very slow reconstruction on firewire disks
From: Jeroen Vriesman <hidden>
Date: 2002-08-23 11:22:43
Dear mr. linux-raid, I'm running kernel 2.4.19 with two firewire disks attached (Maxtor 60G), as a normal disk they work fine, but when I configure it as raid-1 with the md module, the reconstruction would take about 60 days. In the log I get this: Aug 24 11:48:54 syngloom kernel: Read (10) 00 00 00 19 3f 00 00 30 00 Aug 24 11:49:54 syngloom kernel: ieee1394: sbp2: aborting sbp2 command Aug 24 11:49:54 syngloom kernel: Write (10) 00 00 00 1a 3f 00 00 28 00 Aug 24 11:50:54 syngloom kernel: ieee1394: sbp2: aborting sbp2 command Aug 24 11:50:54 syngloom kernel: Write (10) 00 00 00 1c 3f 00 00 18 00 Aug 24 11:51:54 syngloom kernel: ieee1394: sbp2: aborting sbp2 command Aug 24 11:51:54 syngloom kernel: Write (10) 00 00 00 1e 3f 00 00 08 00 Aug 24 11:52:54 syngloom kernel: ieee1394: sbp2: aborting sbp2 command Aug 24 11:52:54 syngloom kernel: Read (10) 00 00 00 2b bf 00 00 18 00 Aug 24 11:53:54 syngloom kernel: ieee1394: sbp2: aborting sbp2 command Note that it is exactly every 60 seconds, don't know if this is an md or an sbp2 problem. Regards, Jeroen.