Multiple Attempts to Power On Before Boot

[Wisdom]

Hello everybody!

My problem: whenever I go to turn on my system, it begins to boot 4-5 times before it finally seems to "catch" and stay online. The screen turns on and everything, the motherboard starts to load everything in, and then I can hear the power supply sounding like it's clicking off. The screen dies, the PSU clicks again, and then it begins again. Once the system does get to the bios/boot screen (where I have the option to go into the bios, pick a drive to boot from, etc), then everything is great and it'll stay on, stress test, and let me run some PLEX stuff without breaking a sweat.
I'm not sure if the power supply or motherboard is to blame. Both are brand new and seem like overkill for what I'm trying to do right now, so I would be surprised if they were really the issue, but I'm not sure what to expect. I'm happy to provide more information if this isn't descriptive enough.

Please note that my setup isn't yet complete. I'm missing my drives and enclosure, so I'm just in the process of stress testing the hardware I do have, before I get everything all put together in case something needs to get RMA'd.

My hardware: https://pcpartpicker.com/user/Wisdom//saved/#view=dnLCmG
https://pcpartpicker.com/user/Wisdom//saved/#view=dnLCmG What's purchased is what I'm running.

MB: ASRock E3C236D2I
CPU: G4400
RAM: Kingston ECC ValueRAM (this particular model is certified to work with the MB by the manufacturer, hence why the particular model)
PSU: Corsair RM550
USB(s): SanDisk Fit CZ33 16GB (to boot from, using two for redundancy)
Drive: leftover WD 80GB model. Just one. Just to test.

Yes, I am working outside of an enclosure, but it's in a very clean workspace on top of the motherboard's box, and once it's booted then it's been great to run for hours at a time, handling a single full 1080p PLEX transcode like a champ.

Everything is basically brand new, save the cooler (which is overkill, I know, but I had it lying around) and the drive, which is temporary and just for testing things. I haven't used FreeNAS before, but between Fester's guide on the wiki, the Hardware guide in the Resource section, and CyberJock's FreeNAS 9.10 guide, I've tried to do my homework. Let me know what you think, and thank you for taking the time to wander through this with me!

 

[Ericloewe]

Sounds like a fairly typical issue with AMI BIOSes. My desktop and my main server do something similar, where they start booting but either hang long enough for the watchdog to reset them or reset spontaneously...

Since the systems are rock solid from the moment an OS takes over and successfully loads itself, I haven't pursued much in the way of solutions.

 

[Wisdom]

Sounds like a fairly typical issue with AMI BIOSes.

Excellent, this is exactly the kind of information I was hoping to hear. Since this is my first time trying FreeNAS with a setup like this I've been unsure if this is par for the course or if there was an underlying issue.

I'll keep an eye on things (as well as listen as whatever anyone else might have to say) but I certainly appreciate this!

 

[joeschmuck]

It's always been my perception that you see this type of operation when the BIOS notices a change that needs to be made in the system such as the RAM speed may be set to Auto but the auto settings fail so it re-adjusts, likely slowing the timing down until it can pass the POST. You could manually slow the RAM timing down if you want. Or better yet, you may be able to see what the timing is using MemTest86 and then go into the BIOS and see if it matches, or manually set it. Give that a shot first if you want to pursue this.

 

[pirateghost]

It's been common on Intel gear for years to power on, power off and then power back on to stay on.

 

[Wisdom]

You may be able to see what the timing is using MemTest86 and then go into the BIOS and see if it matches, or manually set it. Give that a shot first if you want to pursue this.

I ran MemTest for a little over 16 hours yesterday/last night and everything checked out just fine (still at 2133 so far). I also haven't encountered the power cycling issue happening more than one in the several boots I've done. I think the system may have smoothed out whatever kinks it was having, perhaps in part due to being blasted by MemTest for so long? Either way, it's been smooth sailing, but I'll keep an eye on things.