The search functionality has gotten really buggy recently.

16 Replies

daveOP•2mo ago

Anyone else notice this? I can't seem to switch to my local timezone anymore. Always stuck in UTC. I'm in America/Cayman, not sure if that makes a difference.

MrBithles•5w ago

Does anyone know how to setup sentry with for workers with WorkerEntrypoint class (RPC). I see theres a SDK for workers that wraps the exported function calls for fetch. But that doesn't work for class based workers as the class is initialised by CF.

Thomas Ankcorn•4w ago

are you using this? https://docs.sentry.io/platforms/javascript/guides/cloudflare/

Cloudflare | Sentry for Cloudflare

Official Sentry SDK for Cloudflare Workers and Cloudflare Pages.

Unknown User•4w ago

Message Not Public

daveOP•4w ago

Correct.

Unknown User•4w ago

Message Not Public

PixelPulse•4w ago

Hi there, I was wondering if I could get assistance with my organisations Baslime account? We were successfully using the service a few months ago but around the time we were trying to create another admin user in the dashboard we now have an issue where it looks like our environment has been wiped clean and no one is actually an admin anymore. The fact that no one in the org is an admin is making it hard to explore options for setting up necessary configuration to get our logs back up and working in baselime. We would appreciate assistance to help us get back to recieving value from our Baselime account. Thank you

😈 Donkey 💫•4w ago

1. The default mode is the Invocations tab, which is really annoying but acceptable. 2. Even though there are a lot of logs inside the request, the default visible line is still empty, which is very, very annoying. This makes the dashboard mostly useless to me. I hope you guys can revert the default visible line to what it was in the previous version.

Walshy•4w ago

1. I'd be interested to know why you don't prefer invocations as the default 2. I'm not sure what you mean by

the default visible line is still empty,

I'm guessing since you have no message you're printing an object or something as your log? can you expand what your logs are exactly

😈 Donkey 💫•4w ago

I'm with CF worker for a long time so I believe consoling log in object format is recommended

😈 Donkey 💫•4w ago

I don't prefer using invocation as the default because I customize my own data logs, instead relying on default logs. Even I disable the invocation logs for major internal workers since it's useless in that case.

Thomas Ankcorn•4w ago

The message being empty is a bug, thanks for flagging. I will get it fixed But I would recommend changing _message to message so it gets picked up as the message field 🙂

😈 Donkey 💫•4w ago

The message field is default for the String logs, that's why I want to avoid 🙂 Thanks for recommended

mr.niko.la•4w ago

      } catch (groqError) {
        console.log(`Groq API failed: ${groqError.message}. Attempting Workers AI failover.`);
        
        try {
          // Workers AI Whisper Failover
          console.log(`Using Workers AI failover for ${audioFile}`);
          const aiResponse = await this.env.AI.run('@cf/openai/whisper-large-v3-turbo', {
            audio: [...new Uint8Array(audioData)],
            language: 'en'
          });
          
          transcriptionResponse = { text: aiResponse.text };
          usedFailover = true;
          failoverType = 'workers-ai';
          
        } catch (workersAiError) {
          // If Workers AI also fails, let the existing EC2 fallback handle it
          console.log(`Workers AI failover failed: ${workersAiError.message}. Falling back to EC2.`);
          throw groqError; // Rethrow original error to trigger EC2 fallback
        }
      }

      } catch (groqError) {
        console.log(`Groq API failed: ${groqError.message}. Attempting Workers AI failover.`);
        
        try {
          // Workers AI Whisper Failover
          console.log(`Using Workers AI failover for ${audioFile}`);
          const aiResponse = await this.env.AI.run('@cf/openai/whisper-large-v3-turbo', {
            audio: [...new Uint8Array(audioData)],
            language: 'en'
          });
          
          transcriptionResponse = { text: aiResponse.text };
          usedFailover = true;
          failoverType = 'workers-ai';
          
        } catch (workersAiError) {
          // If Workers AI also fails, let the existing EC2 fallback handle it
          console.log(`Workers AI failover failed: ${workersAiError.message}. Falling back to EC2.`);
          throw groqError; // Rethrow original error to trigger EC2 fallback
        }
      }

Error processing queue message: Error: Durable Object's isolate exceeded its memory limit and was reset.
2025-04-02 14:29:51:489
UTC
Error processing f576d688-9957-4cfd-816e-a0c75c34b02c_1743604179380.webm: Durable Object's isolate exceeded its memory limit and was reset

Error processing queue message: Error: Durable Object's isolate exceeded its memory limit and was reset.
2025-04-02 14:29:51:489
UTC
Error processing f576d688-9957-4cfd-816e-a0c75c34b02c_1743604179380.webm: Durable Object's isolate exceeded its memory limit and was reset

I sent a 35mb audio file to Workers AI Whisper v3 Turbo. I am on paid plan. Any idea why this didnt work and i hit the DO memeory limit?

Thomas Ankcorn•4w ago

In your code it looks like you are making multiple copies of the audio file. It would be best to raise this with the workers AI & DO team though. Sorry I can't help more

daveOP•4w ago

Looks fixed as of this morning (or earlier, only noticed now).

Gaming

Programming

The search functionality has gotten really buggy recently.

Did you find this page helpful?