Server lights green, but no access to anything


One of our clients has an HP Proliant ML350 that has been having a recurring problem. Of course, it doesn't happen all the time, but it happens often enough that it seems like there is a pattern. Here is the most recent event log entry related to this issue:

The previous system shutdown at 10:07:19 PM on 9/16/2011 was unexpected (Event ID: 6008)

When this happens (usually on Friday evening, it seems), the server lights are all showing a green status (either solid, or flashing, as appropriate), but no server resources are accessible, nor can we access the server via its own monitor, nor can we address it via RDP. We are forced to shutdown the server manually, then start it back up. After restarting, we get the above-mentioned log entry that the shutdown was unexpected.

We initially suspected that the old UPS attached to it was causing the trouble, since it was not an adaptive sinewave-capable UPS, which is required by all the recent Proliant ML 350's. We replaced the UPS with an adaptive sinewave-capable UPS thinking that we would fix the problem, but it did not. There are two power supplies in this server. One goes to the UPS, the other to a dedicated circuit. The dedicated circuit is not otherwise protected, but we are going to fix that right now (surge protector, another UPS, etc.).

There are 17 entries with Event ID 6008 dating back almost exactly a year (back to 9/21/10). Seven of these entries occurred within a couple minutes of 10:10pm local time. The other 10 seem more random. It's weird to me that there are almost half that occur at almost exactly the same time. The backup is scheduled to run at 11:00pm local time, so it's not triggering the problem. We don't have any other scheduled tasks near 10:00pm.

Any ideas where to start? Thanks for your suggestions!

Who is Participating?
marcustechConnect With a Mentor Commented:
When you say you can't access the server from it's own monitor, what's actually on-screen when this crash happens, or is it on a blank screen and won't wake? Have you had a look at the internal health LEDs on the motherboard to see if they're any more informative than the front-panel ones? If nothing's logged in the event log immediately before the crash, it definitely indicates a hardware fault - ML350s can run quite hot, have you had a look to see what the internal case temp is like when the crash is occurring?
JAStillwellAuthor Commented:
Marcustech: the issue recurred last night. No access via local monitor. Only a ping. No access to network shares, etc. I configured a Sonicwall to Sonicwall VPN before this current issue, so I verified that the SBS server was the only problem. Arrived on-site. Lots of green lights on the server, but no local access, etc. Spoke to HP tech support. We reset the NVRAM, then loaded the HP Proliant Support Pack and rebooted. At this point the server is running, and we are waiting for a follow-up call from HP tech support. We previously didn't have the Integrated Log Viewer installed, and it is now. If/when this recurs, we can check the HP hardware via the ILV. At this point, I guess we wait!

Thanks for your follow-up. I'll report back when I have new info.
JAStillwellAuthor Commented:
So far so good after applying the HP Proliant Support Pack. It's been a little over a week and the problem has not recurred. I'll report back again in about a week.
JAStillwellAuthor Commented:
After talking to HP tech support about the issue, they chose to replace the system board. Since then, we have not had the problem recur.
JAStillwellAuthor Commented:
After troubleshooting with HP, they replaced the system board and all is well.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.