ISSUE: UDMA ICRC error reading fsbn in dmesg and /var/log/messages $ dmesg ... FreeBSD 4.7-STABLE #0: Tue Feb 20 22:06:43 EST 2003 ... ICRC error reading fsbn 257696511 of 128848224-128848479 (ad1s1 bn 257696511; cn 16040 tn 220 sn 51) retrying ad1s1e: UDMA ICRC error reading fsbn 6803007 of 3401472-3401727 (ad1s1 bn 6803007; cn 423 tn 119 sn 15) retrying ...
After adding a Seagate 80GB Barracuda IDE (ad1s1e) hard drive recently I began to notice the above errors in dmesg and /var/log/messages. After reading posts and researching I did some testing with transferring files using both "cp" and "mv" on: /dev/ad1s1e results were sporadic at best, sometimes producing the errors and sometimes not. Believing this to be a cabling issue, I replaced the new IDE 80wire/40pin another 80wire/40pin IDE cable only to receive even more errors of this type. After shutting down, reopening the box, and staring blank-facedly at the motherboard and cables I remembered what an old hardware guru suggested once when adding additional hard drives, "Run the IDE cable to from the controller to the interim connector (master drive) to the end connector (slave drive)." i.e., (diagram 1) |<-----------------------IDE 80 WIRE CABLE-------------------------->| ::::::::::-------------------::::::::::---------------------:::::::::: controller interim connector end connector MOTHERBOARD MASTER DRIVE SLAVE DRIVE (My Seagate 80GB) I figured it couldn't be that easy. I tried to find some info on this using google, of course there were differing opinions on the subject. One argument suggests that the drives will fight for bandwidth, so the logic suggested here seems correct. Before this, I had the drives reversed and in the order of: (diagram 2) |<-------------ide cable---------->| controller-------slave-------master (Seagate 80G) repeating this with other cables gave same type of error messages. Following "diagram 1" above I used the original IDE cable, this seems to have solved the problem. 5 days later still no error messages. So it seems this is a possible fix to this issue. Outside my bad diagrams, I figured this might just be worth a try to others having similar issues. All opinions and any additional information on this topic are welcome. Regards, Rob Clark To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-questions" in the body of the message