You can now use async methods in your Python agents

MyProcessor(Processor):
    async def process(self, record):
        await asyncio.sleep(1)
        return [record]

Another good point for Streaming in Generative AI application is about the fact that it is super easy to “replay” your data. You just need to point your application (subscription/consumer group) to the beginning of the “input topic” and see how the application would behave with the new version. (this is a typical property of applications based on streaming). Combining this ability with a framework to evaluate the LLM/application/prompts you can use the same production data as test cases It would be great to write a blog post about this topic.

hello, is it possible to use langstream in windows? because I have been trying to setup it but I am getting this error: $ langstream -v docker run test -app yamls -s secret/secrets.yaml Tenant: default Application: test Application directory: C:\lstream\yamls Running all the agents in the application Instance file: null Secrets file: C:\lstream\secret\secrets.yaml Start broker: true Start S3: true Start Database: true Start Webservices true Using docker image: ghcr.io/langstream/langstream-runtime-tester:0.5.0 Using config file C:\Users\macha\.langstream\config Using default instance file that connects to the Kafka broker inside the docker container profile local-docker-run updated Using config file C:\Users\macha\.langstream\config Internal error: java.lang.UnsupportedOperationException java.lang.UnsupportedOperationException at java.base/java.nio.file.Files.setPosixFilePermissions(Files.java:2166) at ai.langstream.cli.commands.docker.LocalRunApplicationCmd.makeDirOrFileReadable(LocalRunApplicationCmd.java:514) at ai.langstream.cli.commands.docker.LocalRunApplicationCmd.makeDirOrFileReadable(LocalRunApplicationCmd.java:509) at ai.langstream.cli.commands.docker.LocalRunApplicationCmd.prepareAppDirectory(LocalRunApplicationCmd.java:494) at ai.langstream.cli.commands.docker.LocalRunApplicationCmd.executeOnDocker(LocalRunApplicationCmd.java:319) at ai.langstream.cli.commands.docker.LocalRunApplicationCmd.run(LocalRunApplicationCmd.java:263) at picocli.CommandLine.executeUserObject(CommandLine.java:2026) at picocli.CommandLine.access$1500(CommandLine.java:148) at picocli.CommandLine$RunLast.executeUserObjectOfLastSubcommandWithSameParent(CommandLine.java:2461) at picocli.CommandLine$RunLast.handle(CommandLine.java:2453) at picocli.CommandLine$RunLast.handle(CommandLine.java:2415) at picocli.CommandLine$AbstractParseResultHandler.execute(CommandLine.java:2273) at picocli.CommandLine$RunLast.execute(CommandLine.java:2417) at picocli.CommandLine.execute(CommandLine.java:2170) at ai.langstream.cli.LangStreamCLI.execute(LangStreamCLI.java:74) at ai.langstream.cli.LangStreamCLI.main(LangStreamCLI.java:52) Cleaning environment Deleting temporary file: C:\Users\macha\.langstream\tmp\.langstream_app_590692518254300

one more question, sorry if it seems like a silly question but: com.azure.core.exception.HttpResponseException: Status code 429, "{ "error": { "message": "You exceeded your current quota, please check your plan and billing details.", "type": "insufficient_quota", "param": null, "code": "insufficient_quota" } } I am getting this error on one of the examples (solr example) however, I can't understand which service is exceeded the quota as logs are not really clear. I also tried to search this exception inside of the langchain and langstream but still nothing. do you have any advice on debugging or finding which service causes this problem?

Hi We did a fresh install of Langstream 0.5.0, When deploying an App we get the following error: In the log of control-plane pod:

io.fabric8.kubernetes.client.KubernetesClientException: Failure executing: PATCH at: <https://172.30.0.1:443/api/v1/namespaces/genai-langstream-default/secrets/langstream-app-1?fieldManager=fabric8>. Message: Apply failed with 1 conflict: conflict with "Mozilla" using v1: .data.secrets. Received status: Status(apiVersion=v1, code=409, details=StatusDetails(causes=[StatusCause(field=.data.secrets, message=conflict with "Mozilla" using v1, reason=FieldManagerConflict, additionalProperties={})], group=null, kind=null, name=null, retryAfterSeconds=null, uid=null, additionalProperties={}), kind=Status, message=Apply failed with 1 conflict: conflict with "Mozilla" using v1: .data.secrets, metadata=ListMeta(_continue=null, remainingItemCount=null, resourceVersion=null, selfLink=null, additionalProperties={}), reason=Conflict, status=Failure, additionalProperties={}).
	at io.fabric8.kubernetes.client.KubernetesClientException.copyAsCause(KubernetesClientException.java:238)
	at io.fabric8.kubernetes.client.dsl.internal.OperationSupport.waitForResult(OperationSupport.java:518)
	at io.fabric8.kubernetes.client.dsl.internal.OperationSupport.handleResponse(OperationSupport.java:535)

error message from the langstream cli: langstream-app-2 % langstream apps deploy langstream-app-2 -app ./application \ -i ../config/instance.yaml \ -s ../config/secrets.yaml packaging app: /Users/xyz/Documents/genai-langstream-samples/langstream-app-2/./application app packaged deploying application: langstream-app-2 (4 KB) Http request to https://langstream.rbi.cloud/api/applications/default/langstream-app-2?dry-run=false failed: { "type" : "about:blank", "title" : "Conflict", "status" : 409, "detail" : "409 CONFLICT \"Application already exists\"", "instance" : "/api/applications/default/langstream-app-2" } langstream-app-2 % what are we missing???

Hey folks, Today I and @Nicolò Boschi we have a local meetup in Italian about Realtime Generative AI (and LangStream.ai) https://www.linkedin.com/posts/enrico-olivelli-984b7874_langstream-an-event-driven-developer-platform-activity-7134880925892378624-rGN6?utm_source=share&utm_medium=member_android

Hello, I have a question about crawler agent. when I try to run crawler and if there is pdf hosted in the website, it says this: URL lead to a application/pdf content-type document. Skipping sample: https://s2.q4cdn.com/299287126/files/doc_downloads/governance/2023/amzn-2022-conflict-minerals-report.pdf is it possible to allow crawler to fetch content of pdf? thanks from now!

Check out the new blog about LangStream 0.4 by @David Dieruf! https://langstream.ai/2023/11/22/langstream-0_4/

Recordings of my talk in Milan https://www.linkedin.com/posts/sease-ltd_generativeai-with-apache-solr-and-langstreamai-activity-7136357938142584834-eG0h?utm_source=share&utm_medium=member_android

Hi <!here>, I'm new to langstream, I'm having trouble in installing langstream in my local ubuntu server running in docker.. The command I used curl -Ls "https://raw.githubusercontent.com/LangStream/langstream/main/bin/get-cli.sh" | bash Also tried the below one for downloading get "https://raw.githubusercontent.com/LangStream/langstream/main/bin/get-cli.sh" But getting this error Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|49.44.79.236|:443... failed: Connection refused. Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|24052001607282041:36|443... failed: Cannot assign requested address. Retrying. Is the URL I'm using is valid(it's the one in docs) Please do help if you know how to resolve

Hi <!here>, I'm new to langstream, I am using ubuntu system After installing docker java17 python3.11 langstream when i run a example application which is in github repo of langstream The example is

langstream docker run test \
   -app <https://github.com/LangStream/langstream/blob/main/examples/applications/openai-completions> \
   -s <https://github.com/LangStream/langstream/blob/main/examples/secrets/secrets.yaml>

after running the above example I had a ui poped up after clicking the connect button and asked my query I got a error as 105626.460 [Thread-10] INFO ai.langstream.runtime.tester.Main -- Shutdown hook 105626.461 [Thread-10] INFO a.langstream.impl.nar.NarFileHandler -- Not closing classloaders, in order to avoid class loading issues during development while the JVM is shutdown 105626.821 [SpringApplicationShutdownHook] INFO a.langstream.impl.nar.NarFileHandler -- Not closing classloaders, in order to avoid class loading issues during development while the JVM is shutdown Process exited with code 1 Cleaning environment Help me resolving this issue

Hi <!here>, I downloaded the cli in windows and tried to run my application using docker run, but getting the below exception. Any idea how to resolve this. my system conf java version: openjdk 11.0.14.1 2022-02-08 LTS python version : Python 3.9.7 windows version : 11 docker version: Docker version 24.0.6, build ed223bc and my langstream cli version is 0.5.0. If someone encountered this please help! Thanks! Exception:

C:\Users\DELL\Desktop\Data-aces\Luna Streaming\langstream-cli-0.5.0>.\bin\langstream docker run super-cool-app -app ./examples/applications/astradb-sink/ -i ./examples/instances/pulsar-docker.yaml

Tenant: default Application: super-cool-app Application directory: C:\Users\DELL\Desktop\Data-aces\Luna Streaming\langstream-cli-0.5.0\.\examples\applications\astradb-sink Running all the agents in the application Instance file: .\examples\instances\pulsar-docker.yaml Start broker: true Start S3: true Start Database: true Start Webservices true Using docker image: ghcr.io/langstream/langstream-runtime-tester:0.5.0 profile local-docker-run updated Internal error: java.lang.UnsupportedOperationException Cleaning environment

New blog post about LangStream. Please share on social media! https://x.com/langstream_ai/status/1732493285359317097?s=20 https://www.linkedin.com/feed/update/urn:li:activity:7138258659058311169

Hi <!here> how to get this export ASTRA_VECTOR_DB_ENDPOINT=https://....astra.datastax.com

@Nicolò Boschi found out that LangStream has been reviewed here https://stackshare.io/langstream This is great !

Is it possible to dynamically set a value in a agent? In example, using the Kafka Connect sink agent I wanted to use the PagerDuty sink connector. The pagerduty URL changes based on a header value in each message. Normally that value would be set in the pipeline but could it be dynamic?

Hey everyone! I'm trying to get LangStream set up as a docker container inside UnRaid OS. I see that there are 6 docker images (runtime, cli, control plane, runtime-tester, api gateway, etc.) I'd prefer to avoid doing the mini-kube setup first, and see if I can just get the CLI created as it's own docker container.... Is it necessary to combine all 6 docker images into a Dockerfile first, and then run using docker-compose? I work for a large footwear brand and we're exploring ways to expedite development of RAG applications for various team.

Any help would be greatly appreciated 🙂

Hi everyone We have migrated s3-source sample to use Cassandra instead of herdDB. Everything was already successfully running and now we are running in to an issue, when we are generating the embeddings for a new pdf file. The problem is when writing to the DB. We are not aware that we changed anything and we are also not able to read something meaningful out of the error: I reduced the embedding vector nad the text.

12:21:44.608 [main] INFO  ai.langstream.agents.s3.S3Source -- Skipping already processed object OReilly-DataMesh-earlyreading.pdf
12:21:44.608 [main] INFO  ai.langstream.agents.s3.S3Source -- Nothing found, sleeping for 5 seconds
12:21:45.802 [mapping-0] WARN  a.l.a.v.cassandra.CassandraWriter -- Error decoding/mapping Kafka record TransformRecord[context=MutableRecord(keySchemaType=STRING, keyNativeSchema=null, keyObject=OReilly-DataMesh-earlyreading.pdf, valueSchemaType=STRING, valueNativeSchema=null, valueObject={"name":"OReilly-DataMesh-earlyreading.pdf","chunk_text_length":"1481","text":"o'reilly?\n\ndata mesh\n\ndelivering data-driven value at scale\n\nraw & unedited\n\ncompliments of\n\nshe starburst\n\nzhamak dehghani\n\ our corporate/institutional\nsales department: 800-998-9938 or <mailto:corporate@oreilly.com|corporate@oreilly.com>.\n\nacquisitions editor: melissa duffield\ndevelopment editor: gary o?brien\nproduction editor: beth kelly","text_num_chunks":"156","chunk_id":"0","chunk_num_tokens":"385","filename":"OReilly-DataMesh-earlyreading.pdf","language":null,"embeddings_vector":[-0.016116302,-0.009884279,-0.02481215,-0.026594799,0.024464317,0.03304422,-0.012536513,-0.014869899,-0.035508044,-0.055769365,0.017072847,0.016507616,0.0137829175,0.0053298296,-0.006945808,0.015203239,0.012464047,0.0038877684,-0.018812016,-0.030522423,-0.009145132,0.009406008,0.018203307,-0.011427793,-6.20032E-4,-0.002365995,0.04214587,-0.0039203777,0.00370117,-0.02537738,0.0023207043,-0.019174343, TRUNCATED ]}, key=null, properties={name=OReilly-DataMesh-earlyreading.pdf, chunk_text_length=1481, text_num_chunks=156, chunk_id=0, chunk_num_tokens=385}, inputTopic=null, outputTopic=null, eventTime=1702902096311, dropCurrentRecord=false, customContext={}, recordObject=null), headers=[SimpleRecord.SimpleHeader(key=name, value=OReilly-DataMesh-earlyreading.pdf), SimpleRecord.SimpleHeader(key=chunk_text_length, value=1481), SimpleRecord.SimpleHeader(key=text_num_chunks, value=156), SimpleRecord.SimpleHeader(key=chunk_id, value=0), SimpleRecord.SimpleHeader(key=chunk_num_tokens, value=385)]]: null
12:21:45.816 [reactor-http-io_uring-1] INFO  a.l.r.agent.StandardErrorsHandler -- Handling error java.util.concurrent.CompletionException: java.util.NoSuchElementException for source record ai.langstream.agents.s3.S3Source$S3SourceRecord@41bacf48, errors count 1 (max retries 0)
12:21:45.816 [reactor-http-io_uring-1] ERROR a.l.runtime.agent.AgentRunner -- Unrecoverable error while processing some the records, failing
java.util.concurrent.CompletionException: java.util.NoSuchElementException
	at java.base/java.util.concurrent.CompletableFuture.encodeThrowable(CompletableFuture.java:332)
	at java.base/java.util.concurrent.CompletableFuture.uniRunNow(CompletableFuture.java:813)
	at java.base/java.util.concurrent.CompletableFuture.uniRunStage(CompletableFuture.java:803)
	at java.base/java.util.concurrent.CompletableFuture.thenRun(CompletableFuture.java:2195)
	at ai.langstream.agents.vector.VectorDBSinkAgent.write(VectorDBSinkAgent.java:54)
	at ai.langstream.api.runner.code.AgentCodeAndLoader$2.lambda$write$0(AgentCodeAndLoader.java:160)
	at ai.langstream.api.runner.code.AgentCodeAndLoader.callNoExceptionWithContextClassloader(AgentCodeAndLoader.java:80)
	at ai.langstream.api.runner.code.AgentCodeAndLoader$2.write(AgentCodeAndLoader.java:159)
	at ai.langstream.runtime.agent.AgentRunner.writeRecordToTheSink(AgentRunner.java:776)
	at ai.langstream.runtime.agent.AgentRunner.processRecordsOnTheSink(AgentRunner.java:757)
	at ai.langstream.runtime.agent.AgentRunner.lambda$runMainLoop$8(AgentRunner.java:704)
	at ai.langstream.runtime.agent.AgentRunner.lambda$runProcessorAgent$11(AgentRunner.java:931)
	at ai.langstream.runtime.agent.CompositeAgentProcessor.lambda$invokeProcessor$0(CompositeAgentProcessor.java:198)
	at ai.langstream.ai.agents.GenAIToolKitAgent.lambda$process$0(GenAIToolKitAgent.java:87)
	at java.base/java.util.concurrent.CompletableFuture.uniWhenComplete(CompletableFuture.java:863)
	at java.base/java.util.concurrent.CompletableFuture$UniWhenComplete.tryFire(CompletableFuture.java:841)
	at java.base/java.util.concurrent.CompletableFuture.postComplete(CompletableFuture.java:510)
	at java.base/java.util.concurrent.CompletableFuture.complete(CompletableFuture.java:2147)
	at com.datastax.oss.streaming.ai.ComputeAIEmbeddingsStep.lambda$processBatch$0(ComputeAIEmbeddingsStep.java:148)
	at com.datastax.oss.streaming.ai.ComputeAIEmbeddingsStep.lambda$processBatch$2(ComputeAIEmbeddingsStep.java:212)
	at java.base/java.util.concurrent.CompletableFuture$UniAccept.tryFire(CompletableFuture.java:718)
	at java.base/java.util.concurrent.CompletableFuture.postComplete(CompletableFuture.java:510)
	at java.base/java.util.concurrent.CompletableFuture.complete(CompletableFuture.java:2147)
	at reactor.core.publisher.MonoToCompletableFuture.onNext(MonoToCompletableFuture.java:64)
	at reactor.core.publisher.FluxMapFuseable$MapFuseableSubscriber.onNext(FluxMapFuseable.java:129)
	at reactor.core.publisher.MonoFlatMap$FlatMapMain.onNext(MonoFlatMap.java:158)
	at reactor.core.publisher.MonoFlatMap$FlatMapMain.secondComplete(MonoFlatMap.java:245)
	at reactor.core.publisher.MonoFlatMap$FlatMapInner.onNext(MonoFlatMap.java:305)
	at reactor.core.publisher.FluxSwitchIfEmpty$SwitchIfEmptySubscriber.onNext(FluxSwitchIfEmpty.java:74)
	at reactor.core.publisher.FluxMapFuseable$MapFuseableSubscriber.onNext(FluxMapFuseable.java:129)
	at reactor.core.publisher.FluxMapFuseable$MapFuseableSubscriber.onNext(FluxMapFuseable.java:129)
	at reactor.core.publisher.Operators$BaseFluxToMonoOperator.completePossiblyEmpty(Operators.java:2071)
	at reactor.core.publisher.MonoCollect$CollectSubscriber.onComplete(MonoCollect.java:145)
	at reactor.core.publisher.FluxMap$MapSubscriber.onComplete(FluxMap.java:144)
	at reactor.core.publisher.FluxDoFinally$DoFinallySubscriber.onComplete(FluxDoFinally.java:128)
	at reactor.core.publisher.FluxMap$MapSubscriber.onComplete(FluxMap.java:144)
	at reactor.core.publisher.FluxMap$MapSubscriber.onComplete(FluxMap.java:144)
	at reactor.netty.channel.FluxReceive.onInboundComplete(FluxReceive.java:413)
	at reactor.netty.channel.ChannelOperations.onInboundComplete(ChannelOperations.java:431)
	at reactor.netty.channel.ChannelOperations.terminate(ChannelOperations.java:485)
	at reactor.netty.http.client.HttpClientOperations.onInboundNext(HttpClientOperations.java:712)
	at reactor.netty.channel.ChannelOperationsHandler.channelRead(ChannelOperationsHandler.java:113)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:444)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:420)
	at io.netty.channel.AbstractChannelHandlerContext.fireChannelRead(AbstractChannelHandlerContext.java:412)
	at com.azure.core.http.netty.implementation.AzureSdkHandler.channelRead(AzureSdkHandler.java:222)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:442)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:420)
	at io.netty.channel.AbstractChannelHandlerContext.fireChannelRead(AbstractChannelHandlerContext.java:412)
	at io.netty.channel.CombinedChannelDuplexHandler$DelegatingChannelHandlerContext.fireChannelRead(CombinedChannelDuplexHandler.java:436)
	at io.netty.handler.codec.ByteToMessageDecoder.fireChannelRead(ByteToMessageDecoder.java:346)
	at io.netty.handler.codec.ByteToMessageDecoder.channelRead(ByteToMessageDecoder.java:318)
	at io.netty.channel.CombinedChannelDuplexHandler.channelRead(CombinedChannelDuplexHandler.java:251)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:442)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:420)
	at io.netty.channel.AbstractChannelHandlerContext.fireChannelRead(AbstractChannelHandlerContext.java:412)
	at io.netty.handler.ssl.SslHandler.unwrap(SslHandler.java:1466)
	at io.netty.handler.ssl.SslHandler.decodeNonJdkCompatible(SslHandler.java:1340)
	at io.netty.handler.ssl.SslHandler.decode(SslHandler.java:1380)
	at io.netty.handler.codec.ByteToMessageDecoder.decodeRemovalReentryProtection(ByteToMessageDecoder.java:529)
	at io.netty.handler.codec.ByteToMessageDecoder.callDecode(ByteToMessageDecoder.java:468)
	at io.netty.handler.codec.ByteToMessageDecoder.channelRead(ByteToMessageDecoder.java:290)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:444)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:420)
	at io.netty.channel.AbstractChannelHandlerContext.fireChannelRead(AbstractChannelHandlerContext.java:412)
	at io.netty.channel.DefaultChannelPipeline$HeadContext.channelRead(DefaultChannelPipeline.java:1410)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:440)
	at io.netty.channel.AbstractChannelHandlerContext.invokeChannelRead(AbstractChannelHandlerContext.java:420)
	at io.netty.channel.DefaultChannelPipeline.fireChannelRead(DefaultChannelPipeline.java:919)
	at io.netty.incubator.channel.uring.AbstractIOUringStreamChannel$IOUringStreamUnsafe.readComplete0(AbstractIOUringStreamChannel.java:299)
	at io.netty.incubator.channel.uring.AbstractIOUringChannel$AbstractUringUnsafe.readComplete(AbstractIOUringChannel.java:482)
	at io.netty.incubator.channel.uring.IOUringEventLoop.handleRead(IOUringEventLoop.java:290)
	at io.netty.incubator.channel.uring.IOUringEventLoop.handle(IOUringEventLoop.java:256)
	at io.netty.incubator.channel.uring.UserData.decode(UserData.java:30)
	at io.netty.incubator.channel.uring.IOUringCompletionQueue.process(IOUringCompletionQueue.java:90)
	at io.netty.incubator.channel.uring.IOUringEventLoop.run(IOUringEventLoop.java:203)
	at io.netty.util.concurrent.SingleThreadEventExecutor$4.run(SingleThreadEventExecutor.java:997)
	at io.netty.util.internal.ThreadExecutorMap$2.run(ThreadExecutorMap.java:74)
	at io.netty.util.concurrent.FastThreadLocalRunnable.run(FastThreadLocalRunnable.java:30)
	at java.base/java.lang.Thread.run(Thread.java:840)
Caused by: java.util.NoSuchElementException: null
	at com.datastax.oss.driver.shaded.guava.common.collect.AbstractIndexedListIterator.next(AbstractIndexedListIterator.java:75)
	at com.datastax.oss.driver.internal.core.type.codec.CqlVectorCodec.encode(CqlVectorCodec.java:68)
	at com.datastax.oss.driver.internal.core.type.codec.CqlVectorCodec.encode(CqlVectorCodec.java:33)
	at com.datastax.oss.dsbulk.codecs.api.ConvertingCodec.encode(ConvertingCodec.java:71)
	at com.datastax.oss.common.sink.RecordMapper.bindColumn(RecordMapper.java:299)
	at com.datastax.oss.common.sink.RecordMapper.bindColumnsToBuilder(RecordMapper.java:203)
	at com.datastax.oss.common.sink.RecordMapper.map(RecordMapper.java:122)
	at com.datastax.oss.common.sink.AbstractSinkTask.mapAndQueueRecord(AbstractSinkTask.java:211)
	at com.datastax.oss.common.sink.AbstractSinkTask.lambda$put$1(AbstractSinkTask.java:108)
	at java.base/java.util.concurrent.CompletableFuture$AsyncRun.run(CompletableFuture.java:1804)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:635)
	... 1 common frames omitted

We are trying to develop templates for different application where we would like to run multiple instances. For example "Chat we your different document-sets". For that we try to standardise as much a possible and work intensively with the yaml-templating. so far everything works nicely except when I try to use globals for constructing names. for example: in my instance.yaml

instance:

globals:

app-name: "${DEFAULT_APP_NAME:-defaultappname}"

in my configuration.yaml

configuration:

resources:

- type: open-ai-configuration

name: "${ globals.app-name } - Open AI configuration"

configuration:

url: "${ secrets.open-ai.url }"

access-key: "${ secrets.open-ai.access-key }"

provider: "${ secrets.open-ai.url }"

export DEFAULT_APP_NAME=app1

I would have expected something some like that

configuration:

resources:

- type: open-ai-configuration

name: "app1 - Open AI configuration"

config ...

but globals.app-name does not get replaced with env variable or the default text. Is this supported by langstream??

I'm developing a custom python package to be used with LangStream. In the python folder I have: • myCustomProcessor.py • requirements.txt • smart_logs-0.0.1-py3-none-any.whl The content of requirements is:

./smart_logs-0.0.1-py3-none-any.whl

When I run

langstream python load-pip-requirements -app ./application

I get

Using docker image: <http://ghcr.io/langstream/langstream-runtime-tester:0.5.6|ghcr.io/langstream/langstream-runtime-tester:0.5.6>
Logging to file: /tmp/langstream11196309498169874851.log
WARNING: The directory '/.cache/pip' or its parent directory is not owned or is not writable by the current user. The cache has been disabled. Check the permissions and owner of that directory. If executing pip with sudo, you should use sudo's -H flag.
Processing ./smart_logs-0.0.1-py3-none-any.whl (from -r requirements.txt (line 1))
Installing collected packages: smart-logs
Successfully installed smart-logs-0.0.1
ERROR: Exception:
Traceback (most recent call last):
  File "/usr/lib/python3.11/shutil.py", line 825, in move
    os.rename(src, real_dst)
OSError: [Errno 18] Invalid cross-device link: '/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info' -> '/code/application/python/lib/smart_logs-0.0.1.dist-info'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/local/lib/python3.11/dist-packages/pip/_internal/cli/base_command.py", line 180, in exc_logging_wrapper
    status = run_func(*args)
             ^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/dist-packages/pip/_internal/cli/req_command.py", line 248, in wrapper
    return func(self, options, args)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/dist-packages/pip/_internal/commands/install.py", line 510, in run
    self._handle_target_dir(
  File "/usr/local/lib/python3.11/dist-packages/pip/_internal/commands/install.py", line 569, in _handle_target_dir
    shutil.move(os.path.join(lib_dir, item), target_item_dir)
  File "/usr/lib/python3.11/shutil.py", line 841, in move
    copytree(src, real_dst, copy_function=copy_function,
  File "/usr/lib/python3.11/shutil.py", line 561, in copytree
    return _copytree(entries=entries, src=src, dst=dst, symlinks=symlinks,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/lib/python3.11/shutil.py", line 515, in _copytree
    raise Error(errors)
shutil.Error: [('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/top_level.txt', '/code/application/python/lib/smart_logs-0.0.1.dist-info/top_level.txt', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/REQUESTED', '/code/application/python/lib/smart_logs-0.0.1.dist-info/REQUESTED', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/LICENSE', '/code/application/python/lib/smart_logs-0.0.1.dist-info/LICENSE', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/RECORD', '/code/application/python/lib/smart_logs-0.0.1.dist-info/RECORD', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/direct_url.json', '/code/application/python/lib/smart_logs-0.0.1.dist-info/direct_url.json', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/INSTALLER', '/code/application/python/lib/smart_logs-0.0.1.dist-info/INSTALLER', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/WHEEL', '/code/application/python/lib/smart_logs-0.0.1.dist-info/WHEEL', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info/METADATA', '/code/application/python/lib/smart_logs-0.0.1.dist-info/METADATA', '[Errno 1] Operation not permitted'), ('/tmp/pip-target-atzg9f4_/lib/python/smart_logs-0.0.1.dist-info', '/code/application/python/lib/smart_logs-0.0.1.dist-info', '[Errno 1] Operation not permitted')]

[notice] A new release of pip is available: 23.2.1 -> 23.3.2
[notice] To update, run: python3 -m pip install --upgrade pip

But if I effectively do what that command is doing:

cd python
pip install -r requirements.txt -t ./lib

The lib folder is loaded correctly with my custom package. What have I missed with the langstream command?

@Enrico Olivelli I'll start a new thread for this but it's in the same idea of keeping your AI applications "on premises". I want to use a local LLM file (.gguf) in my custom python processor. If I include it in the "python" folder, will it make in to the agent? Could I reference it as "./super-cool-llm.gguf"? I'll do some discovery here, just thought I would put it out there.

What's the best way to set environment variables when running

langstream python run-tests

?

@Enrico Olivelli: I am trying to explore search chatbot capabilities on few internal websites using Solr. I was looking at Langstream query-solr example. But i am trying to understand do Langstream use OpenAI/VertexAI/Hugging faces to index contents into Solr Vector DB and how it does it? Also when user submit a question to prompt, does LangStream analyze it using AI and generate vectors possibilities for questions?

@Dheeraj Chaudhari starting a new thread for this question:

does LangStream analyze it using AI and generate vectors possibilities for questions?

@Enrico Olivelli I am having trouble using the

docker run

command on my Ubuntu machine. It seems that there is a race condition between unpacking the Kafka client NAR and starting the agents. I keep getting the error below. Sometimes my app will start successfully, but many times it will not. I recently upgraded my machine and it has gotten worse. Now I have a very low chance of actually getting my app to start using

docker run

. Any thoughts on how to work around this? I have tried restricting the CPUs (

--cpus

) and increasing the memory (

--memory

) but neither has worked.

14:48:59.319 [compute-embeddings-runner-7ce9f6ba-cf92-449c-a1d1-f2e0d5cf1faf] ERROR a.l.r.tester.LocalApplicationRunner -- 7ce9f6ba-cf92-449c-a1d1-f2e0d5cf1faf Error on AgentPod compute-embeddings
java.lang.NoClassDefFoundError: org/apache/kafka/connect/util/TopicAdmin
        at ai.langstream.kafka.runner.KafkaTopicConnectionsRuntime$1.start(KafkaTopicConnectionsRuntime.java:148)
        at ai.langstream.api.runner.topics.TopicConnectionsRuntimeAndLoader$1$4.lambda$start$0(TopicConnectionsRuntimeAndLoader.java:266)
        at ai.langstream.api.runner.topics.TopicConnectionsRuntimeAndLoader.executeNoExceptionWithContextClassloader(TopicConnectionsRuntimeAndLoader.java:42)
        at ai.langstream.api.runner.topics.TopicConnectionsRuntimeAndLoader$1$4.start(TopicConnectionsRuntimeAndLoader.java:266)
        at ai.langstream.runtime.agent.AgentRunner.runJavaAgent(AgentRunner.java:367)
        at ai.langstream.runtime.agent.AgentRunner.run(AgentRunner.java:191)
        at ai.langstream.runtime.agent.AgentRunner.runAgent(AgentRunner.java:125)
        at ai.langstream.runtime.tester.LocalApplicationRunner.lambda$executeAgentRunners$2(LocalApplicationRunner.java:235)
        at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:539)
        at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264)
        at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1136)

This looks cool ! https://docs.docker.com/engine/reference/commandline/debug/

@Chris Bartholomew you may be interested in this article https://blog.arkey.fr/2020/11/30/off-heap-reconnaissance/

Is there any simple code example to show us how to use langstream to trigger chatbot alert to users of any real time stock news, events or prices movement since they are time sensitive?

Is there any way to limit the number of inflight requests being processed by a LangStream agent. Sometimes I have agents that are backed up saying they are working on over 10K requests. This blows up memory requirements and I end up hitting OOM issues.