December 19, 20187 yr Getting "Your server has detected hardware errors. You should install mcelog via the NerdPack plugin, post your diagnostics and ask for assistance on the unRaid forums. The output of mcelog (if installed) has been logged" from CA Fix Common Problems. Not sure what it could be, iv attached the diagnostic file. Any help would be great, thanks z820-diagnostics-20181219-1813.zip
December 20, 20187 yr Author 48 minutes ago, trurl said: Have you done a memtest lately? Last July. Ill run one again once I get a chance. Do the logs show bad memory?
December 20, 20187 yr Community Expert 1 hour ago, macieksoft said: Last July. Ill run one again once I get a chance. Do the logs show bad memory? That is my interpretation, or at least a possibility. What do you mean once you get a chance? You do know that bad RAM can corrupt data don't you?
December 20, 20187 yr Author 1 hour ago, trurl said: That is my interpretation, or at least a possibility. What do you mean once you get a chance? You do know that bad RAM can corrupt data don't you? Ya, just backup up what I needed and the server is off until I can access it physically tomorrow.
December 21, 20187 yr Author On 12/19/2018 at 9:15 PM, trurl said: That is my interpretation, or at least a possibility. What do you mean once you get a chance? You do know that bad RAM can corrupt data don't you? Ran MemTest86 and found no errors with a full test. Do you know of anything else it could be? Edit: Turned the server back on and its not showing any memory errors. When the error showed up last time I notice something was wrong with handbrake. It was converting x264 to x265 at 2-3 fps a second instead of its normal 9-11 fps. I started handbrake back up and its running fine at about 10fps. Ill monitor the server carefully for now. Edited December 21, 20187 yr by macieksoft
December 21, 20187 yr Author 34 minutes ago, trurl said: Take a look at the syslog in the diagnostics you posted. Looked up some of the errors, while not specific to unraid they mentioned bad microcode update, possible ECC RAM warning failure, update causing older mobos to act up. Ill take it that 6.6.6 caused some issues or that one of the ram modules is dieing and ill have to look out for that. For now everything is running fine, ill grab backups more often for now.
December 22, 20187 yr 10 hours ago, macieksoft said: Ran MemTest86 and found no errors with a full test. You won't see any errors. ECC is correcting the bad chip for you. But, bad memory is still bad memory, and should be replaced. Point of ECC memory is to correct any errors until the dimm is replaced, not to keep the dimm running.
December 22, 20187 yr Author 4 minutes ago, Squid said: You won't see any errors. ECC is correcting the bad chip for you. But, bad memory is still bad memory, and should be replaced. Point of ECC memory is to correct any errors until the dimm is replaced, not to keep the dimm running. Ah, I thought it just checked for errors that might occur every so often, I didn't think it would correct its self if it would start to fail. How would I got about looking for the bad dimm then? I have 8 sticks of 4gb.
Archived
This topic is now archived and is closed to further replies.