Failed Units: 1 rm-assets.service

What keywords did you search in tectonic-installer issues before filing this one?

Failed Units: 1
rm-assets.service

Is this a BUG REPORT or FEATURE REQUEST?

BUG REPORT

Versions

• Tectonic version (release or commit hash): 1.9.6-tectonic.1 or commit e23765f
• Terraform version (terraform version): v0.11.10
• Platform (aws|azure|openstack|metal): digitalocean

What happened?

Failed clean up install assets from S3

sudo systemctl status rm-assets.service
● rm-assets.service - Clean up install assets from S3
   Loaded: loaded (/etc/systemd/system/rm-assets.service; disabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Mon 2018-11-19 18:39:26 UTC; 23min ago
  Process: 11058 ExecStartPre=/usr/bin/bash /opt/rm-assets.sh (code=exited, status=1/FAILURE)

Nov 19 18:39:19 kubernetes-master-0 bash[11058]: + cp /tmp/assets.zip /tmp
Nov 19 18:39:19 kubernetes-master-0 bash[11058]: cp: '/tmp/assets.zip' and '/tmp/assets.zip' are the same file
Nov 19 18:39:19 kubernetes-master-0 bash[11058]: + docker pull aknudsen/do-spaces-tool:v0.2.1
Nov 19 18:39:24 kubernetes-master-0 bash[11058]: + docker run -t --net=host -e ACCESS_KEY_ID -e SECRET_ACCESS_KEY -e REGION -v /tmp:/spaces aknudsen/do-spaces-tool:v0.2.1 upload /spaces/assets.zip akuber>
Nov 19 18:39:25 kubernetes-master-0 bash[11058]: Uploading file /spaces/assets.zip...
Nov 19 18:39:26 kubernetes-master-0 bash[11058]: + rm -f /tmp/assets.zip
Nov 19 18:39:26 kubernetes-master-0 bash[11058]: rm: cannot remove '/var/tmp/tectonic.zip': No such file or directory
Nov 19 18:39:26 kubernetes-master-0 systemd[1]: rm-assets.service: Control process exited, code=exited status=1
Nov 19 18:39:26 kubernetes-master-0 systemd[1]: rm-assets.service: Failed with result 'exit-code'.
Nov 19 18:39:26 kubernetes-master-0 systemd[1]: Failed to start Clean up install assets from S3.

What you expected to happen?

The cluster work normally, but i think this is should installed without error

How to reproduce it (as minimally and precisely as possible)?

I'm not really sure what happened in this process.

maybe rm /var/tmp/tectonic.zip in rm-assets.sh not necessary

#!/bin/bash
set -e
set -o pipefail

# Instead of deleting object, just overwrite it for simplicity's sake
touch /tmp/assets.zip

# shellcheck disable=SC2086,SC2154,SC2016
/opt/do-pusher.sh /tmp/assets.zip akubernetes-9ee6cea24a83dde81763355993b43b7a/assets.zip
rm /var/tmp/tectonic.zip                 <== this is not necessary?

/var/tmp/tectonic.zip already removed in init-assets.sh

#!/bin/bash
set -e
set -o pipefail

# Download the assets from Spaces

# shellcheck disable=SC2154,SC2086
/opt/do-puller.sh ${spaces_bucket}/assets.zip /var/tmp/tectonic.zip

# shellcheck disable=SC2154,SC2086
/opt/do-puller.sh ${spaces_bucket}/kubeconfig /etc/kubernetes/kubeconfig
unzip -o -d /var/tmp/tectonic/ /var/tmp/tectonic.zip
rm /var/tmp/tectonic.zip                 <== removed
# make files in /opt/tectonic available atomically
mv /var/tmp/tectonic /opt/tectonic

exit 0

Anything else we need to know?

Just reporting a small issue