I have created a new volume with more space to store my data files. I want to use some of the data in Galaxy but there are more files than space in the default Galaxy volume. Is it possible to move the default Galaxy file storage to the larger volume?
Cheers,
Donna
Hi Donna,
This is definitely possible, but I need to preface this with a PLEASE BE CAREFUL warning, because if you do something wrong here, it could cause data loss.
Rather than move the Galaxy filesystem, what I recommend you do is mount a new, larger volume where Galaxy stores its user’s files. In order to make the transition seamless, we’ll move Galaxy’s current user files onto the new volume as well.
Make sure you have a local copy of all of the data that you have in Galaxy!
- Follow the tutorial here, upto the point where you have a new volume mounted at a directory within your home directory. The final steps are here, and I’ll use the same mountpoint here as in the tutorial.
sudo mount /dev/vdb1 /home/ubuntu/example
sudo chown ubuntu:ubuntu /home/ubuntu/example
- Copy the Galaxy userfiles onto the new volume. I’m using
sudo rsync
to preserve permissions
sudo rsync -avhP /mnt/galaxy/files/* /home/ubuntu/example
- Now that we have a copy of the Galaxy files on the new volume, unmount it from
/home/ubuntu/example
and remount it at /mnt/galaxy/files
sudo umount /dev/vdb1
sudo mount /dev/vdb1 /mnt/galaxy/files
- Change ownership of
/mnt/galaxy/files
so that the galaxy user can read/write to it
sudo chown galaxy:galaxy /mnt/galaxy/files
- Go to
http://<ip_address>/cloud/root/admin
and restart galaxy with the button in the middle of the page on the right-hand side
This is what df -h
looks like on my test instance, with the volume mounted at the correct point (see /dev/vdb
):
Filesystem Size Used Avail Use% Mounted on
udev 16G 0 16G 0% /dev
tmpfs 3.2G 14M 3.2G 1% /run
/dev/vda1 117G 29G 88G 25% /
tmpfs 16G 104K 16G 1% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 16G 0 16G 0% /sys/fs/cgroup
tmpfs 3.2G 24K 3.2G 1% /run/user/119
tmpfs 3.2G 0 3.2G 0% /run/user/1000
tmpfs 3.2G 0 3.2G 0% /run/user/121
/dev/vdb 100G 33M 100G 1% /mnt/galaxy/files
tmpfs 3.2G 0 3.2G 0% /run/user/1001
REMEMBER TO REMOUNT THIS VOLUME AND RESTART GALAXY WHEN YOU REBOOT YOUR INSTANCE!
I’ve been having some issues with this this week, i can add the data to the extra volume no problem and then galaxy can read and write to the volume but during some of the analyses it is also putting stuff onto the original volume and filling it up, (i mounted the new volume, synced it and added the ownership as you outlined) and then won’t run anything else through galaxy. I’m not really sure where i’m going wrong. Any suggestions?
Edit: It looks like galaxy is filling up the original volume by putting stuff in the mnt/galaxy/tmp/job_working_directory. Is there a way i can get it to put these files on the larger volume too?
Thanks.
Hi Donna - apologies for this being more difficult than we anticipated!
You could mount a volume at /mnt/galaxy/tmp
, owned by galaxy:galaxy
, or consider deleting some intermediate files from your Galaxy history using the gear icon on the right-hand side of the galaxy page.
We could also look at mounting a big volume at /mnt
, but because you’ve got lots of data on this instance, I’d be happier helping you with this on a new GVL instance.
I had a go at mounting the working jobs file on my other volume and it all looks to be working ok (touch wood).
Thanks for all your help!
1 Like