Opsman Support Bundle Fails to Generate

Products

• VMware Tanzu Platform Core

Issue/Introduction

Opsman shows "Failed" after some time when trying to generate a support bundle. 

You see a "No space left on device" error in /var/log/production.log:

E, [2026-04-20T16:20:36.722385 #1294] ERROR -- : 2026-04-20T16:20:36+0000: [Worker(delayed_job host:opsmanager-example pid:1294)] Job CreateSupportBundleJob [########-####-####-####-############] from DelayedJob(default) with arguments: [6] (id=378) (queue=default) FAILED (1 prior attempts) with Errno::ENOSPC: No space left on device - copy_file_range

df -h shows plenty of space on the device:

ubuntu@opsmanager-example:~$ df -h /
Filesystem      Size  Used Avail Use% Mounted on
/dev/sda2       158G   73G   79G  49% /

Environment

Operations Manager

Cause

This can be caused by /var/log/lastlog and /var/log/faillog showing really big file sizes. These are sparse files and can show a file size much bigger than they actually are if you have added users with large UID numbers or integrated Opsman with Active Directory.

You can check the reported file size and the actual file size with the ls -lh and ls -sh commands:

ubuntu@opsmanager-example:~$ ls -lh /var/log/lastlog
-rw-rw-r-- 1 root utmp 1.1T Apr 20 16:15 /var/log/lastlog

ubuntu@opsmanager-example:~$ ls -sh /var/log/lastlog
12K /var/log/lastlog

In this example, the lastlog shows a 1.1T file size, but it is actually only 12K bytes.

Resolution

This issue will be fixed in the following Opsman versions:

• v3.1.10 or later
• v3.2.6 or later
• v3.3.0 or later

As a workaround, you can temporarily clear the lastlog and faillog files

WARNING: You will lose the latest login and failed login timestamps after running this command. They will get generated again the next time users log back in.

echo -n | sudo tee /var/log/lastlog /var/log/faillog

This workaround is temporary and the issue may reoccur when users with large UIDs log back in.

Additional Information

Sparse Files