Install taking several days to run, still not completed, possibly looping

Bug Report Template

Description

I tried deploying a new server using the bash script below. Variables were set but not included in this post. Upon checking the server a few hours later, I found the bootstrapper was still running. Two days later, the bootstrapper is still running, with the apnscp-bootstrapper.log file at 282 MB.

Summary
# Set up hostname
hostnamectl set-hostname $HOSTNAME
echo "127.0.0.1 $HOSTNAME" >> /etc/hosts

# Set up basics
yum upgrade -y;
yum install epel-release -y;
yum install p7zip p7zip-plugins -y;
yum install wget curl unzip -y;

# Set up ApisCP
echo "curl https://raw.githubusercontent.com/apisnetworks/apiscp-bootstrapper/master/bootstrap.sh | bash -s - -s use_robust_dns='true' -s always_permit_panel_login='true' -s php_install_ioncube='true' -s dns_default_provider='null' -s apnscp_admin_user='$ADMIN_USER' -s apnscp_admin_email='$ADMIN_EMAIL' -s system_hostname='$HOSTNAME' -s whitelist_ip='$WHITELISTED_IP_CIDR'" > ~/apiscp_setup.sh
bash ~/apiscp_setup.sh

Steps to Reproduce

Run script above in details

Expected Behavior

Completed install in an hour

Actual Behavior

Continuously running, possibly looping

Environment

ApisCP version:

revision: 32cf2c23d67e304c3812262abd9c12a90112206c
timestamp: 1751907369
ver_maj: 3
ver_min: 2
ver_patch: 0
ver_pre: 32cf2c23d
dirty: false
debug: false

Operating System:
4.18.0-553.58.1.el8_10.x86_64

Additional relevant information (e.g., PHP version, database, etc.):
Clean install

Additional Information

Using Linode, 4 GB dedicated server, 80 GB storage, two dedicated CPU cores, Rocky 8.

I can attach the log file but it’s huge. Here’s the last 51k lines of the log…

2025-07-09 logs.txt (6.0 MB)

1 Like

Thanks for the bug report. Linked back to calling dbus_shutdown() preventing proper shutdown at end.

Run upcp -m '2025_07_09*' to apply the new migration that rebuilds dbus.so.

Thank you. I ran the command and it completed. It appears the log is still going, though.

It sent me an email that it was complete, but there are tons of accounts… something’s still not right.

Yes, those test accounts failed to cleanup because D-Bus was generating a segfault upon shutdown.

This should clean them up:

cd /home/virtual
DeleteDomain --force apiscp-int-*

Got it. Thanks for your help!