iLO (Integrated Lights-Out) is a separate management controller on HPE ProLiant servers that keeps working even when the operating system is off or frozen. Through iLO, you can power the server on and off remotely, see its screen and, most importantly, monitor hardware health.
What Does iLO Show You?
- Disk and RAID status: failed disks, disks with a Predictive Failure warning, and arrays in a Degraded state.
- Memory (RAM): memory errors, which are often a prelude to module failure.
- Fans and temperature: failed fans or high component temperatures.
- Power supplies: failure or loss of power to one of the redundant power supplies.
- Logs: the Integrated Management Log (IML) records hardware events, and the iLO Event Log records events for iLO itself.
A disk showing “predictive failure” still works, but it should be replaced as soon as possible. In RAID 5, if a second disk fails before the first is replaced, you lose the entire array.
Step 1: Set Up iLO Correctly
- Connect the iLO port to the management network (a separate VLAN), not the user network.
- Change the default iLO password and create a separate account for each person.
- Update the iLO firmware; older versions have known vulnerabilities.
- Never expose iLO directly to the internet; remote access should be through VPN only.
- Sync the iLO clock with NTP so event timestamps are accurate.
Step 2: Enable Alerts
iLO offers several ways to send alerts:
- AlertMail: sends an email on important events. Just enter the mail server (SMTP) and recipient address in the iLO settings.
- SNMP Trap: sends alerts to network monitoring software.
- Remote Syslog: sends logs to a central log server.
- Redfish API: a standard RESTful interface that monitoring software can use to read health status directly.
After configuring them, send a test alert and make sure it reaches the right person.
Step 3: Turn Alerts into Action
A common problem is that alert emails go to a mailbox nobody reads, or there are so many that the important ones get lost. A few suggestions:
- Critical alerts (disk failure, power supply, high temperature) should reach the responsible person immediately, ideally by messaging app or SMS.
- Less important events should be rolled up into a daily report.
- For each type of alert, it should be clear who must act and by when.
- Keep common spare parts on hand, such as a disk of the same model.
Step 4: Centralized Monitoring of Multiple Servers
When you have several servers across several branches or clients, checking each server’s iLO one by one is not practical. Monitoring software can collect the status of all servers in one place through Redfish or SNMP, build a daily report and alert only when there is a problem. The ITPlus724 early warning system does exactly this: it reads hardware health from the iLO of HP servers and reports on it every day alongside backup status.
Checklist
- iLO on a separate management network with no internet access
- Default password changed and firmware up to date
- AlertMail or SNMP enabled and tested
- Critical alerts reach the responsible person immediately
- The IML log is reviewed regularly
- Critical spare parts are on hand
If you want your servers’ health monitored without daily manual checks, ITPlus724 engineers can configure iLO and connect your servers to the early warning system.