This is the equivalent command for applying the fix:

sed -i 's_du -sb $dest | cut -f1_du -sb $dest 2>/dev/null | cut -f1 || true_g' /usr/share/cloudstack-common/scripts/vm/hypervisor/kvm/nasbackup.sh

We haven't confirmed the exact root cause of the du failure yet. As a precaution, we applied this fix to all servers and will monitor backups over the next few days.

So, I am running tests with 2>>/var/log/cloudstack/agent/nasbackup.err so I can see what is the error message.

Proposed Changes Rationale

backup_size=$(du -sb "$dest" 2>/dev/null | cut -f1) || true

• NFS issues may cause du command to fail.
• A size retrieval failure should not invalidate a successful backup.

timeout 60 umount "$mount_point" 2>/dev/null || true
rmdir "$mount_point" 2>/dev/null || true

• Another process may keep the device busy (e.g., parallel backups).
• Network issues may cause hangs on NFS.
• Cleanup failures should not invalidate a successful backup.

echo -n "$backup_size"

• Outputs the size at the end to confirm the script completed past the potentially problematic commands.

Dear @abh1sar , do you think you can help me with this bug? Regards,

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 18.89%. Comparing base (957bfbb) to head (edc80b1).
⚠️ Report is 2 commits behind head on main.

@@            Coverage Diff            @@
##               main   #13424   +/-   ##
=========================================
  Coverage     18.88%   18.89%           
- Complexity    18223    18226    +3     
=========================================
  Files          6174     6174           
  Lines        555226   555226           
  Branches      67774    67774           
=========================================
+ Hits         104872   104895   +23     
+ Misses       438834   438810   -24     
- Partials      11520    11521    +1     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.