Intermittent Power-On Failure with Multiple PCs

[mschoyen]

Hey guys,

Here's the problem:

1. The unit is powered on via the case power switch
2. The HDD & Power LEDs on the case flash once but the unit does not turn on. No fans spin up or other activity occurs
3. Pressing the case power switch again results in no response at all
4. Unplugging the PSU from the wall (or turning it off with the PSU power switch) and allowing the unit to sit for a few minutes before reconnecting power and turning the computer on will either cause it to function normally or repeat the same behavior. It occasionally takes 2 or 3 iterations of this to "fix" the issue.

I have so far been unable to reproduce the behavior on-demand or identify any common point of failure in the hardware. The fact that incidents are only happening every few days in labs that are regularly in-use has made it very difficult to troubleshoot, especially with a stop-gap solution already in place.

Here's some background:
I help oversee 4 computer labs for a college with a total of about 110 identical computers. The problem occurs in about 5-6 computers per lab (the same each time, but not necessarily all at once). It only occurs every few days or few weeks and the computers may be shut off and started successfully several times between incidents.

Computer specs are as follows:

• Intel Pentium D 3GHz Dual-Core Processor
• 3GB DDR2 RAM
• Asus P5B-VM Mainboard
• ATI Radeon X1050 Video Card
• Attansic L1 Gigabit NIC (in addition to an onboard NIC)
• Pioneer DVD-RW
• Startech IDE HDD Caddy Bay
• AcePower 400W PSU

These machines have no internal HDD other than the one supplied by the student via the IDE caddy bay. This problem occurs regardless of whether a caddy HDD is connected or not.

Has anyone run into anything like this before?

[compdoc]

I have never personally experienced anything like that, but it sounds like it could be a power issue. If it is at all possible try to identify if the problem computers are plugged into the same circuit. if they are there may be a problem with the power on the circuit or maybe to many computers on that one circuit.

[mschoyen]

I have never personally experienced anything like that, but it sounds like it could be a power issue. If it is at all possible try to identify if the problem computers are plugged into the same circuit. if they are there may be a problem with the power on the circuit or maybe to many computers on that one circuit.

That's been my thought as well. I, frustratingly, have no ability to check that option at the moment though.