Shuffle
Security
Features
Usecases
Docs
Sign InGet Started

Search

Ctrl+K

Usability

5

  • Getting Started
  • About Shuffle
  • Tenants & Multi-Tenancy
  • Features Overview
  • API Reference

Automation

3

Security

4

Infrastructure

4

Troubleshooting

23 min read
yashsinghcodes
0x0elliot
frikky
ayush0033
Edit on GitHub
23m to read

On this page

Debugging Executions
Debugging in the non scale mode
How to get logs
Docker logs
Kubernetes logs
Health check endpoints
Common issues and how to fix them
UI does not load
Backend starts but login or setup fails
Apps are missing or won't download
Executions stay queued or never start
Worker starts but app actions fail
OpenSearch is crashing or unhealthy
High memory usage or host slowing down
Need help? Here's what to share with us
Quick script to bundle your logs
Resetting MFA
Orborus backend connection problems
Orborus can not reach backend
Load all apps locally
How to stop executions in loop
Abort all running executions of a specific workflow
Opensearch permissions error
Delete user
Docker client version too new
Recover admin user
Useful OpenSearch Queries
Find a user and their Tenants
Find all tenant
OpenSearch Dashboards
Adding to docker-compose.yml
Accessing the Dashboard
Using Dev Tools
Security Note
Extract all workflows
Moving from Docker in a VM to Kubernetes or SaaS
Before you start
Step 1: Create a backup before migration
Step 2A: Move from Docker/VM to Kubernetes
Step 2B: Move from Docker/VM to Shuffle SaaS
Step 3: Validate the migration
Step 4: Decommission the old environment
Summary
Rebuilding an OpenSearch index
OpenSearch consumes too much memory
Workflows are stuck
Disclaimer
Updates failing
Database not starting
OpenSearch TLS certificate setup reference
TLS Timeout and EOF Errors
Shuffle on ARM
Permission denied on files
Docker Permission denied
Server is slow
How to handle wrong or bad images on old versions of docker.
Docker not working
Troubleshooting for executions not running in swarm mode
Swarm init advertise address error manual fix
Find code OpenAPI app
Tenants/Suborgs seem to be lost
Find top index items opensearch
Add user to lost tenant
No module named xyz in execute python Shuffle Tools
Changing timestamps for an index from Unix to timestamps

Documentation for troubleshooting and debugging known issues in Shuffle.

Debugging Executions

Please check the Debugging section in the Configuration documentation

Debugging in the non scale mode

If you're running Shuffle in the non-scale mode, you can get the worker logs like this:

Bash

Editable

Response

Similarly, for any of the app logs, you can use the same command, but replace "worker" with the app name (this gives the shuffle tools logs):

Bash

Editable

Response

How to get logs

When debugging or reporting an issue, the first thing to check is logs from both the server (frontend, backend, database) and the runtime (Orborus, workers, and apps).

Docker logs

Check what's actually running:

Bash

Editable

Response

Get the last 300 lines of logs from the main services:

Bash

Editable

Response

Follow logs live while you test or reproduce the problem:

Bash

Editable

Response

To see worker and app containers currently running:

Bash

Editable

Response

To get logs from a specific worker or app container:

Bash

Editable

Response

Kubernetes logs

Check pods and events in your shuffle namespace:

Bash

Editable

Response

Get logs from the core deployments:

Bash

Editable

Response

Find and inspect worker/app pods:

Bash

Editable

Response

Health check endpoints

You can ping Shuffle's health endpoints directly to see if the API is responding:

API Call

GET
cURL
Python
HTTP

Response

No response received.

To check OpenSearch health:

  • From Docker:

    Bash

    Editable

    Response

  • From Kubernetes:

    Bash

    Editable

    Response


Common issues and how to fix them

Here's a breakdown of the most common issues people run into with Shuffle, and how to fix them.

UI does not load

If you open Shuffle in your browser and get a blank page or connection refused:

  1. Make sure shuffle-frontend is running (docker compose ps or kubectl get pods -n shuffle).
  2. Make sure port 3001 (HTTP) or 3443 (HTTPS) is open in your firewall (ufw, iptables, AWS security groups).
  3. Make sure the frontend container can reach shuffle-backend:5001.
  4. Check the frontend and backend logs:

    Bash

    Editable

    Response

    If on Kubernetes:

    Bash

    Editable

    Response

Backend starts but login or setup fails

If you can't create your first admin user, or get a 401 Unauthorized / 500 Internal Error:

  1. Check if OpenSearch is running and healthy:

    Bash

    Editable

    Response

  2. Double check that SHUFFLE_OPENSEARCH_PASSWORD in your .env matches the password OpenSearch was initialized with (OPENSEARCH_INITIAL_ADMIN_PASSWORD).
  3. Check permissions on your database folder — user 1000 needs ownership:

    Bash

    Editable

    Response

  4. Check backend logs to see the exact error:

    Bash

    Editable

    Response

    PS: If you changed SHUFFLE_ENCRYPTION_MODIFIER, existing credentials won't be decryptable.

Apps are missing or won't download

If you go to /apps and the page is empty, or you get errors when clicking "Download from Github":

  1. Shuffle needs to reach https://github.com/shuffle/python-apps to pull apps. Check if the backend has internet access:

    Bash

    Editable

    Response

  2. If you are behind a corporate proxy, make sure HTTP_PROXY and HTTPS_PROXY are set in .env.
  3. Make sure ./shuffle-apps directory is writable by Docker.
  4. If you don't have internet access at all, see how to load all apps locally.

Executions stay queued or never start

If your workflows stay in WAITING status or execution never progresses:

  1. Check if Orborus is running. Orborus is what polls the backend and spawns workers:

    Bash

    Editable

    Response

  2. Make sure Orborus can reach the backend. If you're using Docker, OUTER_HOSTNAME needs to be an IP or hostname that newly created containers can reach from their network (don't use localhost or 127.0.0.1!).
  3. If running Docker Swarm, make sure the shuffle_swarm_executions network exists and isn't corrupted. If in doubt: delete the network and restart Orborus (it will recreate it).
  4. If on Kubernetes, make sure Orborus has RBAC permissions to create deployments in the shuffle namespace:

    Bash

    Editable

    Response

  5. Check if your concurrency limit is maxed out (SHUFFLE_ORBORUS_EXECUTION_CONCURRENCY).

Worker starts but app actions fail

If a workflow starts, but an app node fails immediately with FAILURE or exit code 1:

  1. Find the failing worker or app container and inspect its logs:

    Bash

    Editable

    Response

  2. Check if Docker can pull the app image (from ghcr.io or Docker Hub).
  3. If the app needs to talk to internal systems through a proxy, make sure SHUFFLE_PASS_APP_PROXY=true is set.
  4. If using internal HTTPS with a self-signed certificate, you may need to mount your root CA into the app container or disable SSL verification in the app's configuration.

OpenSearch is crashing or unhealthy

If OpenSearch restarts in a loop, exits with code 137 (OOM), or gives permission errors:

  1. Check host RAM (free -m). OpenSearch is memory-hungry; we recommend at least 8 GB RAM on the host.
  2. Check vm.max_map_count:

    Bash

    Editable

    Response

    If it's not 262144, set it:

    Bash

    Editable

    Response

  3. Turn off swap:

    Bash

    Editable

    Response

  4. Fix folder permissions:

    Bash

    Editable

    Response

High memory usage or host slowing down

If Shuffle is eating up too much RAM or CPU:

  1. Check what's taking up resources:

    Bash

    Editable

    Response

  2. If app logs are filling up memory under heavy load, turn off log streaming by adding this to .env:

    Bash

    Editable

    Response

    Restart Orborus after changing this.
  3. Clean up stopped containers and dangling images:

    Bash

    Editable

    Response

  4. Lower the execution concurrency in .env:

    Bash

    Editable

    Response


Need help? Here's what to share with us

If you're stuck and need help from the Shuffle team or community, you can reach out on Discord, open an issue on GitHub, or email us at support@shuffler.io.

Before sending logs, please make sure to remove any passwords, API keys, or sensitive customer data!

Here's what helps us debug your issue fastest:

  • How you're running Shuffle (Docker Compose, Docker Swarm, Kubernetes, or Hybrid)
  • What version of Shuffle you're on (e.g. 2.1.1 or nightly)
  • Your sanitized .env or values.yaml
  • A screenshot or description of the error
  • Logs from around the time the issue happened

Quick script to bundle your logs

To save time, here's a script that grabs all the relevant info into a single .tar.gz file you can share with us:

For Docker:

Bash

Editable

Response

For Kubernetes:

Bash

Editable

Response


Resetting MFA

MFA can be enabled for your account on the settings User page of a tenant, or on your settings page. If you have lost access to your account due to this however, follow these steps:

Cloud (shuffler.io): Send an email to support@shuffler.io using the email you want MFA removed for. Onprem: It's a bit more tricky onprem, as we'll need to modify the Opensearch database. Here is how:

  1. Follow the backend logs WHILE logging into Shuffle to find your users' ID. It's a UUID in the format 550e8400-e29b-41d4-a716-446655440000.

Bash

Editable

Response

  1. After you've identified the user id you want to remove MFA for, docker exec to get bash session into OpenSearch container or any other container so long as the said container can communicate to the Opensearch container

Bash

Editable

Response

  1. Once in the container from step 2 above, run the following command after modifying the USERID field.

API Call

POST
cURL
Python
HTTP
JSON

Response

No response received.

  1. Restart the backend server: docker restart shuffle-backend. This is to fix potential caching problems.

Orborus backend connection problems

Due to the nature of Shuffle at scale, there are bound to be network issues. As Shuffle runs in Docker, and sometimes in swarm with k8s networking, it complicates the matter even further. Here's a list of things to help with debugging networking. If all else fails; reboot the machine & docker.

  1. Is the Orborus container speaking to an IP in the same network? Check with docker inspect shuffle-oroburs -> is the same CIDR / network address to be seen in the list of Network IPs?
  2. Do all the required networks exist?
    1. WITHOUT swarm: minimum 1 bridge network.
    2. WITH swarm: 3 networks (ingress, shuffle_shuffle, shuffle_swarm_executions). All in overlay mode.
  3. Are all the networks configured properly? If in swarm mode; delete the networks, then restart Orborus (it will remake them).
  4. Are the right network modes in use? Bridge (normal) vs Overlay (swarm). This depends on the server structure (1 vs. many).
  5. Is net.ipv4.ip_forward=1 set in the /etc/sysctl.conf file? If not, add it to the file, then exit it and type sysctl -p. This allows network cards to talk to each other on the same machine.
  6. Is DNS available inside the container? You can configure DNS in ALL docker containers on a server by editing this file: /etc/docker/daemon.json. To add DNS entries, add them as such:

JSON

Response

"Fixes" (in order):

  • Remake all networks (docker network rm)
  • Restart Docker (systemctl stop -> start)
  • Reboot (the whole server)

Orborus can not reach backend

In certain cases there may be DNS issues, leading to hanging executions. This is in most cases due to apps not being able to find the backend in some way. That's why the best solution if possible is to use the IP as hostname for Orborus -> Backend communication.

Load all apps locally

In certain cases, you may have an issue loading apps into Shuffle. If this is the case, it most likely means you have proxy issues, and can't reach github.com, where our apps are hosted.

Here's how to manually load them into Shuffle using git.

Code

Editable

Response

Alternatively: You can go download the latest Shuffle apps in your browser, and manually extract the .zip file into the ./shuffle/shuffle-apps folder.

How to stop executions in loop

  1. Run docker ps

    Code

    Editable

    Response

  2. Run docker inspect -f '{{range.NetworkSettings.Networks}}{{.IPAddress}}{{end}}' container_name_or_id

    Bash

    Editable

    Response

    Output:

    Code

    Editable

    Response

  3. Run curl -XDELETE http://<container_ip>:9200/workflowqueue-shuffle

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

    Output:

    JSON

    Response

Abort all running executions of a specific workflow

Follow Python scripts allows to massively stop all running executions of a workflow

Python

Editable

Response

Copy the script into a file called abort_running_executions.py and run it with

Code

Editable

Response

In order to work requests Python library must be installed in your Python execution env.

Opensearch permissions error

image image Set the ownership of the shuffle-database folder that the shuffle-opensearch container expects.

Bash

Editable

Response

Delete user

We recommend you to know what you're doing when you delete a user! It can have unpredictable consequences

  1. Find the ID of the user (Replace the username of your user in the query with the placeholder!)

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

  2. Find and take the "_id" value of your user from the returned!

    Code

    Editable

    Response

  3. Delete the user

    API Call

    DELETE
    cURL
    Python
    HTTP

    Response

    No response received.

Docker client version too new

If we run into "Failed to build: Error in Docker build: Error response from daemon: client version 1.51 is too new. Maximum supported APl version is 1.43" add the following environment variable into docker-compose under the backend section.

- DOCKER_API_VERSION=1.40

Recover admin user

If you find yourself in a situation where you have forgotten your passowrd and need a reset for your user, you can reset your lost password in your local instance by doing the following:

  1. docker exec to get bash session into OpenSearch container docker exec -it <container_id> bash

  2. Dump the results of users index query into users.log file

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

  3. open the users.log file with less and search for the admin user. Once found, scroll down to the apikey section. this value will be the api key of the admin user.

  4. I jumped onto another server within the same vlan as my Shuffle server but these could be ran on local host too. We will create a new user and update the user's role to admin with the Shuffle API.

  5. (Optional step): If you have multiple tenants, Change active tenant like this and repeat for each tenant:

    API Call

    POST
    cURL
    Python
    HTTP
    JSON

    Response

    No response received.

  6. Create a new user

    API Call

    POST
    cURL
    Python
    HTTP
    JSON

    Response

    No response received.

  7. Retrieve all the users and identify the user_id of the newly created user

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

  8. Assign the new user to the admin role.

    API Call

    PUT
    cURL
    Python
    HTTP
    JSON

    Response

    No response received.

  9. Log into webui with the new user, and you should now have admin rights.

Useful OpenSearch Queries

Find a user and their Tenants

API Call

GET
cURL
Python
HTTP

Response

No response received.

Find all tenant

API Call

GET
cURL
Python
HTTP

Response

No response received.

Find all tenant IDs

API Call

GET
cURL
Python
HTTP

Response

No response received.

OpenSearch Dashboards

If you prefer a visual interface over curl commands, you can add OpenSearch Dashboards to your on-prem Shuffle deployment. This gives you a UI to browse indexes, run queries, inspect documents, and monitor index health.

Adding to docker-compose.yml

Add this service to your existing docker-compose.yml, inside the services: block:

YAML

Editable

Response

Then start it:

Bash

Editable

Response

The Dashboards version must match your OpenSearch major version. Check yours with:

API Call

GET
cURL
Python
HTTP

Response

No response received.

Accessing the Dashboard

Open http://<your-server-ip>:5601 in your browser. Log in with your OpenSearch credentials (default: admin / the value of SHUFFLE_OPENSEARCH_PASSWORD in your .env file).

Using Dev Tools

Go to Dev Tools in the left sidebar to run queries directly against OpenSearch without needing curl. All the queries from the Useful OpenSearch Queries section work here.

List all Shuffle indexes:

Code

Editable

Response

Find a user:

Code

Editable

Response

Security Note

OpenSearch Dashboards exposes read/write access to your database. Do not expose port 5601 to the public internet. Access it via:

  • SSH tunnel: ssh -L 5601:localhost:5601 your-server
  • VPN
  • Firewall rules restricted to your IP

Extract all workflows

This procedure can help you extract workflows directly from OpenSearch even if the Backend and FrontEnd are in an awkward situation.

  1. Extract the index info from OpenSearch. NOTE: You may need to create a bind mount for the location where the workflows will be extracted to.

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

  2. Script to separate all workflows

    Python

    Editable

    Response

This script need to be run on the folder with the file workflows.json, it will create a workflows_loaded directory with all the workflows in it. This can also be very useful to either backup a copy your work or export it from a lab to a prod instance.

Moving from Docker in a VM to Kubernetes or SaaS

If you are running Shuffle in Docker inside a VM today, it is possible to move either to an on-prem Kubernetes deployment or to Shuffle SaaS. The important part is to migrate your data in a controlled way so that existing workflows, files and configuration are not lost.

Before you start

  1. Identify the data you want to preserve:
    • workflows
    • uploaded files
    • OpenSearch data
    • app-related persistent storage, if used
  2. Record your current Shuffle version and image tags.
  3. Verify where your persistent data is stored in docker-compose.yml, especially:
    • shuffle-database
    • shuffle-files
    • shuffle-apps
  4. Schedule downtime for the migration.

Important: Do not run the old Docker-based OpenSearch instance and the new Kubernetes OpenSearch instance against the same raw data path at the same time.

Step 1: Create a backup before migration

  1. Export workflows from OpenSearch as an application-level backup:

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

  2. Back up the persistent storage used by Shuffle:

    • OpenSearch data directory
    • files directory
    • apps directory, if used for hotloading or local app state
  3. Keep a copy of your current docker-compose.yml and .env file.

Step 2A: Move from Docker/VM to Kubernetes

This is the more direct migration path.

  1. Deploy a fresh Shuffle Kubernetes environment first. Use the Helm chart described in the Kubernetes documentation.
  2. Prepare persistent storage for the Kubernetes deployment:
    • create Persistent Volumes / Persistent Volume Claims for Shuffle data
    • ensure OpenSearch storage is dedicated to the Kubernetes deployment
  3. Stop the existing Docker-based Shuffle environment before migrating OpenSearch data.
  4. Move the persistent data to the target storage used by Kubernetes.
  5. Start the Kubernetes deployment and let it come up fully.
  6. Verify that the backend can connect to OpenSearch and that the UI loads correctly.
  7. Log in and verify:
    • workflows are present
    • uploaded files are available
    • app execution works
    • authentication and tenants are intact

Storage note: shuffle-database in Docker is a local volume or bind mount for OpenSearch data. Kubernetes does not reuse the Docker volume name directly. Instead, it uses Persistent Volumes / Persistent Volume Claims.

Step 2B: Move from Docker/VM to Shuffle SaaS

This path is possible, but it is usually handled as an export/import or support-assisted migration instead of reusing raw database files directly.

  1. Export the workflows you want to preserve.
  2. Export or identify any additional data that must be retained, such as files or configuration that cannot be recreated manually.
  3. Contact Shuffle support if you need help preserving more than workflow definitions.
  4. Import the workflows into the SaaS environment or follow the migration process provided by Shuffle support.
  5. Validate that:
    • workflows are visible
    • workflow logic is intact
    • app credentials and environment-specific references are updated where needed

Important: Raw OpenSearch storage from Docker should not be mounted directly into SaaS.

Step 3: Validate the migration

After moving to Kubernetes or SaaS, validate the following before decommissioning the old VM:

  1. Open several workflows and confirm they load correctly.
  2. Run a few existing workflows end-to-end.
  3. Verify that uploaded files are accessible where relevant.
  4. Check app authentication and secrets.
  5. Review backend and OpenSearch logs for errors.

Step 4: Decommission the old environment

  1. Keep the old VM powered off but available until validation is complete.
  2. Confirm that the new environment is stable for normal workload.
  3. Remove or archive the old environment only after the new setup has been verified.

Summary

  • Docker/VM to Kubernetes: usually the simplest way to preserve existing Shuffle data.
  • Docker/VM to SaaS: possible, but normally done through export/import or support assistance.
  • Existing workflows can usually be preserved if you back up and migrate the persistent data carefully.
  • Never let two different OpenSearch deployments write to the same raw data path at the same time.

Rebuilding an OpenSearch index

If you lost an index due to corruption or other causes, there is no easy way to handle it. Here's a workaround we have for certain scenarios. What you'll need: access to another Shuffle instance, OR someone willing to share. Lets do an example rebuilding the environments index. This assumes opensearch is on the same server.

  1. Cleanup the index

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

  2. Start refilling the index with info. For environments, make sure the "org_id" is correct according to the ID you can find in the /admin UI or the tenant index.

    API Call

    POST
    cURL
    Python
    HTTP
    JSON

    Response

    No response received.

  3. Check the index

    API Call

    GET
    cURL
    Python
    HTTP

    Response

    No response received.

OpenSearch consumes too much memory

After narrowing down your problem to opensearch is what is consuming your system resources you need to figure out why? Might be too many indices in OpenSearch or just a java heap size problem

  1. You'll need to list out all indices in elasticsearch opensearch.

    Bash

    Editable

    Response

  2. You could grep to narrow down on your search.

Bash

Editable

Response

  1. Once you see what's causing the problem in our case it was workflowexecution which was at 13 gb. We deleted it using the below command.

Bash

Editable

Response

Workflows are stuck

If you're on the default setup for shuffle and you start to notice that your workflows are getting stuck, It might be because you're running out of CPU needed to run the workflows.

To fix this, You will have to move towards setting up shuffle for production readiness in our configuration documentation.

Disclaimer

If you are doing this in a production server you will have to comb through the indices and delete them manually with respect to you tenants priorities, old executions and such.

  1. You should notice a reduction in memory consumption check this by running top. Do a docker-compose down then a docker-compose up -d for good measure and you are good to go.

  2. If the above steps do not fix the issue then this might mean its a java heap size issue, go into your docker-compose.yml file, move down till you locate the opensearch configurations and navigate to the OPENSEARCH_JAVA_OPTS settings and change them if initially they were running at 4 gb half that to 2 gb and save the file.

Code

Editable

Response

  1. Do a docker-compose down then a docker-compose up -d. You should notice a difference in the memory consumption by Opensearch

Updates failing

  1. After an update, click CTRL+SHIFT+R on your keyboard while in your browser. This runs a hard refresh without cache.
  2. Make sure you have the right version of Shuffle. Even if "nightly" is chosen, download them again with docker-compose pull or docker pull
  3. Ensure environment variables are defined properly for the misbehaving service.

Database not starting

In certain cases, you may experience OpenSearch continuously restarting. PS: All of these can be spotted in the logs. There are a few reasons for this which should be checked in the following order:

  1. Have you set vm.max_map_count=262144 setting?
  2. Did you change the folder ownership (1000:1000 by default)?
  3. Is the folder ownership a proper user (1000:1000) working?
  4. Is there enough RAM on the device?
  5. Is there enough storage space on the device?
  6. Do you have security enabled (https & username & password), but not configured it in the .env file?

OpenSearch TLS certificate setup reference

For full setup instructions for SHUFFLE_OPENSEARCH_CERTIFICATE_FILE (quick start + production hardening), see:

TLS Timeout and EOF Errors

In certain cases, especially when you're running in swarm mode (Make sure ports: 2377, 7946 and 4789 between your machines internally), you may experience timeouts, EOFs. Or maybe, in different cases a TLS timeout error, or a similar network request issue. This is most likely due to the network configuration of your Shuffle instances not matching the server it's running on.

The main configuration is "MTUs", AKA Maximum Transmission Unit. This has to match exactly - with the both the docker network driver bridge and shuffle_swarm_executions.

Find the MTU of your preferred network interface:

Code

Editable

Response

It is usually the network interface in the second line. Get it's MTU!

To set the MTU in Docker, do it in the docker-compose, in the networking section. Say the MTU you found was 1460, then use 1460, as can be seen below.

Code

Editable

Response

Next, if you're running on swarm mode, delete the existing shuffle_swarm_executions network if it already exists. You can do that by using:

Bash

Editable

Response

This might be an essential step to enforce what we did in the last step. Shuffle things a lot of things under the hood and syncing up the right interfaces is one of them so that you don't have to worry about it.

If it requires removing dependant services, proceed to do that.

When done, restart the docker-compose. Now the issue should be automatically taken care of. If not, and you're on swarm mode, Proceed to the next step of manually setting the network MTU:

We need to make a network named the same as the environment SHUFFLE_SWARM_NETWORK_NAME for Orborus (default: shuffle_swarm_executions):

Bash

Editable

Response

If the issue still persists, Please look into changing the environment variable SHUFFLE_SWARM_BRIDGE_DEFAULT_INTERFACE. Shuffle takes care of syncing the docker0 bridge interface to the preferred interface of the container. Changing this value might help docker sync up things better. We assume that the interface name is "eth0" by default, which is the default setting.

If none of this works, Often times it's simply because of the virtualisation used by your cloud provider. For example, We have found these issues to be persistent with providers using VMware underneath, Refer to this for a fix

Shuffle on ARM

ARM is supported on Shuffle since 1.3.0!

Permission denied on files

In certain scenarios, permissions inside and outside a container may be different. This has a lot of causes, and we'll try to help figure them out below. Thankfully most fixes are relatively simple. To test this try to go to /admin?tab=files in Shuffle, and upload a file. If the file is uploaded and it says status "active", all is good. If it's not being uploaded, then it's most likely a permission issue.

23 03 2022_22 51 18_REC

Fix 1: share permissions.

In the docker-compose.yml file, find the "shuffle-files" volume mounted for the backend service. Simply add a ":z" on the end of it like so:

Code

Editable

Response

Then restart the docker-compose (down & up -d), and try to upload a file again.

Fix 2: Selinux problems

Disable Selinux to test. This should take immediate effect (run as root).

Code

Editable

Response

After, try to upload a file again

See permissions on the inside of the container

To find the folder permissions inside the container

Bash

Editable

Response

Docker Permission denied

In certain scenarios or environments, you may find the docker socket to not have the right permissions. To work around this, we've built support for the docker socket proxy, which will give the containers the same permissions. Another good reason to use the docker socket proxy is to control the docker permissions required.

To use the docker socket proxy, add the following to your docker-compose.yml as a service:

Code

Editable

Response

When done, remove the "/var/run/docker.sock" volume from the backend and orborus services in the docker-compose. These containers should route their docker traffic through this proxy. To enable the docker rerouting, add this environment variable to both of them:

Code

Editable

Response

This will route all docker traffic through the docker-socket-proxy giving you granular access to each API.

PS: Adding :z to the end of the volume may fix this issue as well.

Server is slow

If the server Shuffle is running on is slow, it's likely due to the same constraints of any other server. One of these are typically the culprit:

  • Disk space
  • CPU
  • RAM

The normal reason this happens is due to too many processes running concurrently in Docker (too many containers). To look at ideal configurations, look at production readiness in our configuration documentation.


First we check CPU. This is can be done using the "top" command.

Code

Editable

Response

The typical near the top is something like this. If the CPU usage is too high (see line three '%Cpu(s): 11.0 us'" - this means 11% is used total), you've most likely not configured Shuffle to run with the appropriate amount of containers as a maximum, with bad cleanup routines (CLEANUP=true).

Code

Editable

Response

Fix: Stop docker containers and reduce the amount that are allowed to run. If everything is TOO slow, reboot the server and stop all containers when it's started back up:

Bash

Editable

Response


Next up is RAM. This is can also be done using the "top" command.

Code

Editable

Response

As with CPU, the information is near the top of your screen and looks something like this. If the RAM usage is too high (see line four 'KiB Mem : 8008956 total, 142236 free'" - this means that almost no memory is left on the device). This is a typical problem if you've enabled app log forwarding into Shuffle. To disable log forwarding, add the environment "SHUFFLE_LOGS_DISABLED=true" to Orborus, then bring it down and back up again.

Code

Editable

Response

Fix: Stop docker containers and reduce the amount that are allowed to run. If everything is TOO slow, reboot the server and stop all containers when it's started back up:

Bash

Editable

Response


Next up is disk space - can Shuffle save anything? See whether there is space on the machine in the location Shuffle is running

Code

Editable

Response

To get more space, either delete some files, clean up the Opensearch instance or add more disk space.

How to handle wrong or bad images on old versions of docker.

WhatsApp Image 2022-04-06 at 11 06 20 AM

Download the correct app version from shuffle cloud

Once downloaded upload it on your onprem shuffle instance by dragging and dropping it on the activated app list

upload app

Once done check the server for misp images present

list images

You should see previous existing images and the newly added apps image

The last step is to refer the target image to the source image that you uploaded

You do this by using the docker tag command see more information here (https://docs.docker.com/engine/reference/commandline/tag/) docker tag frikky/shuffle:misp_1.0.0 davvyshuffle/shuffle:MISP-e72b9e9c5b0a40753e184c8ce0ba6c2b i.e docker tag source_image:{TAG} target_image:{TAG}.

docker tag

Go back to your shuffle interface and your app should run success.

If you intend on uploading the app in a remote server you could push the app image onto docker hub using the docker push command more info here(https://docs.docker.com/engine/reference/commandline/image_push/)

Sign up on docker here (https://login.docker.com/u/login/) then push the intended image into your docker hub repository, from your server's cli. You might be prompted to enter your password, do so and your image will be uploaded successfully.

docker push

docker hub image

From your remote server cli pull the image from your docker hub repository. for more info about docker image pull see here (https://docs.docker.com/engine/reference/commandline/pull/)

docker pull

Once this is done you have to tag the existing images of this app to the working app you just downloaded from your docker hub repo.

docker tag

Docker not working

In certain cases, Docker may not be working due to too large an amount of containers running, and Docker not being able to keep up. The cause of this is typically Orborus starting too many workflows in unison. To fix this, either reduce the amount of containers able to run, or set up swarm mode (paid).

This can be controlled by the environment variables:

Code

Editable

Response

Then manually clean up the containers:

Code

Editable

Response

PS: You may need to use "systemctl stop docker" instead of using "service".

Now restart the Shuffle stack again, and all the containers should be gone

Troubleshooting for executions not running in swarm mode

  • You'll need to check whether swarm is configured properly and running. Do this by orborus logs using the following command.

Bash

Editable

Response

  • You should get results similar to the image below, if not check here

Inkedswarm2_LI

  • Once you've assertained that swarm mode is running as it should, go ahead and list services running in swarm using the below command.

Bash

Editable

Response

  • You should get results similar to the image below, the REPLICAS column should be AT LEAST 1 replica created for each running service. i.e A 0/1 means that the app isn't available to handle the executions that you need it to, so you'll have to reinstall the app, see here for more information on how to do this.

Inkedswarm1_LI

  • Once you've ensured that all services have replicas created and are running, move on to the next step.

  • Check logs for the worker. First, run the below command to get a list of tasks that are running

Bash

Editable

Response

Inkedswarm3_LI

  • Get the name of the worker and run the following command. If not, this means you will have to go on local of each node to get logs of workers running on it or using a docker management tool.

Bash

Editable

Response

  • All you have to do now is comb the logs and identify where the problem is, if you can't figure out where the problem is reach out to our community on discord.

Swarm init advertise address error manual fix

If you see Orborus logs like this:

Code

Editable

Response

it usually means Docker can not choose a unique host IP automatically.

Use a host NIC IP (not a container IP) and pin it:

  1. Find your host IP (example output: 10.0.10.25):

Bash

Editable

Response

  1. If swarm is partially initialized, reset it first:

Bash

Editable

Response

  1. Initialize swarm with the same IP:

Bash

Editable

Response

  1. Restart Orborus / stack.

Find code OpenAPI app

You may have had problems with an app and need some help getting it fixed. Apps created in the app creator of Shuffle also do generate underlying Python code utilizing the same capabilities as if you make a Python function from scratch. Here's how to find the code for a function.

  1. Decide on an app to find. In our example, let's use "DefectDojo". Start by finding the Docker image:

Bash

Editable

Response

  1. From the list above, you should see an image called "Defectdojo". The next step is to run it and extract the code. Copy the name of it, then type this:

Bash

Editable

Response

  1. You should now have a file called "app.py" locally. Now lets say the function you're having trouble with is called "importscan create". In the app.py file we just created, go into it, and search for that exact name. All spaces ( ) and dashes (-) have been turned into underscores (_), and all text is LOWERCASE. The line should start with something akin to:

Python

Editable

Response

Copy everything indented under this function and sent to support@shuffler.io for further help!

Tenants/Suborgs seem to be lost

If you loose your tenants/suborgs for any reason at all and you need to reinstate them then you will need to do the following;

  1. Get the tenant id's from your backend logs. It's a UUID in the format 550e8400-e29b-41d4-a716-446655440000.

Bash

Editable

Response

  1. After you've identified the tenant id's for the tenants you want to reinstate, docker exec to get bash session into OpenSearch container or any other container so long as the said container can communicate to the Opensearch container

Bash

Editable

Response

  1. Once in the container from step 2 above, run the following command.

API Call

POST
cURL
Python
HTTP
JSON

Response

No response received.

  1. Exit out of the container

  2. Restart docker

Code

Editable

Response

Go back to your shuffle instance and you should see the tenant in question reinstated in the tenants tab.

Find top index items opensearch

As the Opensearch index may fill up over time, it is important to be able to debug the the available indexes. One particular issue we have had has been that it takes >60 seconds to load apps onprem at times. Here is how to resolve them.

Get into the Opensearch container of port 9200 is not exposed by default:

Bash

Editable

Response

Get the indexes and look at them:

API Call

GET
cURL
Python
HTTP

Response

No response received.

Find the largest items in the workflowapp index (for apps in shuffle):

API Call

GET
cURL
Python
HTTP

Response

No response received.

Delete an index if it's too large (normal ones to delete if problems: workflowexecution, workflowqueue-shuffle, environment_stats)

API Call

GET
cURL
Python
HTTP

Response

No response received.

Add user to lost tenant

If you have lost access to Shuffle, it usually due to an unforeseen disconnect to the Database during startup, leading to more Tenants being added. To fix this, your user needs to be re-added to the original Tenant

Find the Tenant and User Id of your account. They are in the UUID format 550e8400-e29b-41d4-a716-446655440000

Bash

Editable

Response

Get into the Opensearch container of port 9200 is not exposed by default:

Bash

Editable

Response

Update the USERID and ORGID, ORGNAME fields, then run this command to re-add your account to the right tenant

API Call

POST
cURL
Python
HTTP
JSON

Response

No response received.

No module named xyz in execute python Shuffle Tools

If you want to install a custom module like pandas (although, 'execute python' isn't made for heavy processing. We recommend making a custom python app for it).

Important: Shuffle automatically pulls and updates the Shuffle Tools image on startup, which may overwrite your custom-built image. To prevent this, set the environment variable SHUFFLE_AUTO_IMAGE_DOWNLOAD=false in your Orborus/worker configuration before building a custom image.

There are two primary ways to it:

  1. Learn about dynamic library loading in your code (not recommended)
  2. Build the Shuffle Tools app again locally and add the libraries you want to it! Library configuration is over in this requirements.txt file

To build this app again, I would:

  • Add in the requirement in requirements.txt linked above.
  • Rebuild the app again by doing:

Bash

Editable

Response

Changing timestamps for an index from Unix to timestamps

This may be necessary if you want to do additional modifications or exploration of an existing Index. See this Issue on how it was fixed by a Shuffle user.