MidWestTechnicalSales
asked on
SAS LTO4, 5, &6 Library Tape drives testing on bench
Hi:
We have not been able to power up and control SAS library tape drives once removed from the library when on the technical bench for testing. What I have been told is that once a SAS LTO4 or 5 has been installed into a library tray it is no longer testable on the bench and out of the tray. Seems several of the LTO4s and 5s. The concept of the personality being installed at the factory does not always hold true for all newer type LTO tape drives.
Many of us in the repair field have the same issue and non of my associates in the business have been able to solve.
Any ideas
Thanks
-Dan
We have not been able to power up and control SAS library tape drives once removed from the library when on the technical bench for testing. What I have been told is that once a SAS LTO4 or 5 has been installed into a library tray it is no longer testable on the bench and out of the tray. Seems several of the LTO4s and 5s. The concept of the personality being installed at the factory does not always hold true for all newer type LTO tape drives.
Many of us in the repair field have the same issue and non of my associates in the business have been able to solve.
Any ideas
Thanks
-Dan
Davis McCarn
Tape diagnostic tools are make and model specific. I need the make and model of, at least, one of the drives?
ASKER
Davis:
Thanks for the response typically we test and repair IBM and HP tape drives. Most of the issues is with the sled design. Typically the Dell 2/4K, IBM TS series and the HP ML series. All are HH tape drives the we see the most of that fit into the sled.
We are aware of the DIP switch setting we need to use in bench testing. No issues with FC or SCSI but the SAS drives are the ones that pose the issue.
Regards,
Dan
Thanks for the response typically we test and repair IBM and HP tape drives. Most of the issues is with the sled design. Typically the Dell 2/4K, IBM TS series and the HP ML series. All are HH tape drives the we see the most of that fit into the sled.
We are aware of the DIP switch setting we need to use in bench testing. No issues with FC or SCSI but the SAS drives are the ones that pose the issue.
Regards,
Dan
ASKER
Davis:
We use HP LT&T tools as well as the IBM and Dell versions of ITDT, some of the HP drives need X-talk to play as well. Standard windows OS some servers and Win7 machines. We also have the XP machines for testing some of the vintage stuff.
Dan
We use HP LT&T tools as well as the IBM and Dell versions of ITDT, some of the HP drives need X-talk to play as well. Standard windows OS some servers and Win7 machines. We also have the XP machines for testing some of the vintage stuff.
Dan
You can kludge a SATA cable to fit an SAS drive; but, getting it recognized and, more problematically, getting it to run diags that way is probably a long shot, at best.
H*ll, I had conniption fits simply finding testing tools for a parallel SCSI, LTO 4 drive just over a year ago!
If I were you, I'd start looking for good deals on SAS backplanes and controllers so I could test them in their respective sleds.
H*ll, I had conniption fits simply finding testing tools for a parallel SCSI, LTO 4 drive just over a year ago!
If I were you, I'd start looking for good deals on SAS backplanes and controllers so I could test them in their respective sleds.
ASKER
Davis:
We have the necessary adapters to make the connection to make the physical connection. When we power up the drive we get EC E so we cannot test with something like ITDT. We can access the serial port on the drive with custom cables and access the tape drive for serial port functions like re calibrate and look for open channels on the head. We also can access the front button pushes to semi control the drive, the E has us stuck which indicates a lack to communicate with the tape drive via the SAS interface.
We have no issues testing these drive but they must be installed a library, for the repair function we like to see what is going on so it makes it more difficult to use the library that is where the bench function for us is useful. We have test beds set up to test any type of autoloader tape drive but most of those FC and SCSI can be tested on the bench with out issues.
We can bench test the FH SAS tape drives on the bench like from a Dell ML or ADIC i500 or the IBM TL series or the HP line as well.
Your suggestion on the SAS backplanes could be a good idea but the library interface that plugs into the tape drive must have a lack of the proper signal. This cable is the one that originates from the sled and interfaces via the control board on the sled. I think that is the real issue.
HH are the problem child.
Dan
We have the necessary adapters to make the connection to make the physical connection. When we power up the drive we get EC E so we cannot test with something like ITDT. We can access the serial port on the drive with custom cables and access the tape drive for serial port functions like re calibrate and look for open channels on the head. We also can access the front button pushes to semi control the drive, the E has us stuck which indicates a lack to communicate with the tape drive via the SAS interface.
We have no issues testing these drive but they must be installed a library, for the repair function we like to see what is going on so it makes it more difficult to use the library that is where the bench function for us is useful. We have test beds set up to test any type of autoloader tape drive but most of those FC and SCSI can be tested on the bench with out issues.
We can bench test the FH SAS tape drives on the bench like from a Dell ML or ADIC i500 or the IBM TL series or the HP line as well.
Your suggestion on the SAS backplanes could be a good idea but the library interface that plugs into the tape drive must have a lack of the proper signal. This cable is the one that originates from the sled and interfaces via the control board on the sled. I think that is the real issue.
HH are the problem child.
Dan
When you say they won't power on is it that they are completely off or can you still send SAS commands to them? Can you talk to it through the RS442 ACI/ADI interface?
Only differences I know between the SAS and FC drives is that SAS have CD-ROM mode (very unlikely to be on with library drives) and they start off with a different ACI level according to https://docs.oracle.com/cd/E38452_01/en/LTO6_Vol2_E1_D8/LTO6_Vol2_E1_D8.pdf.
Can't find the Ultrium 3 technical reference manuals or it would be relatively simple to compare it with the Ultrium 4 version.
Hmm, have you tried applying 3.3 or 5V to ACI_LIB_SEN_L to tell the drive it is standalone , it might just be that it floats low on the SAS drives. http://www.t10.org/ftp/t10/document.02/02-011r0.pdf
Only differences I know between the SAS and FC drives is that SAS have CD-ROM mode (very unlikely to be on with library drives) and they start off with a different ACI level according to https://docs.oracle.com/cd/E38452_01/en/LTO6_Vol2_E1_D8/LTO6_Vol2_E1_D8.pdf.
Can't find the Ultrium 3 technical reference manuals or it would be relatively simple to compare it with the Ultrium 4 version.
Hmm, have you tried applying 3.3 or 5V to ACI_LIB_SEN_L to tell the drive it is standalone , it might just be that it floats low on the SAS drives. http://www.t10.org/ftp/t10/document.02/02-011r0.pdf
ASKER
Hi Andy:
Thanks for the comments. I had posted in the previous comments that it will power on but we get the EC E code lacking communication. We do access the drive via the serial connection for various things but the EC E stops us from running ITDT.
Let me look at the links you sent and share with the other technical folks here and see if something clicks here. What we find is that certain new LTO5s SAS can be bench tested with ITDT but once we put them on a tray and the library configures the drive for a library drive, then we cannot test out of the tray.
We have the 38 pages of documentation and the command set inherent in the IBM drives, I wonder if there is some command line we can run that would clear the state of the drive personality that the drive sled or library changed when the drive was installed on the tray.
If you are familiar with the command statements on the tape drive could you point us into the correct area? We are not even sure what we should be looking for. As stated in my last response we do have access to the port and can do drive recalibration and look at the resistance on each of the data channels. The commands to do this is not limited by the EC E.
Thank You,
Dan
Thanks for the comments. I had posted in the previous comments that it will power on but we get the EC E code lacking communication. We do access the drive via the serial connection for various things but the EC E stops us from running ITDT.
Let me look at the links you sent and share with the other technical folks here and see if something clicks here. What we find is that certain new LTO5s SAS can be bench tested with ITDT but once we put them on a tray and the library configures the drive for a library drive, then we cannot test out of the tray.
We have the 38 pages of documentation and the command set inherent in the IBM drives, I wonder if there is some command line we can run that would clear the state of the drive personality that the drive sled or library changed when the drive was installed on the tray.
If you are familiar with the command statements on the tape drive could you point us into the correct area? We are not even sure what we should be looking for. As stated in my last response we do have access to the port and can do drive recalibration and look at the resistance on each of the data channels. The commands to do this is not limited by the EC E.
Thank You,
Dan
I hadn't seen your reply to Davis when I posted, but the serial connector you're using must be the diagnostic one, the one I'm talking about is the one that connects the drive to the library controller via the sled. Description is on page 26-29 of https://docs.oracle.com/cd/E38452_01/en/LTO6_Vol1_E1_D7/LTO6_Vol1_E1_D7.pdf , it says Tape drive will not appear on SCSI/FC until commanded when low , however it also says it's pulled high by the drive.
ASKER
Andy:
Just took a peek at the link. We were pretty busy yesterday so it got put on the back burner. Going to focus on that today, Looks very relevant.
Thanks
Dan
Just took a peek at the link. We were pretty busy yesterday so it got put on the back burner. Going to focus on that today, Looks very relevant.
Thanks
Dan
ASKER
Andy- Had to chance to look at HP docs very interesting. To further clarify we do not see this issue bench testing HP SAS drives out of the library trays. It is just IBM.
We have found on the newer versions of LTO some 4, most 5s and 6s that if they start out as a stand alone device we can bench test them. If we mount them in a library tray and install in a library, they will load the library personality and work in a library environment. When taking that same drive out of the tray and attempting to test on the bench, no can do.
We suspect that the library device is setting some part of the code on the tape drive when installed in a tray.
When we bench test an IBM FH tape drive out of the tray on the bench we just turn off DIP # 5 all is well. Turn on DIP# 5 to re install drive in tray and into the library.
I apologize for not clarifying my question in the first place, did take some time to get to the real issue.
Thanks
Dan
We have found on the newer versions of LTO some 4, most 5s and 6s that if they start out as a stand alone device we can bench test them. If we mount them in a library tray and install in a library, they will load the library personality and work in a library environment. When taking that same drive out of the tray and attempting to test on the bench, no can do.
We suspect that the library device is setting some part of the code on the tape drive when installed in a tray.
When we bench test an IBM FH tape drive out of the tray on the bench we just turn off DIP # 5 all is well. Turn on DIP# 5 to re install drive in tray and into the library.
I apologize for not clarifying my question in the first place, did take some time to get to the real issue.
Thanks
Dan
I see, so the half height IBM doesn't have a switch to override library mode?
http://t10.org/ftp/t10/document.02/02-022r0.pdf is out of date but as far as I can see you have to send it "Maint_Command for Device Online" through the library interface.
Also mentions "if Disable automatic drive online bit is set in non-volatile storage"... which implies that once you put it in a library the library sets the Disable automatic drive online bit in NVRAM which seems to be your problem. There may still be a connector on one of the interfaces that does the equivalent of the switch that's on the full height drive though, would be easier to pull that high or low instead of having to send it serial commands.
http://t10.org/ftp/t10/document.02/02-022r0.pdf is out of date but as far as I can see you have to send it "Maint_Command for Device Online" through the library interface.
Also mentions "if Disable automatic drive online bit is set in non-volatile storage"... which implies that once you put it in a library the library sets the Disable automatic drive online bit in NVRAM which seems to be your problem. There may still be a connector on one of the interfaces that does the equivalent of the switch that's on the full height drive though, would be easier to pull that high or low instead of having to send it serial commands.
ASKER
Andy:
Correct the HH IBM LTOs do not have DIP switches on the board. We have not attempted to access the serial library port on the drive, do you think there may be a hint on that port on what to do.
Also we are not sure if the pin outs on the library controller port is the same as the diagnostic port. It appears that the diagnostic port has 9 pins and the library port has 8 pins.
Any ideas on the pin outs, we have many of the 8 pin connectors and we just build the cables to interface.
Thanks
Dan
Correct the HH IBM LTOs do not have DIP switches on the board. We have not attempted to access the serial library port on the drive, do you think there may be a hint on that port on what to do.
Also we are not sure if the pin outs on the library controller port is the same as the diagnostic port. It appears that the diagnostic port has 9 pins and the library port has 8 pins.
Any ideas on the pin outs, we have many of the 8 pin connectors and we just build the cables to interface.
Thanks
Dan
The serial (or Ethernet) library port is the one you have to send commands to to enable the host SAS port, but it may also have a signal to override library mode like the HP port has. I don't have the spec of the IBM library port to look it up though.
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
I've requested that this question be closed as follows:
Accepted answer: 0 points for MidWestTechnicalSales's comment #a41793406
for the following reason:
I think the solution of viewing the library control port on the tape drive in question, may work. Once I do this I will re post.
Dan
Accepted answer: 0 points for MidWestTechnicalSales's comment #a41793406
for the following reason:
I think the solution of viewing the library control port on the tape drive in question, may work. Once I do this I will re post.
Dan
How can it be that you are accepting your own comment when it was me that told you that you need to use the library interface port to enabke the host ports?
As per my above comment.
Well leave it open until then then. At the moment I feel rather cheesed off at having given you so much informatiom and helping you clarify the question (which you did not explain at all well to start with) for nothing.
ASKER
To Andy and the administrative comment.
FYI this is the first time I have used this form so the protocol for comments and issue resolvement for me are foreign, I apologize if I made an error as a new user please have patience with me for my errors.
Secondly, the original issue I brought to the Exchange is an issue that all of the tape drive repair facilities have in common and we have been unable to resolve. Most of the time we are so busy with repair and sales in our businesses that we do not have the time to allocate to resolving issue to make our lives easier. We do have work arounds for this issue but it takes additional time which is limited.
I also realize that some folks sign the back of the check instead of the front of the check. So regardless of what they do or accomplish the compensation is the same. As business owners we need to allocate our resources to produce revenue, make expenses and solve and repair and sales to customer issues so time becomes a very important part of the management process.
Dan
FYI this is the first time I have used this form so the protocol for comments and issue resolvement for me are foreign, I apologize if I made an error as a new user please have patience with me for my errors.
Secondly, the original issue I brought to the Exchange is an issue that all of the tape drive repair facilities have in common and we have been unable to resolve. Most of the time we are so busy with repair and sales in our businesses that we do not have the time to allocate to resolving issue to make our lives easier. We do have work arounds for this issue but it takes additional time which is limited.
I also realize that some folks sign the back of the check instead of the front of the check. So regardless of what they do or accomplish the compensation is the same. As business owners we need to allocate our resources to produce revenue, make expenses and solve and repair and sales to customer issues so time becomes a very important part of the management process.
Dan
As a businessman do you find that customers and suppliers tell you when they put you on their list of people they won't do business with after you only provide half the information to start with and waste their time or do you just not hear from them again?