4

I have a 150GB /docker partition mounted on a RHEL 8 VM and I have ran this command as root and also with a local user that has sudo:

export TMPDIR=/docker

to make Podman change its default tmpdir. However, upon import of a 54GB .tar we run into this error below. What config setting am I missing?

$  podman import dockerfile.tar
Getting image source signatures
Copying blob b45265b317a7 done
Error: writing blob: adding layer with blob "sha256:b45265b317a7897670ff015b177bac7b9d5037b3cfb490d3567da959c7e2cf71": Error processing tar file(exit status 1): write /a65be6ac39ddadfec332b73d772c49d5f1b4fffbe7a3a419d00fd58fcb4bb757/layer.tar: no space left on device

Here are my partitions on this VM:

$ df -h
Filesystem                                  Size  Used Avail Use% Mounted on
devtmpfs                                    3.9G     0  3.9G   0% /dev
tmpfs                                       3.9G   84K  3.9G   1% /dev/shm
tmpfs                                       3.9G   51M  3.8G   2% /run
tmpfs                                       3.9G     0  3.9G   0% /sys/fs/cgroup
/dev/mapper/rhel_rhel86--svr-root            38G  7.2G   31G  20% /
/dev/sda2                                   495M  276M  220M  56% /boot
/dev/mapper/rhel_rhel86--svr-var             33G  1.4G   32G   5% /var
/dev/mapper/rhel_rhel86--svr-tmp            4.7G   66M  4.6G   2% /tmp
/dev/mapper/rhel_rhel86--svr-home            43G  1.6G   42G   4% /home
/dev/mapper/rhel_rhel86--svr-var_log        4.7G  104M  4.6G   3% /var/log
/dev/sdb                                    147G   56G   84G  40% /docker
/dev/mapper/rhel_rhel86--svr-var_log_audit  9.4G  168M  9.2G   2% /var/log/audit
/dev/mapper/rhel_rhel86--svr-var_tmp        1.9G   47M  1.9G   3% /var/tmp
/dev/sda1                                   500M  5.9M  494M   2% /boot/efi
tmpfs                                       785M  8.0K  785M   1% /run/user/42
tmpfs                                       785M     0  785M   0% /run/user/1000

Heres part of /etc/containers/storage.conf

# Default Storage Driver, Must be set for proper operation.
driver = "overlay"

Temporary storage location


runroot = "/docker"


Primary Read/Write location of container storage


When changing the graphroot location on an SELINUX system, you must


ensure  the labeling matches the default locations labels with the


following commands:


semanage fcontext -a -e /var/lib/containers/storage /NEWSTORAGEPATH


restorecon -R -v /NEWSTORAGEPATH


graphroot = "/docker"

This VM is running SELinux, so I ran the two commands and rebooted:

semanage fcontext -a -e /var/lib/containers/storage /docker
restorecon -R -v /docker

I was able to successfully import two other .tar files with Podman that were around 800MB, so I know Podman is working correctly. Any ideas why I'm running out of space even though the tmpdir is set to /docker? Thanks!

BMitch
  • 2,302
  • 11
  • 15
808mrb
  • 43

1 Answers1

3

The output of your command has the answer. The file it is trying to write is "/a65be6ac39ddadfec332b73d772c49d5f1b4fffbe7a3a419d00fd58fcb4bb757/layer.tar"

The podman command is trying to write a 54GB file to the / partition which is only 38GB with 31GB available as indicated by your df -h command.

Even though you have /docker as the tmpdir and it has enough space, the / partition does not have enough space. Either increase the size of the / partition, or figure out a way to write the files to /docker since the tmpdir command isn't doing it and neither are the config settings you have in your storage.conf file

Dan H
  • 31