5

I have an LTO4 drive that when given even new LTO4 tapes, only writes around 460GiB to the tape even though the tape's uncompressed capacity is supposed to be 800GB.

I have noticed that in one direction the drive writes close to the stated 80MB/sec, however in the reverse direction it only manages close to 40MB/sec.

The drive is a supported HP model and needed a firmware upgrade which I have done, however the problem remains. The drive's self test fails when run through HP Tape Tools, however there are no error messages just warnings on the lines where it shows the amount of data written at each different tape speed.

What could be causing this problem? The tape motors sound like they are running at the same speed in both directions, so I am wondering whether writing at barely 40MB/sec means the drive is encountering errors and having to write the data again, causing the apparent loss of write speed and capacity. Is this possible?

Does anyone else know what would cause similar problems but have the HP Tape Tools drive assessment come back as otherwise OK?

EDIT: I've tried with encryption on (462GB) and off (467GB) and the data I'm writing is not very compressible so the drive's compression doesn't do much.

EDIT2: @Mark I tried with an LTO3 tape and instead of the expected 400GB it only writes 296GiB. Watching the write amounts, it seems to write 9GiB on the forward wrap but only 4.4GiB on the reverse wrap, but the motors run at the same speed. So looks like your theory is right, and the drive is getting a lot of verify-after-write errors and having to rewrite quite a lot of data. It's a bit weird that a reason for this doesn't show up in any of the Tape Tools tests, they all say the head life is excellent and there are no issues, other than there's an issue with the amount of data written (approx 2GB during the test, but I only know from other tests with other drives this figure is normally around 6GB.)

Malvineous
  • 955
  • 7
  • 27
  • compression, clearly. – olivierg Jun 25 '17 at 20:42
  • @olivierg: How so? The tape's native capacity is 800GB (1600GB compressed 2:1). LTO compression is also block based, and if the compressed block comes out larger than the uncompressed data, the drive writes that particular block as uncompressed, so you will never end up with a larger amount of data on-tape due to drive compression. This means for an LTO4 tape you should always be able to write at least 800GB. – Malvineous Jun 25 '17 at 22:19
  • 1
    The drive could be duplicating data if the verify-after-write pass found minor errors, but if it consistently writes 460GiB out of 800GiB, that seems like a remarkably non-random error rate. Can you try using an LTO3 tape and see if the behavior stays the same? – Mark Plotnick Jun 26 '17 at 02:58
  • @MarkPlotnick: Good idea, I've updated the question with the LTO3 test results. – Malvineous Jun 26 '17 at 22:15
  • At this point, I'd place a service call. They may ask you to run a cleaning tape through the drive and try the tests again, but ultimately I think they'll just swap out the drive and someone in a service center will tinker with it to figure out what the problem was. – Mark Plotnick Jun 26 '17 at 23:00

4 Answers4

3

It sounds like clogged tape drive head. The only thing you can try is:

  • Clean multiple times with a preferably new cleaning tape the drive.

  • Run a test backup to see if something changes.

  • Contact Vendor for replacement (crossing fingers that if you need this you will have still warranty).

Forgot to mention to check if the drive was updated its driver recently use the previous available than current version.

Alexios Pappas
  • 505
  • 3
  • 9
  • This must have been the problem. I got the drive swapped and the replacement works fine, it writes the full encrypted 800GB to an LTO4 tape. The original drive wrote 9GB on the forward wrap and 4.4GB on the backwards wrap, and the new drive writes 14GB each way. – Malvineous Jul 15 '17 at 08:27
  • 1
    @Malvineous I have RMAed lots of LTO4 drives (of a different brand) suffering from the very same problem you describe. And occasionally the replacements I got from the vendor would already suffer from the very same problem as the one I had returned. At some point I realized that one of the bad replacement drives had the very same serial number as one I had RMAed a couple of months earlier. – kasperd Jul 15 '17 at 15:58
2

LTO drives will perform a verification read on the data they have just written. If it is not good enough they can write an extra copy as you mention. The data physically written to the tape contains extra metadata not visible to the backup application, which the firmware can use to deduplicate the extra copy at read time.

Since LTO drives write in both directions it is not enough to have just one read head and one write head. The order in which those two heads were located would mean that the verification read would only be possible in one direction not both.

For that reason the drive has three heads. They don't wear down equally fast. When the first head is wearing down you will see the drop in capacity in one direction because it is a different head which is used in the other direction.

The failure mode you describe is not unique to HP. I have seen multiple non-HP drives fail in the same manner as you describe. The only difference was that I saw it on drives rated for 120MB/s so I saw 120MB/s in one direction and 60MB/s in the other.

kasperd
  • 29,894
  • 16
  • 72
  • 122
  • That's interesting because according to the HP drive tests, the faulty drive had something like 98% remaining on the head life, same as the working replacement drive. It only gave me a warning that the drive didn't write as much data as expected. – Malvineous Jul 16 '17 at 00:03
  • @Malvineous I don't know how that head life percentage is calculated, so it is difficult for me to guess what went wrong with it. Given that your symptoms perfectly match those of a worn down head and that worn down heads is the most common drive failure I have seen, I still think that is what happened to your drive. In my experience measuring write speed gives a better picture of head health than vendor test tools. – kasperd Jul 16 '17 at 09:48
1

I'm no specialist on LTO, but I suspect if the tape is not "streaming" while being written (but uses "stop and go"), the firmware may leave some "safety gaps" between the last write and the next write (the tape has to stop, position back a bit, then re-accelerate forward and then start writing after a safety gap).

So like in the good old days it would be helpful to stand near the drive when it's writing to listen how it sounds. You can clearly hear the stop-n-go mode. Stop-n-go mode may also make any built-in drive compression efficiency worse.

U. Windl
  • 286
  • 2
  • 13
  • Yes this is called tape "shoe shining". It is definitely a valid cause of slowdowns, and as you can say easy to hear. But you can usually tell when this is happening because your disks are idle, the memory buffer is full, but no data is going to the tape for a few seconds. If your memory buffer regularly hits zero then the cause is that you can't get enough data to the tape, but if the memory buffer is usually full then it's a fault within the tape drive itself. In my specific case however there was no shoe-shining happening during the slow writes. – Malvineous Jul 27 '21 at 09:58
-3

Sorry. Real capacity LTO is 800Gb. But encription can reduce it. Try to disable it and run the test again.

Mikhail Khirgiy
  • 2,003
  • 9
  • 7