Datadog agent error: SyncForwarder.sendHTTPTransactions final attempt: error while sending transaction, rescheduling it

[liadt-neosec]

hi, I am using Datadog Agent to send metrics to datadog in my repo. I'm using this 2:
from datadog_lambda.metric import ( init_lambda_stats, lambda_metric as original_lambda_metric, )

from time to time I get this error:
SyncForwarder.sendHTTPTransactions final attempt: error while sending a transaction, rescheduling it

unfortunately, I don't have the traceback of the error, I see this as just one log in AWS CloudWatch, without any further explanation.

I've also seen a similar issue that is closed but the error does not seem to be gone:
#1546

thanks! :)

[quentinbarbault]

Hi, I too am experiencing this error using the datadog lambda extension.

Please note that it only occurs when my lambda is in a VPC.

I tried setting up my proxy using environment variables DD_HTTPS_PROXY and DD_HTTP_PROXY.
Those variables are displayed at the beginning of my lambda logs (confirming their presence) but it really seems to me like the extension is not taking them into consideration while making the request to datadog. Hence giving me this error.

To make sure my proxy was working correctly, I took the request to datadog : "https://6-0-0-app.agent.datadoghq.eu/api/v1/check_run?api_key=" with my apikey and did a small lambda doing the call, in a VPC, with proxy values set manually. It worked !

It seems I must be using the extension wrong (I'm with CDK) or there is a problem somewhere. Can you help me ?

[theon]

We were also seeing this error. Upgrading to the Datadog-Extension Layer v22 seems to have resolved it for us (early days, but haven't seen the error since).
https://github.com/DataDog/datadog-lambda-extension/releases/tag/v22

We were previously on v21.