Questions tagged with AWS X-Ray

I require support for preserving traces from different teams in Xray with the following attributes. The XRay backend used to store all traces will be provisioned in a single centralized account. Note that applications are integrated with OpenTelemetry libraries and will be submitting traces to XRay using ADOT collector. * maintain separate keys, one per team, for encryption of trace data at rest * separate IAM roles to provision access of keys and trace data such a team cannot access traces from other teams. * maintain separate configurations for sampling rates, rate limits, quotas, etc. A feasible option for implementing these requirements seems to be using tracing groups. Groups can be created using filter expressions based on certain pre-defined attributes in a trace such as service prefix, unique per team. However, XRay lacks configuring multiple keys for encrypting traces and IAM policies to control read access to traces and service graphs don't support resource level permissions or condition filters. There is also a limit of 25 groups per account. Due to these limitations, we cannot use groups. Are there alternate approaches for implementing my aforementioned requirements?

I have use-case where the Producer Java lambda publishes file messages to SQS and Consumer Nodejs consumes them. Producer side of X-ray is enabled in serverless.yml; displays a Service map with all AWS components with a new Trace Id. The Consumer side of Nodejs lambda; is enabled with another Trace Id. I see a single service map, but with two clients. I guess it is because of the 2 separate Trace Ids. Is it possible to make single Trace Id by propagating it end to end?

The idea is the following: * Codecommit has a notification rule. * Notification's target it's an sns topic. * The topic's target is a lambda function. * Minimally I want lambda X-Ray trace when the notification occurs. If all trace better. I post a message from the SNS Topic console, the X-Ray trace is generated with SNS and Lambda. However, if I trigger a notification, the trace is not generated (Lambda runs correctly). I would like to know if I can make the complete or partial trace be generated when generating a notification in CodeCommit. # Update My lambda logs insights: * Post message from SNS Topic:```"_X_AMZN_TRACE_ID": "Root=1-62gd6bfa-6ag9233f49fe8faf6ef7ed16;Parent=4b19513e742ad199;Sampled=1"``` * Codecommit notification: ```"_X_AMZN_TRACE_ID": "Root=1-62d56c36-3bg3f6fbaa075b171b9fdba6;Parent=2f5f68ffyf25388e;Sampled=0"``` My lambda Function has active tracing, and has permissions, but I don't know why the second case don't get sampled. It should be tracing all the invocations, being sampled or not sampled...

I'm trying to use Spring Cloud Sleuth with AWS X-Ray. When I am testing my API locally I can see the X-Amzn-Trace-Id header being picked up correctly by Spring Cloud Sleuth and being attached to my logs correctly. The issue is I cannot see any requests being sent from my application to the xray-daemon which is running as a container locally. Has anyone successfully got this working with the latest versions of spring cloud sleuth and zipkin-reporter-xray-udp to forward the requests to the xray-daemon? I have only been able to find guides/resources from 2019 but these are no longer relevant. I have configured the following bean and using spring-cloud-starter-sleuth 3.1.3 and zipkin-reporter-xray-udp 0.23.4. ``` @Bean(ZipkinAutoConfiguration.REPORTER_BEAN_NAME) Reporter<Span> myReporter() { return XRayUDPReporter.create(); } ``` I am running the X-Ray daemon using the following command ``` docker run -d -v ~/.aws/:/root/.aws/:ro -e AWS_REGION=us-west-2 --name xray-daemon -p 2000:2000/udp public.ecr.aws/xray/aws-xray-daemon -o ```

Hi all I'd like to set up AWS Open Distro for OpenTelemetry for Java with Elasic Beanstalk (https://aws-otel.github.io/docs/getting-started/java-sdk/trace-auto-instr). I've added the Java agent (using `java -javaagent:path/to/aws-opentelemetry-agent.jar -jar myapp.jar`). Starting the application, it logs a line like this every 15 seconds: ``` [otel.javaagent 2022-06-08 19:21:03:940 +0000] [OkHttp http://localhost:4317/...] ERROR io.opentelemetry.exporter.internal.grpc.OkHttpGrpcExporter - Failed to export spans. The request could not be executed. Full error message: Failed to connect to localhost/127.0.0.1:4317 ``` So I guess the ADOT Collector (https://aws-otel.github.io/docs/getting-started/collector) isn't running. When I log into my instances (platform: Corretto 11 running on 64bit Amazon Linux 2/3.2.15), I can see the CloudWatch Agent under `/opt/aws/amazon-cloudwatch-agent` (version: 1.247350.0). It includes the Collector under `/opt/aws/amazon-cloudwatch-agent/cwagent-otel-collector`: ``` $ pwd /opt/aws/amazon-cloudwatch-agent $ ls LICENSE NOTICE RELEASE_NOTES THIRD-PARTY-LICENSES bin cwagent-otel-collector doc etc logs var $ ls cwagent-otel-collector/ etc logs var $ ls bin CWAGENT_VERSION amazon-cloudwatch-agent-config-wizard config-downloader cwagent-otel-collector amazon-cloudwatch-agent amazon-cloudwatch-agent-ctl config-translator start-amazon-cloudwatch-agent ``` I try to start the OTel Collector using `sudo bin/amazon-cloudwatch-agent-ctl -a start -m ec2 -o default`. It doesn't start, though. Here's the process status: ``` $ sudo bin/amazon-cloudwatch-agent-ctl -a status { "status": "running", "starttime": "2022-06-09T07:12:30+0000", "configstatus": "configured", "cwoc_status": "stopped", "cwoc_starttime": "", "cwoc_configstatus": "configured", "version": "1.247350.0" } ``` And the log: ``` $ tail cwagent-otel-collector/logs/cwagent-otel-collector.log 2022/06/09 07:09:20 E! User aoc does not exist: no matching entries in passwd file 2022/06/09 07:09:20 E! Failed to ChangeUser: no matching entries in passwd file 2022/06/09 07:09:20 no matching entries in passwd file ``` Has anyone gotten the OTel Collector working with Elastic Beanstalk? In tips for how to configure this? -martin

Hi, we are using X-Ray for application logs tracing but X-Ray is not able to show case all the logs and after some times it drops off the tracing. Is it some limitations with X-Ray or how we can enables X-Ray for the complete logs tracing.

Hello, We are implementing X-Ray Notification for Insights using EventBridge. We would like to test it on our Dev environment. Is there any way to create an Insight manually? Or would you suggest some option to create Insight for testing purposes? Thank you.

HI, If I have a business process that include several events that trigger each other is there any tool I can use to see a chart of this chain so it will be easy to manage and understand the flow ? Thanks

Hello, I have added xray to my django app and it works correctly for every api call , Custom segment and subsegments I created except custom segments of functions that are called inside async_django_q_task. It looks like it does create traces for them but they are empty, example of such trace: ``` { "Id": "some_id", "Duration": null, "LimitExceeded": false, "Segments": [] } ``` Example of the way I try to trace the function in async_django_q_task: ``` def dummy_function(): xray_recorder.begin_segment('segment') do_something() xray_recorder.end_segment() async_django_q_task(dummy_function) ``` Thanks in advance for help.

Hello. Suppose that I have the following architecture: API Gateway (service name: gateway api) -> Lambda Function (service name: gateway handler) -> API Gateway (service name: internal service) -> Lambda Function (service name: internal service handler) The "internal service" api gateway has 2 endpoints (`POST /endpointA` and `POST /endpointX`), which both proxy to the same "handler" lambda function. Depending on a feature flag, incoming requests are requests to the "gateway handler" are sent to either `POST /endpointA` or `POST /endpointX`. All of these services are set up for tracing with AWS X-ray. I want to query X-Ray to gain insights into the performance of requests send to `X` vs `A`. I tried the following query ``` service("gateway handler") AND service("internal service") AND http.url ENDSWITH "/endpointX" // and also // service("gateway handler") AND service("internal service") AND http.url ENDSWITH "/endpointA" ``` However this does not seem to work. In general I cannot query using http.url, when that http request is not the first request in a trace. Is there a way around this? Bonus points if it does not require me to add additional code to annotate traces for the xray system.

I have set up a node app following the guide here https://docs.aws.amazon.com/xray/latest/devguide/xray-daemon-ecs.html Both containers are running but the X-ray daemon didn't receive data from the nodejs app. I check the x-ray daemon logs and found the below logs. Can anyone help with what I missed please? ``` [Error] Get instance id metadata failed: RequestError: send request failed caused by: Get http://169.254.169.254/latest/meta-data/instance-id: dial tcp 169.254.169.254:80: connect: invalid argument ```

Hi, I have a Feathersjs/eclipse Node.js API running on Elastic Beanstalk. I've just instrumented the app according to the "Getting Started" section on Node.js/eclipse. X-Ray tracing is already enabled for the Elastic Beanstalk app. I then deployed the newly instrumented app... and nothing is traced. Does anyone know if there is some additional instrumentation/configuration for Feathersjs app?

We're using AWS X-Ray to trace our long-running processes, but we're not able to view the traces until the root segment has ended. As soon as the long-running process finishes we're able to see the entire trace with all the sub-segments included, but nothing is displayed before this, even though many sub-segments finish while the process is running. The X-Ray API doesn't return the data either, so it's not just an issue in the X-Ray console. Is this expected behaviour? If not, how would we troubleshoot this issue?

Hi, We're trying to instrument our services using AWS X-Ray. Most of the app that we're interested in instrumenting doesn't get passed X-Ray trace ids but instead are called every quarter of a second via a timer to generate data and then pass downstream to another service (via TCP) before sending to clients over websockets. As far as we have found this requires manual instrumentation and we have setup the app to create a segment, create subsegments for parts of the code we would like to additionally instrument before passing the `trace-id` and `parent-id` downstream for the other services to pick up and continue using for their own subsegments. Running the app locally we can see X-Ray trace information in the dashboard and all appears okay. From reading the documentation however, it's unclear whether this is enough or whether we need to do additional work to ensure we're picking up the sampling rules or whether this is something the AWS-SDK will be doing automatically. It is also unclear whether we need to pass more than the `trace-id` and `parent-id` downstream as I have seen in some of the examples also passing a flag indicating whether the current trace is being actively traced. If anyone can let me know the ideal way of manually generating traces and passing between services that'd be extremely helpful. Regards

Firstly, the Lambda code itself has no issue, and executes fine. What I am seeing in the logs is an error that returns very little info on Google searches and is now a complete mystery to me. I have an Email alert scanning logs for "ERROR" and because this mysterious error has that in it, I keep getting emails for it. 2022-02-21T10:11:46.665+00:00 [ERROR] [1645438306665] LAMBDA_RUNTIME Failed to get next invocation. No Response from endpoint 2022-02-21T10:11:46.666+00:00 An error occurred while attempting to execute your code.: LambdaException 2022-02-21T10:11:46.678+00:00 terminate called after throwing an instance of 'std::logic_error' 2022-02-21T10:11:46.678+00:00 what(): basic_string::_M_construct null not valid Some extra info, which may be red herrings, but including them anyway. Lambda is written in .NET Core This seemed to start after enabling X-Ray (Possible red herring) The error shows up around 5 minutes after Lambda inactivity (Possible red herring) Any help would be greatly appreciated. Thanks

Hi Team, I have a lambda which basically performs the data transformation on service response. The lambda duration metrics is slowly spiking when the traffic is increasing. #### Referred already : * https://docs.aws.amazon.com/lambda/latest/dg/runtimes-context.html #### Things already configured for lambda: * Auto scaling the provisioned concurrency of lambda. * I have enabled the X - Ray trace for the lambda. * Found out that, Invocation phase duration is increasing when there is more incoming traffic. #### Screenshots for reference: * [RECENT_TRACES](https://paste.pics/e2046899130c7c0e9918f0c683ce9b96) * [BAD_SCENARIO](https://paste.pics/94ac3350eb3da24a3ea2f42035777384) * [GOOD_SCENARIO](https://paste.pics/2ee5a3711f04820edc0da3a8f1e3dc5c) Any idea why the invocation phase is increasing & how to handle this ?

Need some help debugging this long response time I'm seeing from my cloudfront CDN for images that have not been cached. The outline of our setup is that we have a cloudfront cdn that responds with cached images when available. If no cached image is available, there's a lambda that pulls the requested image from s3 and resizes it using sharp.js, then sends the resized image as the response to the request. Cloudfront caches this image and then uses it for subsequent requests for the same image. The problem is that this handling takes 2-3s usually. You can see in [this](https://i.stack.imgur.com/uCt4W.png) screenshot. I'm only partially aware of the breakdown of those 2-3s. That screenshot is of logs from cloudfront, so that means the problem must lie somewhere within our cloudfront setup. The lambda itself takes 800-1300ms from start to finish, and that includes the time it takes to pull the image from s3, resize it, convert it to a buffer, then respond to the request. We already use the [http keepAlive](https://aws.amazon.com/blogs/networking-and-content-delivery/leveraging-external-data-in-lambdaedge/) optimization to reduce the latency of pulling the image from s3. However the lambda's running time is often 50% or less of the total response time, so that means there's another significant bottleneck elsewhere that I haven't discovered, and I'm not sure how to go about finding it. I've tried enabling AWS X-Ray to get more insights into the problem but our lambda is on Lambda@Edge, which doesn't support X-Ray. What else can I investigate and where else could I look?

Is there any plan to support X-Ray subsegment creation for all AWS PHP SDK Calls?

Is there any tool, solution, framework, service or alike to allow memory profiling for NodeJS based lambda functions? Maybe even running on AWS directly and not on a local development environment? What would be the recommendation for such kind of profiling? I'm aware that local profiling can be done using Chrome's Inspect / DevTools, but that actually means debugging a "long running process", not a short-lived one like a lambda function. Additional Context: We migrated a reporting service from an EC2 instance to a ECS Fargate based docker container using a basic lift & shift approach, but with the catch that we assigned less memory to the ECS Task than the original configuration of the EC2 instance purposely, as usually the instance had quite a lot of unused resources. So, now we are transitioning the "outliers" that fail with OutOfMemoryError into lambdas, but we'd like to improve their code and not just move them to a lambda container.

https://docs.aws.amazon.com/xray/latest/devguide/xray-sdk-dotnet-configuration.html The configuration method for .NET apps only shows how to grab rules from a local JSON file. But how can an app query and grab sampling rules already defined beforehand in AWS console? And beyond that, if the online sampling rules are adjusted, do they get dynamically propagated to all existing related app systems?

One thing that bothered me right from the beginning learning about X-Ray is no native construct to define segregated whole systems and environments. An AWS account can be used to host multiple non-related or mutually-exclusive entire systems, and other services like VPC, EC2, S3, etc provide the necessary "fencing" features to keep them cleanly apart and secure. X-Ray on the other hand, seems to just want to collect traces from _all_ sources and dump everything into a single "bucket". Even if we define highly elaborate annotation protocols to clearly differentiate the systems/environments, what's to stop a team from looking at another team's traces and possibly sensitive data?

Hello, I'm trying to add a XRay segment to my NodeJs code that user express router to process incoming POST requests. I've added the openSegment/closeSegment calls around my router.post() method, but I don't see the segment in the XRay trace. I'm very new to this, so I'm probably missing something. Any help would be appreciated. Source code below: var AWSXRay = require('aws-xray-sdk'); const XRayExpress = AWSXRay.express; const router = require("express").Router(); const DB = require("../../db/oracle/oracleDB"); router.use(XRayExpress.openSegment('MemberGroupId')); /** * POST route */ router.post("/", async (req, res) => { const { patientId, dbName } = req.body; let memberLookUpObj = { payload: { cardHolderId: "", groupId: "", relationshipCode: "", }, msg: "", }; ... remaining source }); router.use(XRayExpress.closeSegment()); Thanks, Chris

I have a Python / Flask application running in a Docker container under Elastic Beanstalk. The X-Ray daemon (v3.2.0) is installed inside the container. The daemon's logs show that it is successfully sending batches of data to the X-Ray service. However, my application logs show a failure when the Python X-Ray SDK (v2.6.0) attempts to get sampling rules, with a stack dump ending as follows: File "/usr/local/lib/python3.8/site-packages/botocore/endpoint.py", line 269, in _send return self.http_session.send(request) File "/usr/local/lib/python3.8/site-packages/botocore/httpsession.py", line 283, in send raise EndpointConnectionError(endpoint_url=request.url, error=e) botocore.exceptions.EndpointConnectionError: Could not connect to the endpoint URL: "http://127.0.0.1:2000/GetSamplingRules" \[2021-02-11 19:53:22 +0000] \[17] \[INFO] No effective centralized sampling rule match. Fallback to local rules. I can confirm that the daemon is running, although a request to the URL shown above returns a 403 Forbidden error. I don't know if that's expected or not. Can anyone suggest what might be going on here? Is there an incompatibility between the 3.x daemon that's running and the 2.x SDK? Any help would be greatly appreciated. Thank you!

I was reading about the standardization efforts around distributed tracing, in particular, this document: https://www.w3.org/TR/trace-context/ While looking at AWS XRay's implementation, it seems to be going completely custom, with custom HTTP Headers, and custom formating of the hierarchy IDs. When is XRay going to switch to a more standardized format that is compatible with other systems that also implement that standard? I found this project when randomly searching for this topic: https://aws-otel.github.io/ I don't quite understand what that is and why it even coexists with XRay. Would appreciate clarification here.

We configured provisioned concurrency for our Java Lambda function to reduce its startup time. Part of the function's startup time is the Java code initializing itself though, which isn't reduced by provisioned concurrency. We'd like to know how long it takes for our code to initialize before it starts processing a request, so we can assess if it's worth the time and effort to optimize. I created a test system with a Java Lambda function, and activated AWS X-Ray with Active tracing, but I don't see a breakdown for initialization, invocation, and overhead. What's the best way to track initialization metrics for a Lambda function?

HI, I have a spring boot application (2.3.1) running in a AWS EC2 instance. I followed AWS documentation for Spring and AOP:https://docs.aws.amazon.com/xray/latest/devguide/xray-sdk-java-aop-spring.html I am able to see the traces and service map after I deploy on my AWS EC2 instance. But when I run my code on my local machine in dev environment(NOT on EC2), it throws an error. ================================================================================= 2020-08-19 10:22:55.655 \[pool-1-thread-1] INFO c.a.x.s.sampling.pollers.RulePoller -Polling sampling rules. 2020-08-19 10:22:56.670 \[pool-1-thread-1] ERROR c.a.x.s.sampling.pollers.RulePoller -Encountered error polling GetSamplingRules: com.amazonaws.SdkClientException: Unable to execute HTTP request: Connect to 127.0.0.1:2000 \[/127.0.0.1] failed: Connection refused: connect at com.amazonaws.http.AmazonHttpClient$RequestExecutor.handleRetryableException(AmazonHttpClient.java:1175) at com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeHelper(AmazonHttpClient.java:1121) at com.amazonaws.http.AmazonHttpClient$RequestExecutor.doExecute(AmazonHttpClient.java:770) at com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeWithTimer(AmazonHttpClient.java:744) at com.amazonaws.http.AmazonHttpClient$RequestExecutor.execute(AmazonHttpClient.java:726) at com.amazonaws.http.AmazonHttpClient$RequestExecutor.access$500(AmazonHttpClient.java:686) at com.amazonaws.http.AmazonHttpClient$RequestExecutionBuilderImpl.execute(AmazonHttpClient.java:668) at com.amazonaws.http.AmazonHttpClient.execute(AmazonHttpClient.java:532) at com.amazonaws.http.AmazonHttpClient.execute(AmazonHttpClient.java:512) at com.amazonaws.services.xray.AWSXRayClient.doInvoke(AWSXRayClient.java:1257) at com.amazonaws.services.xray.AWSXRayClient.invoke(AWSXRayClient.java:1226) at com.amazonaws.services.xray.AWSXRayClient.invoke(AWSXRayClient.java:1215) at com.amazonaws.services.xray.AWSXRayClient.executeGetSamplingRules(AWSXRayClient.java:568) at com.amazonaws.services.xray.AWSXRayClient.getSamplingRules(AWSXRayClient.java:539) at com.amazonaws.xray.strategy.sampling.pollers.RulePoller.pollRule(RulePoller.java:65) at com.amazonaws.xray.strategy.sampling.pollers.RulePoller.lambda$start$0(RulePoller.java:46) at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515) at java.base/java.util.concurrent.FutureTask.runAndReset(FutureTask.java:305) at java.base/java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:305) at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128) at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628) at java.base/java.lang.Thread.run(Thread.java:834) ======================================================================================= Questions: 1.) We do not need the traces to be generated and sent to aws. How do we TURN OFF/DISABLE Xray tracing feature on our dev machines ?

Is there a way to enable tracing for my ECS services that connects each other via Amazon MQ ? I have two services which running in ECS cluster. Service A receives incoming requests via ALB. It does.something and post message to Queue (Lets say it is QueueA). Listener that listens message from QueueA is part of service B. It recieves message from QueueA , process it and sends to another Queue/some other services. I know that currently Amazon MQ doesn&#39;t support Xray. Is there anyway to trace the complete flow? I have lots of micro services that is integrated via Amazon MQ.

I am working with two application, one is .net core, the other is asp.net. I would like to configure both, to exclude my health check requests. I noticed that the Ruby version has a configuration for just this type of thing. <https://github.com/cookpad/aws-xray> I spent quite a bit of time pouring over the .net version of the documentation, but was not able to find anything similar. I am under the assumption that I might be able to accomplish excluding out these requests, by setting up a sampling rule. However I was hoping there was some sort of configuration that would make this a bit easier.

I am looking at the doc here: https://docs.aws.amazon.com/xray-sdk-for-nodejs/latest/reference/module-http_p.html There is a downstreamXRayEnabled parameter but I don't think I understand what exactly it does. I looked at the source code to try to understand what it does. All I can see is that it added a `traced` attribute but it isn't apparent to me how the attribute is used. I also tried searching for usage of it by others but there aren't any meaningful results. So what's the effect of this flag exactly? Thanks

Customer has an AWS Landing Zone (ALZ) implementation where they are sharing a Transit Gateway (TGW) between accounts. Sharing a TGW results in error unless `Allow external accounts` is checked, even though the account is in the same organization. The account that they are trying to share the TGW with, is under the same root Organization by ALZ and AWS Control Tower configuration Why these accounts are considered externals?. Once allow external accounts is checked the TGW can be shared and the principal type shows "Account (External)"

I use sturctured logs of which the fields are automatically picked up by Insights. All logs have a field traceId. One log has the field itemId. This logs also has a traceId field. I want to view all logs based on the itemId. So I need to find the traceId related to the itemId. And use this traceId to view all logs with this traceId. This feels like a nested query in SQL. How can I achieve a nested query with Insights? Does something exist in the likes of : ``` fields @message | filter traceId=(fields @traceId | filter itemId="123" | limit 1) ```

Hi, I am trying to introduce Xray tracing in one of my services which basically is responsible for executing asynchronous tasks using SQS. This code that needs to be traced does not run in response to an HTTP request. I am not able to find a way where I can control the sampling of this code through the Sampling Rules that I configure on the Xray console. Since this application component spawns worker threads that run in background, I am resorting to creating and ending new segments around each such thread execution. I understand that for application code triggered through HTTP request, the AWSXRayServletFilter polls for Sampling Rules when a CentralizedSamplingStrategy is configured. Is there a way of leveraging such sampling for application code that is not invoked from an HTTP request? Any help would be greatly appreciated.

Hi, I am deploying AWS X-Ray daemon on an EC2 Instance (AWS Linux AMI). I am doing this by following the instructions provided in <https://docs.aws.amazon.com/xray/latest/devguide/xray-daemon-ec2.html> which asks to have the following in UserData script: ``` #!/bin/bash curl https://s3.dualstack.us-east-2.amazonaws.com/aws-xray-assets.us-east-2/xray-daemon/aws-xray-daemon-3.x.rpm -o /home/ec2-user/xray.rpm yum install -y /home/ec2-user/xray.rpm ``` The above install launches the X-Ray Daemon and it listens to the default UDP bind address 127.0.0.1:2000 (Loopback address). I require to make the daemon process listen to the public network (0.0.0.0:3000) and also want changes to the region and logging configuration. <https://docs.aws.amazon.com/xray/latest/devguide/xray-daemon-configuration.html> describes how this can be achieved. But I am facing the following issues: 1.) I am facing problems killing the X-Ray daemon on instance start up through the User Data script. pkill is not working for xray. 2.) I am unsure of how to start X-Ray daemon in the user data script as it is not as straight forward as executing the X-ray command to run in background mode i.e ``` /usr/bin/xray -c "/home/ec2-user/xray_config/daemon_config.yaml" & ``` will not work. I am wondering if it would be possible to specify a configuration file during X-Ray daemon installation so that the default daemon process that kicks off after install is launched with the configuration parameters (UDP bind address etc) that I specify. Can anyone please help me with this? Regards, Edited by: shricloud on May 27, 2019 12:41 PM Edited by: shricloud on May 27, 2019 12:42 PM

Hi, For the past week, I've been trying to access the X-Ray console for Paris (eu-west-3) region, but the service seems to be unavailable: the page keeps "Loading" indefinitely. It works just fine for other regions. Is there an ongoing issue with XRay in Paris ? Thanks

Hello, I have a pipeline which is as follow: lambda1 -> S3 -> lambda2 Lambda1 put an object in an S3 bucket and lambda2 is invoked by the S3 notification X-Ray doesn't make the link between lambda1 and lambda2. I tried to do it programmatically by making lambda1 pass its trace_id and segment id to lambda2 and then in lambda2 call putTraceSegments. Actually it worked but then nothing more is traced from lambda2. I guess it's because the trace_id computed by AWS in lambda2 doesn't match the one used to create the segment in lambda2. However, even if it were working, it's quite a lot of work to make this link. 1) Is there any plan to integrate this kind of scenario in X-Ray ? I saw in this thread https://forums.aws.amazon.com/thread.jspa?threadID=292267&tstart=0 that there is work going on for SNS integration. Would S3 notification be part of it ? 2) In such scenario lambda1 -> S3 -> lambda2 Wouldn't be better if X-Amzn-Trace-Id is forwarded between lambda1 and lambda2? Correlation between components of our pipeline is made difficult , it seems that only custom code would achieve it but I would like to avoid it at maximum. Thanks in advance

Hello, I am trying to use X-Ray on my ECS Fargate service. I have followed the steps here: <https://docs.aws.amazon.com/xray/latest/devguide/xray-daemon-ecs.html> I have used the first Dockerfile from that page. ``` FROM amazonlinux RUN yum install -y unzip RUN curl -o daemon.zip https://s3.dualstack.us-east-2.amazonaws.com/aws-xray-assets.us-east-2/xray-daemon/aws-xray-daemon-linux-2.x.zip RUN unzip daemon.zip && cp xray /usr/bin/xray ENTRYPOINT ["/usr/bin/xray", "-b", "0.0.0.0:2000"] EXPOSE 2000/udp ``` I have added the Docker image to an ECR repo and added one of these containers to my existing task definition. In the logs for my container I see this: ``` 2018-07-09 13:17:172018-07-09T20:17:17Z [Error] Get instance id metadata failed: RequestError: send request failed 2018-07-09 13:17:17caused by: Get http://169.254.169.254/latest/meta-data/instance-id: dial tcp 169.254.169.254:80: connect: invalid argument 2018-07-09 13:17:162018-07-09T20:17:16Z [Info] Initializing AWS X-Ray daemon 2.1.3 2018-07-09 13:17:162018-07-09T20:17:16Z [Info] Using buffer memory limit of 37 MB 2018-07-09 13:17:162018-07-09T20:17:16Z [Info] 592 segment buffers allocated 2018-07-09 13:17:162018-07-09T20:17:16Z [Info] Using region: us-west-2 ```