Blade Posted September 1, 2018 Share Posted September 1, 2018 I was running a supermicro X8SIL-F and the parity check was very quick (just under a hundred MB per sec) Parity check took 13 1/2 hours on the X8SIL-F I upgraded my motherboard to a supermicro X8DTH-6F and now the parity check is about 28MB/sec This one says it was going to take over 2 days I have 15 hard drives and 2 SSD cache drives Is there something I could tweak to speed up the parity check? I am attaching a diagnostic tower2-diagnostics-20180901-1012.zip Link to comment
JorgeB Posted September 1, 2018 Share Posted September 1, 2018 Didn't you notice the errors on the failing disk3? 197 Current_Pending_Sector -O--CK 195 195 000 - 1356 Notifications disabled? Link to comment
JorgeB Posted September 1, 2018 Share Posted September 1, 2018 BTW, you should immediately cancel the parity check, since disk errors on a correcting check can corrupt parity, scheduled checks should always be non correct. Link to comment
Blade Posted September 1, 2018 Author Share Posted September 1, 2018 I have cancelled parity check it looks like I need to replace disk3 then? Link to comment
Blade Posted September 1, 2018 Author Share Posted September 1, 2018 Would disk3 having read errors cause the parity check to take almost 3 days to run? Link to comment
Squid Posted September 1, 2018 Share Posted September 1, 2018 Every read error means that the system has to recalculate the information that is supposed to be in that sector from all of the other drives, and re-write the offending sector. So, yes Link to comment
Blade Posted September 1, 2018 Author Share Posted September 1, 2018 Thanks Squid I will replace the drive immediately then - I have a spare here Link to comment
JorgeB Posted September 2, 2018 Share Posted September 2, 2018 And make sure you enable system notifications, you would get warnings of the pending sectors and the read errors. Link to comment
Vr2Io Posted September 2, 2018 Share Posted September 2, 2018 13 hours ago, Squid said: Every read error means that the system has to recalculate the information that is supposed to be in that sector from all of the other drives, and re-write the offending sector. So, yes I also think it should be , but recent disk read error event show negative. So, if it is true, then tick correction or not during parity check also need consider. Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.