Core VM is full again


#1

*1. Describe the problem. Are there any error messages you can paste? Can you provide a screenshot of the problem (with sensitive information removed)?

I am sorry to open this issue again. My core memory is full again.

I made sure to install miniconda3 on a volume so that the core memory is spared from getting full.

2. What is the name of the VM or volume that the problem is affecting?

My IP address is ubuntu@137.205.69.13](mailto:ubuntu@137.205.69.13) and its the Kraken VM.

3. Where is the VM or volume hosted; Birmingham, Cardiff, Swansea or Warwick?
It is hosted in warwick

Could you please help me with this

Thank you!

Anu


#2

Dear Anu,

I would suggest storing all of your data on a mountable volume rather than on the default boot volume. The mountable volumes can be sufficiently large in order to deal with your analysis. Details on how to create and attach a volume can be found here - Creating and attaching volumes using Horizon.

All the best,
Sion


#3

yes thank you for the suggestion. I do this every time. However, now it is now so full, I cannot even open .bashrc to check my settings.

When I do du -h -s *, this is what I get
4.0K Anu_new

4.0K Anu_new.pub

4.0K basespace

4.0K bin

4.0K Desktop

4.0K Documents

16K Downloads

4.0K id_rsa.pub

4.0K key_Anu.pem.pub

4.0K mpallen

4.0K Music

4.0K nohup.out

4.0K Pictures

224K Public

4.0K Templates

4.0K Videos

du: cannot read directory ‘volume_5/lost+found’: Permission denied

121G volume_5

4.0K volume_Anuradhr

Can you please help me in freeing up some space so that I can access the basic files.

Thanks!

Anu


#4

Hi Anu,

You are going to need to delete some files from the boot volume in order to have sufficient memory available to do anything. If you have any data on there you can do without then I would copy it to a large mounted volume.

If conda is the issue and it is installed on a separate, non-boot, volume (volume_5?) then that might be the problem. If that volume is full then I would recommend making a larger volume 1-2Tb and reinstalling conda on said volume.

All the best,
Sion


#5

I am sorry for not able to understand. I dont know what files to delete. From the du -h -s *, there are no files that are greater than 1 Mb. I am not sure what is taking up all this space. How can I check the file sizes of everything in the boot volume?

I have 5 volumes attached to this VM. The miniconda is found in volume2 which is located in /mnt/volume2. The volume5 is a volume attached, do you think if this is taking up all the space?

I had a problem with my boot volume expanding and Maciej could log into my VM and identified miniconda as the problem. This problem was rectified by moving miniconda to volume2.

Is it possible for you to log into this Vm and help me fix this.

Thank you

Anu


SSH connection refused- Warwick
#6

Hi Anu,

Your core VM is completely full and sshd will now refuse to start. It looks like you rebooted your instance and it has failed to reboot, as it is entirely full. You will need to make a new clean VM and transfer your volumes over to it. This should resolve the issue you are having in this thread - Core VM is full again.

All the best,
Sio