Replies: 3 comments 3 replies
-
|
I am a little confused about where exactly you are running your commands, you mention in step 2 that you "entered the postgres container" but I'm not 100% sure what that means. You then mention manage commands you are running in steps 3 and 4 but are not entirely clear on where they are being run. In general, when using docker you want to run commands as It seems like the state of your database right now is pretty broken, so I would recommend you delete all the docker postgres databases by clearing out any data in the mounted volume for that pod and start again cleanly. Then I would recommend you make sure you have a working empty seqr instance. Run
Keep the output from each step and let me know if any of them fail Also, how old exactly is the version of the manual instance you are running? If it is pre 2019 and has all the old xbrowse models you will need to locally transfer to the updated schema before you can switch to docker compose |
Beta Was this translation helpful? Give feedback.
-
|
Per my previous comment:
If this is the case you need to migrate your current, non-docker seqr to use the new schema first, and then dump that migrated db and prceed with the steps to transfer to the docker compose seqr. Instructions can be found here, and you will need to check out the secific commit that they are found at (3f85f1f) : https://github.com/broadinstitute/seqr/blob/3f85f1f226e374c4dff50f3f642991ef7aaa6065/deploy/MIGRATE.md |
Beta Was this translation helpful? Give feedback.
-
|
Hi, Regards, |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Hello!
We've got an old seqr instance that was manually installed back in 2017 that I'm interested in migrating to the newer version using the seqr docker-compose. I've tried a couple of things based off this issue from 2020:
pg_dump -U postgres seqrdb | gzip -c - > backup.gzdocker-compose up -d seqr) and entered the postgres container, and ran these commands to drop the database and replace it with my backup:This all seems to run fine.
I then do:
3.
python -m manage makemigrationsWhich results in 'no changes detected', I assume because the migrations folder of each app (like seqr/migrations) is full of old migrations that were uploaded to the github repo.
4.
python -m manage migrateAnd at this point it finally errors out:
I have tried clearing out the migrations folders (by removing the 00*.py and pycache files and folders) for all the apps (seqr, matchmaker, panelapp, reference_data), and trying the last two commands again:
Which seems to create new migrations, followed by:
Which errors out again. At this point I'm a little out of my depth, so I was wondering whether you'd have any help or other instructions for porting an old non-docker instance of seqr to a newer docker instance of it.
Thanks in advance!
Chris
Beta Was this translation helpful? Give feedback.
All reactions