df -h output shows that the root partition is using more than 80% of its capacity.https://<vcenter-FQDN/IP>:5480) >> Summary >> Storage will show a Critical message:File system /root has run out of storage space.increase the size of disk /root.
vCenter 8.x
/root directory to fill up./root directory :
root@example [ / ]# ls -lrthatotal 84Klrwxrwxrwx 1 root root 7 MMM DD YYYY srv -> var/srvlrwxrwxrwx 1 root root 8 MMM DD YYYY sbin -> usr/sbinlrwxrwxrwx 1 root root 9 MMM DD YYYY media -> run/medialrwxrwxrwx 1 root root 7 MMM DD YYYY lib64 -> usr/liblrwxrwxrwx 1 root root 7 MMM DD YYYY lib -> usr/liblrwxrwxrwx 1 root root 7 MMM DD YYYY bin -> usr/bindrwxr-xr-x 12 root root 4.0K MMM DD YYYY usrdrwxr-xr-x 5 root root 4.0K MMM DD YYYY optdrwxr-xr-x 21 root root 4.0K MMM DD YYYY ..drwxr-xr-x 21 root root 4.0K MMM DD YYYY .drwx------ 2 root root 16K MMM DD YYYY lost+founddrwxr-xr-x 2 root root 4.0K MMM DD YYYY rpmsdrwxr-xr-x 13 root root 4.0K MMM DD YYYY homedrwxr-xr-x 2 root root 4.0K MMM DD YYYY tftpbootdrwxr-xr-x 3 root root 4.0K MMM DD YYYY .cachedrwxr-xr-x 18 root root 4.0K MMM DD YYYY vardrwxr-xr-x 3 root root 4.0K MMM DD YYYY vasecuritydrwxr-xr-x 4 root root 4.0K MMM DD YYYY mntdrwxr-xr-x 5 root root 4.0K MMM DD YYYY bootdrwxr-xr-x 102 root root 12K MMM DD YYYY etcdrwxr-xr-x 32 root root 4.0K MMM DD YYYY storagedr-xr-xr-x 13 root root 0 MMM DD YYYY sysdr-xr-xr-x 413 root root 0 MMM DD YYYY procdrwxr-xr-x 30 root root 4.4K MMM DD YYYY devdrwxr-x--- 5 root root 4.0K MMM DD YYYY rootdrwxr-xr-x 40 root root 1.2K MMM DD YYYY rundrwxrwxrwt 37 root root 1.4K MMM DD YYYY tmp
Scenario: vSphere Client (Tomcat) work cache bloat
On long-running VCSA systems, the vSphere Client service (vsphere-ui) can accumulate temporary/compiled artifacts in:/usr/lib/vmware-vsphere-ui/server/work
Expected size: ~100 MB–1 GB. Abnormal: 10 GB+.
This content resides on the root filesystem (/), so large growth here directly triggers the VAMI “/root low on storage” alert once / ≥80%.
How to verify:
# Identify largest areas on /
du -sh /usr /var /opt /tmp /etc /home 2>/dev/null | sort -h
Take a snapshot or backup of the vCenter virtual machine.
If the vCenter is in enhanced linked mode, refer "Snapshot Best practices for vCenter Server Virtual Machines" and "VMware vCenter in Enhanced Linked Mode pre-changes snapshot (online or offline) best practice" for more information.
If vCenter HA is in use on any of the nodes it must be destroyed before attempting this process. refer "Remove a vCenter HA Configuration".
cd /root
du -shc *
echo > <file-name>
rm -rf <directory-name>
Note : For assistance in deleting directories or truncating files, reach out to Broadcom support.
Impact: Brief unavailability of vSphere Client (HTML5) during vsphere-ui restart
No impact: Running VMs, ESXi host connectivity, core vCenter services.
Pre-checks
df -h /
service-control --status vsphere-ui
du -sh /usr/lib/vmware-vsphere-ui/server/work
Steps
# 1) Stop vSphere Client service
service-control --stop vsphere-ui
# 2) Clear Tomcat work cache (safe; auto-regenerated on start)
rm -rf /usr/lib/vmware-vsphere-ui/server/work/*
# 3) Start service and verify
service-control --start vsphere-ui
service-control --status vsphere-ui
Post-checks
Confirm VAMI → Summary → Storage shows / < 80% and the alert is cleared.
df -h
Advise users to refresh their browser tabs after the UI comes back.
Optional accompanying cleanup – Journald size (non-disruptive)
# Keep last 7 days (adjust to 14/30d or use --vacuum-size)
journalctl --vacuum-time=7d
journalctl --disk-usage
df -h /