Posted September 1, 20186 yr Monthly parity check found one error. I've been running this server for a few years now and this is the first time I've run into the parity check finding an error. Is this something that is no big deal or is it a sign of a bigger issue and what do I do to figure that out? Syslog shows no recent errors and there have been no unclean shutdowns recently either. I ran SMART tests on all the drives and found no errors. So what could've caused this error? Syslog attached for Info. TIA tower-diagnostics-20180901-1545.zip Edited September 9, 20186 yr by nickro8303
September 2, 20186 yr So certainly not an expert in this, but I have seen in once or twice before in my server. Could be simply be a bit was flipped during the original write to the parity drives or maybe bit rot? I just made sure all my cable were properly attached and no issues with any HBA cards if you are using them. Once I checked all those things, I just waited to see if the issue appeared again.
September 2, 20186 yr If you're running a non-correcting monthly parity check (as you really ought to be doing) then after finding an error you need to investigate and then run a second, correcting check because parity with errors is not much use. If you happen to be running correcting monthly parity checks then the correction will already have been made. The reason why I say that you ought to schedule your monthly check to be non-correcting is because a hardware fault could overwrite good parity with bad.
Archived
This topic is now archived and is closed to further replies.