Big problems with firewire drive and OSX

[bchern 0]

I want to try to revive this topic, as at the end of my previous thread I had verified that the (identical) problems did not exist in OS 9, and then no one had any more to say on the matter. AS a result, I hardly ever back-up since it is so hard to take a day to switch back and be stuck in OS 9 doing nothing but waiting for retrospect to finish.

 

We are talking about an Onstream FW30 firewire tape drive, (now) OS X 10.2.3, 533 MHz G4, Retrospect Server 5.0.238, driver update 3.2.104.

 

On my latest attempt to see if I could get through at least 1.2 tapes and get my OS X drive backed up (another problem, I have kept this drive isolated from OS 9, it never had the OS 9 drivers installed, so of course it never gets backed up since I have to back up under 9) using tapes that had seemed to work before. Everything seemed to be going fine, until the verify. This is another problem. It is very hard to tell when a backup has failed. Retrospect will just go along writing and nothing shows up in the log. Then, I finally had to go to bed. apparently it waited until I was no longer watching it to start reporting the bad headers. Actually, I have no way of knowing when it did this. It was asking for tape two when I woke up, so I assumed it had gotten through the first tape okay. It then spit out tape two immediately with a 105 error. So, I tried a recatalog, same deal, so I erased the tape and started the backup again. Then I got enough coffee in me to realize I needed to look at the log again (even though I had kept an eye on it yesterday for hour after hour with no sign of trouble....)

 

I don't know when the trouble started, because the log said there were another 38,000 lines it couldn't show me. One bad backup header entry after another. I started to verify tape one (wait wait wait wait....) and about 5 gigs in I got the first bad backup header entry.

 

"+ Executing Verify at 2/2/2003 3:35 PM

To backup set Xist A…

Bad backup set header found (0x940166ce at 4,870,012).

2/2/2003 4:56:50 PM: Execution stopped by operator.

Remaining: 272600 files, 10.7 GB

Completed: 28672 files, 4.3 GB

Performance: 54.5 MB/minute

Duration: 01:20:59 (00:00:49 idle/loading/preparing)

"

 

It seems like something interrupts communication, but Retrospect just keeps writing away and things get worse and worse. (I had the verify stop on error this time. but, it would have just gone on for 66,000 errors or so.)

[AmyJ 0]

Customers have reported problems with Onstream firewire drives after upgrading to 10.2.3. All reports to date indicate that the 10.2.4 update resolved their issues. Please try an upgrade.

[bchern 0]

I missed this last reply because I had pretty much given up. Did the OS update come out before the last Retrospect Driver update? If so, I almost certainly tried it, because I tried running a backup in OS X again after the dirver update came out. At first I thought it was working, but pretty soon the 205 errors came back. I have only tried to backup whan I bite the bullet and go to OS 9 for a couple days. However, I am beginning to doubt that that is working either, because even though the backup phase completes successfully I always get a 105 error if the verify has been interrupted when I try to do the next incremental backup. Recatalogging from tape will go all the way through and say it wa successful, and then as soon as I try to write to that last tape (any last tape) it fails immediately.

 

 

[bchern 0]

still doesn't work under 10.2.5 upgrade.

 

Seemed to back up one entire volume, spanning two tapes. first tape verified fine. second tape got unexpected end of data.

 

repeated attempts to rebuild a catalog from the first tape failed, seemingly in a degenerating manner. Back to the old error 205 lost contact. what's up with this?

 

a second attempt to backup to the _same_ tape that had worked the first time, ended with the infinite bad backup headers errors on verify (after seemingly writing both tapes successfully) about as far into the tape as the last attempts to rebuild the catalog had failed at.

 

what _should_ I have done the first time if trying to rebuild from tapes is wrong? As I said, that first tape had written and verified fine. Now it seems to be trashed.

[AmyJ 0]

All the problems and errors you are outlining tend to point towards hardware failure. You're seeing multiple hardware generated errors in OS X and OS 9. This is one of the hallmarks of hardware failure - the problems gradually worsen until the configuration is completely unusable.

 

The following troubleshooting should help you isolate the cause of the issue:

 

Check to make sure you have the latest FireWire firmware update for your computer. Apple often releases firmware updates that may help to resolve FireWire issues. Search the Apple Software Updates Web site for relevant updates.

 

Try a new piece of media to see if the problem is related to a faulty or damaged medium.

 

If possible, try using a different brand of media. Often drives are picky regarding the brand of media. Follow the manufacturer's recommended list of brands.

 

Clean the heads with a cleaning cartridge.

 

Isolate the device. Another device on the chain may be interfering with the backup device's communication. If your backup device is connected to your computer through a hub or another FireWire device, unplug it and connect it directly to a port on the computer. If it is already connected directly to the computer, try changing ports. If the problem persists, do not reconnect the other devices, and continue down the checklist.

 

You may have a bad cable. Replace the cable that connects the device to the computer.

 

Completely uninstall any other third-party backup software that may be on your machine, including any drivers that software may have loaded for the device.

 

The system board or FireWire adapter in the computer may be having a problem. Install Retrospect on another computer, if available, and try the device there as the only device.

 

The device itself may be defective. If you have implemented all of the preceding steps and get failures on new media after changing media, cables, software, firmware and computers, then the backup device, being the only factor that has not changed, is the the last variable. Contact your drive vendor for further diagnostics, tests, or to inquire about repair or replacement.

[bchern 0]

Quote:

---

All the problems and errors you are outlining tend to point towards hardware failure. You're seeing multiple hardware generated errors in OS X and OS 9. This is one of the hallmarks of hardware failure - the problems gradually worsen until the configuration is completely unusable.

 

 

 

---

 

Completely different, seperate and consistent errors in the two operating systems.

 

In OS 9 tapes become unreadable after operating system crashes.

 

In OS X the OS loses contact with the device during normal operation.

[shadowspawn 0]

Quote:

---

Everything seemed to be going fine, until the verify.

 

---

 

That is why the verify step is so important. Retrospect writes the data to the tape during the first pass.

The hardware says no problem...

The operating system says no problem...

So no problem, right?

Not so fast!

Retrospect then goes through all the files again in a second pass, reading from the backup and comparing the original on disk version with the backup version. This is where major data integrity problems get discovered, like the tape drive silently writing garbage.

 

You have some nasty corruption problem, and need to so some serious troubleshooting to isolate where the problem is occurring. Amy posted a checklist to work through. You can also find similar troubleshooting steps in the manual, and in the online Knowledge Base. Basically since you have posted twice and no one can identify from your description exactly what the problem is, you have to slog through eliminating every possibility until you pin down the problem!

 

Good luck!

[wkegel 0]

I have had similar problems when I moved my backup system from an USB30 tapedrive under 9.2.2 to a FW30 tapedrive under OS-X 10.2.5. I had to switch tapedrives since the USB tapedrive is not supported under OS-X and this new iMac can't boot in 9.2.2.

 

However, two weeks ago I took the following steps:

 

• Upgraded the OS to 10.2.6

• Upgraded the drivers to 3.5.104

• Gave the tape drive a FW port of it's own (previously it was connected through a Belkin FW hub)

• Mounted the tapedrive upright on my desk using the stand provided by OnStream. I did this because I had noticed that the tapes could become quite warm, so I figured the drive needed all the cooling it could get.

 

I'm happy to report that since then the configuration worked quite well. I have seen one failure on an older tape, but I have written some 50 sessions without further trouble.

Since I made all the changes at the same time I'm not sure which one did the trick. I think we can rule out

• media (I have seen the problem on several tapes and I use only OnStream ADR cartridges, some of which were brand new)

• wear and tear to the tapedrive (it was new)

• the cable (cable problems tend to produce random intermittent problems, while my problems were reproducible).

 

Hope this helps.

 

[bchern 0]

well, I already had my drive standing on its side, and I used many, many different tapes so I doubt it was the media. I always did have it on its own port. however, I'm sorry to report that it will be on that port no longer, since I removed the drive to send it in for warranty repair and/or replacement given my failure to find any clear answers on this end. (and, to all appearances, it has gone missing entirely now, and I will have to find out whether onstream or the post office saw it last.) anyway, back to the firewire port, after I removed the drive (turned off for safety, even) my motherboard firewire ports never worked again. I had the infamous fried port experience. So, if I ever do see the drive again I will have to be trying it through a PCI card, giving yet another variable.

 

I suppose the fried ports could be a sign there was something wrong with the drive all along. I remain unconvinced one way or another. but, I'm glad someone got one working, anyway.

[pt44 0]

I'm having the same problem (+ a couple of others thrown in :-(

 

I'm on a Dual 1.25 G4 MDD, 10.2.6, Latest version of Retrospect Workgroup and all the latest drivers. My Onstream FW30 is also having its own problems with tapes getting stuck in the machine, just to add to the fun & games.

 

Ontream is apparently having financial problems and have temporarily (I hope that's only temporary) gone bust. They are not answering the phones in the UK, US or their head office in the Netherlands. I have spoken to a UK distributor, who advised me that this has happened before with them, and they should be back in action within a month or two. We'll see.

 

However back to the problem. I haven't fully gone through the check list, I'll do that next. However it is extremely strange, I find, that it is specifically FW30 drives that seem to be all producing this same fault. Sure it may just be coincidence, but it was working fine a couple of months ago, now I simply can't back up. Its always the last tape that fails on verify.

 

I'm very near to the point of giving up with the FW30 and buying something else. I have over $1500 invested in tapes etc. but having no backup is useless. Also I have data of over 200Gb which is growing rapidly (I'm a digital photographer, shooting around 2-3Gb/week) So backup's on the FW30 take at least two days, usually more like 4. Has anyone got anything good or bad to say about AIT1 or AIT2 drives? they'd certainly speed things up, perhaps its worth the xtra investment.

 

Regards Paul

[wkegel 0]

You may have some problems contacting OnStream since they are indeed bankrupt. In The Netherlands, in such cases one or more lawyers are appointed as trustees. I found three phone numbers for trustees in the info pages of the Dutch Chamber of Commerce for the Eindhoven region Kamer van Koophandel, handelsregister :

For OnStream Holding BV and OnStream Data BV the listed number is +31 40 2444360 (no names given)

for OnStream BV the names are mr. A.A.M. Deterink (+31 40 2626600) and mr. G.T.C. van der Bilt

(+31 40 26380600). I suspect the last number is a typo (it has too many digits) and should read +31 40 2380600 - it is the closest match to a number of a lawyer firm in Eindhoven I could find in the yellow pages.

 

Since these pages are in Dutch I thought I should pass on this information.

 

 

[bchern 0]

man, has it really been almost two years?

 

Just to complete the joke, in case anyone is still reading...

 

I decided to get an RMA for the drive and give up on the idea it might be a software issue.

 

They sent me RMA instructions, I carefully packed the drive and sent it Priority Mail, for what was supposed to be a 5 day turn-around.

 

And never heard from it/them again. Attempts to contact any portion or OnStream repeatedly failed.

 

Anyone want to buy a huge stack of ADR 30 tapes?

[natew 0]

Hi

 

There was someone on the forum looking for Onstream carts just recently. Although I can't seem to find the post at the moment.

 

Nate