I am running the community edition from AWS marketplace without our aws account. This is a POC/test account so over the weekend (cost savings) I stopped the master coordinator EC2 instance from the console. This morning I resumed the instance and went through quite a few hoops to restart the service, once I did, it basically started by install fresh (no lake connections, no spaces, no saved queries, etc.).
I am curious about a few things:
a. why does this happen? Is there a way to stop the master coordinator without this happening?
b. I moved the mounted files to the efs (sudo cp -r /mnt/c1 /var/dremio_efs/c1/) and still was unable to restore my work–how and where was it lost to?
Is anyone out there facing something similar? What if you are forced to restart the instance or there is a hardware failure?
Thanks in advance; Let me know what else you are in need of as far as details
@whit Just restarting the project (coordinator) should not reset anything. What are your sources? Like is it S3 and PARQUET? The EFS volume should have prior server.log, are you able to send those?
So is it valid to stop the coordinator instance, not terminate but stop via the aws console? Interesting as when I started it back up the service did not start, which is what took me down this path. My sources are s3 and parquet, yes
I ended up deleting the Cloud Formation and re-launching. From there I found my original project, great. I then clicked the Stop Project in the dremio UI, which terminated all of the instances. From there, how do you reopen your existing project? Do you have to re-launch the CloudFormation each time?
@whit YOu had mentioned “when I started it back up the service did not start,”, do we know why? Also, have you followed instructions below and when it came up, you lost everything?