FreeAgent Products

[brlcad]

I just recently purchased a 1TB Seagate drive for one of my Mac OS X systems (attached via firewire) and have noticed a problem after the drive spins down that makes it unusable. The main symptom is that after the drive spins back up, the mount point is invalid. If you pop open a command-line, an 'ls' on the mount point returns nothing while accessing a subdirectory may actually work. Similarly, traversing with Finder seems to work simply because it holds cached directory values, but will report failure if it's reset. This invalid remount causes most applications to fail when accessing a file/dir on the drive since they cannot validate the path.

I suspect this is a bug in the kernel extension that was installed or some similar problem in the drive's state management. I have found several workarounds for the problem, but wanted to report the issue nonetheless as both options are horrible.

The two workarounds that (so far) seem to be doing the trick:

1) close all apps and working directories using the drive so you can manually 'eject' the drive. then use "Disk Utility" or disktool/diskutil to remount the drive (which takes several attempts if the drive has spun down). the mount should be fine until the disk spins down again.

2) add a crontab entry to stat the drive every minute so it won't spin down. I used something like this in /etc/crontab:
* * * * * root cd /Volumes/MyDriveName && touch .Trashes

Again, both solutions are *horrible* workarounds that are far from ideal for even extended use, but perhaps useful to some other poor souls that have found themselves in the same situation. I read the other thread at http://stx.lithium.com/stx/board/message?board.id=freeagent&message.id=1406&query.id=46761#M1406 and agree that an update to the FreeAgent software is very much needed.

Being able to disable spindown is only half the problem. Spindown could and should work .. and should be adjustable without Windows .. but presently the remount after spindown leaves the filesystem in an invalid and unusable state.

Look forward to seeing an update.

Cheers!
Sean

[fredman1365]

Im having a similar problem with the same drive. could not figure out what it was. If the drive is failing or not. At one point it caused my system to freeze. I have run hard drive tests using disk utility and occasionally shows corrupted sectors. Did you have similar problems? If so the way to fix it is to keep the drive from spinning down? Does that also wear the drive out faster but keeping it spinning all the time?

[brlcad]

I haven't seen any filesystem errors other than the remount problem I mentioned. That said, I also didn't use the drive in that state either beyond testing. It is conceivable with the improper remount that it could end up causing filesystem errors; you may want to reinitialize the drive volume just to keep things simple.

If it's the remount that caused the problem, it's fairly reasonable to infer that keeping it spinning will avoid that problem. It should not wear the drive out faster -- there are plenty of systems and configurations never spin their drives down for purposes of reliability. The main difference will be that the drive pulls a little more power to stay on all the time.

For what it's worth, the cronjob has continued to do the trick for me. No spin downs, no remounts, no problems. Horrible intermediate solution, but a solution nonetheless.

Cheers!
Sean

[Steve000J]

Hi Sean,

You've just answered a vexing question, and explained why the replacement drives all 'fail' the same way! BTW Seagate warranty swap-out has been brilliant, no questions immediate replacement, especially when I say they have only been used on a Mac

I can even see the reason for this failure in the Time Machine System log now I know what to look for.

-/backupd[5392]: Volume at path /Volumes/Backup does not appear to be the correct backup volume for this computer. (Cookies do not match)
-/backupd[5392]: Backup failed with error: 18

And its repeatable repeatable with all of my FreeAgent Pros (750G and 1T)! And worse it does not happen with another brand of Firewire external.

It seems as if the Freeagent hardware causes a new BDS tag to be created every time it wakes up. Very bad practice, I can't see if there is a customer driver loaded (but there really should not be for a basic Firewire).

These drives are not fully Mac compatible. I have four in the field, and all on Mac with 'Time machine' problems. Unfortunate as they are the cheapest FW400 drives around.

Your fix (cron touch) works, and proves the problem beyond doubt, but I am not sure I can get away with it on customer sites, the 750s get too hot to touch after about three hours, they actually smell hot!

BTW Seagate warranty swap-out has been brilliant, no questions immediate replacement, especially when I say they have only been used on a Mac

Any chance of a fix Seagate?

Steve

[zackk89]

To Edit the spindown time in Mac OS X
go to terminal
type in:
pmset -a spindown 30
0=Never
then you will be asked for admin's password than your good to go!

=)
=)

[nickdyna]

Is there a way to use Terminal to make such a command for only the Seagate drive? I don't want my Macbook's HD never spinning down...

[mihalis]

Is anyone getting this _just_ with the firewire cable? I've played around with it on the USB cable and it seemed to work fine.

Either way, it's very annoying and inconvenient. Hopefully it will be addressed soon. 

[mihalis]

Bump

Anybody have a solution for this? 

[brlcad2]

You need to read the entire thread.  My original posting (from 'brlcad') that started this thread described a solution that will work for individual drives.  Just add a crontab entry for each drive.

[davidb]

Sean (brlcad), I think your suggestion to run a crontab to keep prodding the external drive every minute might be my only solution to this nightmare problem. However, I'm not a terminal user and could do with some specific instructions on how to do this. Do you think you could supply a step-by-step? I know it's asking a lot, but it would be really helpful - thanks!