectBase': constructor parameter 1 is not of type 'DurableObjectState'

Failed to construct 'DurableObjectBase': constructor parameter 1 is not of type 'DurableObjectState'

PixelPulse•2/12/25, 2:19 PM

I was wondering if I could get some assistance on our baselime account which appears to be acting very buggy.
We saw some strange behaviour in the dashboard when our admin tried to give admin status to another team member.

This removed admin status from everyone and now we are all “contributors”. I’m not sure if these problems are connected but we also started seeing that our query pages were failing. When we would try going into a query it would flash something useful and then immediately go to an empty screen. Essentially we would see some UI and then it would almost immediately go blank.
Our dashboard right now has gone completely empty - we have zero environments and no logs displaying