Image corrupt -SATA drive

Had an interesting development today. Based on somebody's post I was checking how long it took to do an image with verify if No Compression is used. I use Normal all the time. TI9.0 Build 2337.

I did an image of C with Normal and it created and verified OK. I then did the same image with No compression. In both cases the images are HD to HD both SATA drives. PC is 2.8G P4, Asus P4P800MB, 1GB dual-channel Corsair memory. All testing done within Win XP.

The No compression image showed as corrupt. Subsequent verifies after restarting TI, rebooting etc still showed it as corrupt.

Deleted bad image and created it again. I slowed down the RAM in BIOS even though it had run Memtest86+ 25 hours continuously (just for fun) at the faster settings a few days ago. Also, the Corsair's faster setting is what it is spec'd to run at on this MB.

Created new no compression image. Took just over 2 minutes. Verify completed OK but took an astounding 35 minutes!!! We are dealing with only a 6.6GB image.

Looked in the event log. Disk errors flagged. The 35 minute successful verify showed all sorts of Event ID 51 messages. Prior to that it indicated some controller errors and a parity error. Event viewer recommendation was that the cable was bad.

Changed the SATA cable to the backup drive. Verified same image in about 4 minutes! No errors in Event Log.

Later created another no compression image and it created in 2min 8s and verfied in 3min 28S without any problem with either TI or the Event Viewer.

Interesting that the normal compression imaging was OK although without compression the backup disk is really getting a workout. It writes more data (uncompressed) in less time. 6.6GB were written in 128S versus 4.3GB in 188S for compressed. While verifying the throughput on the no compression case is about 10% higher.

Anyway it appears that the cable was bad. Obviously it had to be marginally bad to only fail under such circumstances. I had to replace the other SATA cable on this PC before but I unfortunately can't tell if this cable is the mate that came with the other one with the motherboard. The first replacement was nothing to do with TI but the event viewer showing the odd disk parity error.

I should mention that I had 2 cases where an image wouldn't verify the first time but was OK the second time. This happened a while ago and now I wonder if that wasn't related but there is no way of knowing.

If you are having trouble check the Event Viewer! It may give you a clue.

 

Very useful post! It perfectly illustrates that problems can be hardware related. Your point about checking the event viewer will help many who have problems. That is providing they can find it )

Xpilot

 

When I replaced the first cable I remember thinking that it was funny how specific the change the SATA cable fix was. Normally, for a parity or other error you might expect it to say bad drive, bad controller or whatever not just "replace the cable" all by itself. Makes me think that SATA cables are known as a weak link in the system.

It would be interesting to know if others have had a similar experience with SATA cables.

 

I used to have an intermittent problem with one of my non-raid SATA drives. When the problem occurred the drive was inaccessible. If the problem occurred during boot, it would take a very long time to complete booting. After the boot the drive was not visible in Window XP (i.e. no drive icon). I thought the connection might be loose so I tried reconnecting the cable at both ends each time the problem occurred though the connection seemed tight. That didn’t help. I also looked at the drive SMART data and scanned all sectors. The drive had no bad sectors and was healthy. So a bad drive was ruled out. Finally I replaced the SATA cable though it looked “new”. Since then the problem is gone!