This topic has been marked solved and closed to new posts due to inactivity. We hope you'll join the conversation by posting to an open topic or starting a new one.
I can't run the Smart Control command on the RN because it will no longer boot no matter what I do. It just comes up with ReadyNAS in the display and never does anything further. This is regardless of which or how many disks are in it or if I'm holding the reset pin on the back or not. I realize that sounds like a failing PS but it has a new PS.
I put the drives in my spare Linux PC, one at a time (I have a hot swap bay) and ran the smartctl - x command on them all and captured outputs into txt files. Of the six original drives one has Device Error Count=0, four have 1 error each and the bad one has 151. The new replacement shows 0 error count.
I mean, they all have several years of use but the above doesn't seem like it should be an issue.
Perhaps bad memory? Sure be nice if I could run a memory test but I can't get it running to get to the menu. Should the RN normally be able to boot or at least get to the menu with 1 disk or what is the minimum number it can boot with?
I mean, what else is there? I'm not taking the CPU out and rubbing on balloons or something, the box hasn't been apart since I put the new PS in and has logged plenty of hours when I copied the data off.
I put the drives in my spare Linux PC, one at a time (I have a hot swap bay) and ran the smartctl - x command on them all and captured outputs into txt files. Of the six original drives one has Device Error Count=0, four have 1 error each and the bad one has 151. The new replacement shows 0 error count.
I mean, they all have several years of use but the above doesn't seem like it should be an issue.
I can't run the Smart Control command on the RN because it will no longer boot no matter what I do. It just comes up with ReadyNAS in the display and never does anything further. This is regardless of which or how many disks are in it or if I'm holding the reset pin on the back or not. I realize that sounds like a failing PS but it has a new PS.
Perhaps bad memory? Sure be nice if I could run a memory test but I can't get it running to get to the menu. Should the RN normally be able to boot or at least get to the menu with 1 disk or what is the minimum number it can boot with?
You should normally be able to get to the memory test with the NAS diskless. A normal boot would require at least one disk of course.
Unfortunately I don't have any thoughts on a good next step, other than to measure voltage on the new supply. It is new, but maybe is having an early failure?
Although I don't have a PC that uses the same memory as the RN to allow me to check it, I swapped the two memory sticks in the RN, just exchanged their places. No workie. Then I pulled the one out of slot 2 and it booted! So I shut down and swapped to the other stick in slot 1 and no workie. To my way of thinking I have one bad SIMM. They are Kingston KHX6400D2LLK2/4G, for what it's worth.
I did a factory reset last night and this morning it's about 48% complete in its initial sync. I don't want to be too optimistic, but I've got my fingers crossed...
This makes me curious again about the other RN chassis, the one that is seemingly stone dead, won't even power up the PS. I recall pulling RAM and reseating several times but I think it's worth revisiting that chassis. The symptoms are different but who knows?
What's the current recommendation on optimal RAM amount for these machines and does RAM timing class make an appreciable difference? I expect this older style memory won't be cheap, so looking for recommendations...
Actually, if you stick with a total of 4GB of RAM via two 2GB sticks, it's dirt cheap. Unless you are running specific apps that require a ton of RAM, 4GB is more than enough. The right kind of 4Gb sticks (unbuffered, non-ECC, 2Rx8) are uncommon because PC's of that generation were mostly running a 32-bit OS, so they could only use 4GB. Server OS's were mostly 64-bit, but servers usually use buffered and/or ECC RAM.
The motherboard does not have dual-channel RAM, so there is no reason to match the two DIMMs and you gain nothing with 2 x 2GB vs. 1 x 4GB.
Thanks Sandshark, helpful as always. I won't look to expand memory capacity then. And it seems that this older memory is quite expensive but only from retailers, used sources seem plentiful and relatively cheap. I sure hope this is the reason for the failure of the sync and instability. If so, I'm even more amazed it stayed up long enough for me to back up the data. I'll be even more pleasantly surprised if it turns out to be the same issue with the other chassis that is seemingly stone dead!
After a bit of reading here it appears that the memory can be of the 1066Mhz variety, no need to limit my search to 800Mhz apparently. The CPUs were updated to the fastest possible just before I converted to OS6 a few years back.
Just a quick note to close the loop on this thread.
The "stone dead" RN lives on... It was bad memory on this box as well. Hard to believe but of the four memory modules in two boxes, three were bad or chose the moment I replaced the PS to go bad. I've seen bad memory prevent booting but never seen it prevent even triggering the PS on. In any case, fresh memory immediately allowed it to start.
Both RNs are now running, are fully synched, have several terabytes of data and have not been the least bit unstable.
Oddly, one of them constantly reports that its UPS has been disconnected then immediately reports it being plugged in and it's a "new" UPS. Like 50 time per day. Swapping cables, swapping UPS's, using different USB ports etc makes no difference at all. But that's another thread perhaps.
Thanks to you both for your patience, assistance and valuable input.
