-
Notifications
You must be signed in to change notification settings - Fork 161
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Exception thrown on Organisation creation #239
Comments
Hey, thanks for the report. Some follow up questions and comments, The summary seems like there's an issue with your database. It either didn't migrate to the latest schema correct or somehow it was reverted.
FYI Organizations have no practical function in the system today. They're just a random attribute on a users profile. If you don't have meaningful data in the system I recommend just deleting your volumes and starting over from scratch. If this is happening on first stand up with fresh volumes, it's likely related to the deployment method. The system loads the "GoldStandard" you saw in the logs and then migrates it up to current. There's a chance it's loading correctly and some migration is failing, but the system still comes up and somewhat works. If you have data you need to recover, that would require some offline discussion. Feel free to email the team at, vectrops@sra.io to discuss. We're more than happy to see community members getting VECTR running on any flavor of k8s. Our "supported platforms" is more due to community support burden and functionality validation than a lack of compatibility. If you'd like to share your work feel free to start a topic under "Discussions" with examples of your Helm chart and such for other community members to engage with. If you don't feel comfortable sharing it publicly you can shoot it to our email above so we can consider rolling it into our public documentation. |
Hey @SRAPSpencer, thank you for your speedy response! Answering your questions:
Yes it is. I can't progress beyond the organisation creation page.
Yep.
No. I'm running this locally for testing purposes before I migrate to AWS EKS (in which case I'll be using EFS volumes for database storage - which persist). Since you noted that it appears to be a database issue, I did notice a few mongo exceptions in the logs that occur every time I run docker compose, before the webapp launches. Here they are:
I took your advice and deleted all volumes and then relaunched the app, and unfortunately still received the same original exception in this issue (as well as the ones above). I also noticed this error right before the exception.
|
Yeah, this looks like a environment specific issue. Something prevents the mongorestore from running, which is step 0 in setting up the database for first use. If it can't deploy the database there's no way to workaround that. This is where other container runtimes and platforms get tricky. The "java.io.ioexception error=0" is a very generic error, it couldn't spawn the process with no indication as to why. This is a non-exhaustive list of the potential issues. Could be file permissions with the container runtime not reading mongorestore or being able to execute it. Could be a run as issue, our containers have a VECTR user they run as. See details here, https://docs.vectr.io/upgrading/non-root-migration/faq/ Could be something specific to the Rosetta translation layer causing the process error. We ship VECTR as x86-64 so running on an ARM Mac wouldn't be translating or emulating in some manner. If I had to guess it's the "run as" causing issues. K8s, namespaces and such can be a challenge for some containers. I can leave the issue open if anyone else in the community wants to comment, but this goes beyond what we're able to troubleshoot for the community. |
This makes sense @SRAPSpencer, thank you for your help. I'll play around with run as and see what I can do. I'm not all that familiar with mongo so thanks for putting it plainly. No worries at all, I'll close this issue! |
I'll link this back to our feature request for general k8s support, @kimberleyhallifax feel free to reach out to vectrops@sra.io if you have time to share some of your requirements for deployment and environments. We are actively seeking detailed feedback on the matter as we plan for the future. |
Hey @SRAPSpencer, so I gave your advice a go and tried using This is the Dockerfile I used:
As you can see, I changed ownership of all files used in execution (including the mongodb files) to |
This looks like an ARM issue that occurs with Java applications. It looks like there are a few possible workarounds like changing the java_opts spawn method to vfork. |
Omg that worked @thebleucheese! Thank you so much! I was wracking my brain with this for days. Much much appreciated!! |
Hey all
So I just want to preface this by saying that I have read your documentation on operating systems such as macOS and kubernetes not being supported (which is indeed what I am running on here), but I'm raising this bug issue in confidence that this exception isn't related to my choice of OS (although I haven't attempted this on any other OS to verify that). But if I'm wrong and it is, please let me know.
Describe the bug
I'm running vectr locally on my macOS and whenever I attempt to create an organisation after logging in, I get this exception.
To Reproduce
You should be running on macOS.
wget https://github.com/SecurityRiskAdvisors/VECTR/releases/download/ce-8.8.1/sra-vectr-runtime-8.8.1-ce.zip`
.env
file with the following:https://localhost:8081/
and entering the default root credentialsExpected behavior
I expected an organisation to be created.
Desktop (please complete the following information):
I'm running macOS Monterey 12.2.1 on Apple M1 Pro.
Additional context
I have also received this exact same error when I've run vectr locally on minikube (kubernetes). I basically converted the docker compose file to a helm chart, reproduced the same steps, and received the same error at the same point. But I'm just raising the issue in relation to the docker compose method because there are less files involved.
I did find some additional logs printed during startup, relating to this GoldStandard (if this is of any help):
Here is my
docker-compose.yml
file:And here is my
.env
file:Thank you in advance for your help!
The text was updated successfully, but these errors were encountered: