HPE MSL4048 9B 3B ( robotic failure)

[iqtc]

Dear community,

We have a tape machine HPE MSL4048 that we have been using to backup data of our organisiation until this monday 13th May with 46 LTO4 tapes and 2 LTO5 tapes. 

We recently adquired new LTO5 tapes and wanted to replace the old LTO4 with this ones.  After replacing the 46 tapes we got the following message in the machine's display:

Robotic Failure

unable to locate magazine slot

9B 3B

The MSL4048 documentation also describe this error as:

9B - Sled position sensor not found

3B - Error slot position sensor not found

After trying several things we decided to go back and remove the new LTO5 and put back again the old LTO4 tapes. We have power cycled the machine, done Wellness tests, check the magazines are ok and also the tapes, etc. But the error still remains.

We have checked also that the robotic part of the machine moves around the several tapes and scans them.  It is also able to mount one tape if requested.

For more log details, this is what is logged after powering on the machine and do a scan:

19.04.28 12:03:11.23  Libctrl task Error - HE: sled position not found
     Return status: 0x80 - command failed
     Global error code: 0x9B - HE: sled position not found
     Robotic movement error: 0x3B - unable to locate magazine slot positions
     Current command: 0x14 - scan with range
     Task originator: 0x02 - CDB interpreter task
     Extender library specifier: 0x00 - master library
     Source element type.: 0x04 - tape drive
     Element number: 0x01, 1

The machine in its display tell us that the Drive Status is correct but the Library status indicates 'Robotic Failure'

We are not sure which part of the machine is causing problems to see if it may be replaced. After searching in forums and the machine documentation we think it could be related with some of these parts:

• base chassis
• library controller
• magazines

But we don't know how to determine which of them is the problem or if there could be any other reason causing it.

Does anybody experienced a similar situation or knows what could we try?

Thank you very much for any help!

[NarayanKK-SO]

Hello,

Hope you were able to refer the following link.

https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-c01557092#N10217

The sensor failures results in robot failure. Before replacing the Chassis. Then make sure you remove the drives and magazines. Inspect the Chassis for any broken parts or obstruction in the Robot movement path.  Reseat all the components and test. If the issue repeats. You need to replace the Chassis unit in this case (Base Chassis). 

Following link will help you in removing the magazines when the unit is powered off.

https://support.hpe.com/hpsc/doc/public/display?docLocale=en_US&docId=emr_na-c01517695#N101C1

Hope this helps.

[Moderator edit: Removed the broken link. Please refer to https://support.hpe.com/]

[iqtc]

Thanks NarayanKK-SO ,

We have removed the drives and magazines and haven't seen any broken parts or obstruction to the robot movement path. After resetting all components the issue repeats. 

Just to test, we also opened the rear part (where a second drive could be placed) to see the robot behavior while working. It is apparently capable of doing all movements, and we've checked that scans all tapes one by one before giving that error.  It's been also successfully able to load/unload a tape into the drive afterwards. Although it takes a lot of time in doing it, and seems to do lots of unnecessary movements before finishing the task and some times it does not respond while requested any operation.

Is there any way to determine if the problem resides in the Base Chassis or it could also be in the Drive Controller?

[Barry_Reider]

@iqtc , given that this error occurred after removing the magazines and swapping out tapes, and the "slot position sensor not found" message; I would suggest taking a look at all of the magazines again. The first thing to check is that the magazine with the optional mailslot is in the correct location. You will notice one of the left magazines has a small white dot located in the lower right corner (see attached photo). This magazine should be in the lowest magazine slot on the left side of the library where there is another small white dot that matches up. If the magazines are reversed, remove them and swap them to the correct locations to see if the error goes away. If they are in the correct position, or this does not resolve the error, remove the magazines from the library, and remove all the tapes. When you look at the slot openings in the magazine, you will notice a clear plastic piece in the bottom middle section of each slot. These clear pieces have 6 small rectangular surfaces on each part. (see attached photo) The exception to this is - on the left side magazine where the optional mailslot is (the front 3 slots), the plastic piece is toward the rear of the slot and only has 4 rectangles. (see attached photo)

Check each of these small plastic pieces to verify they are in each slot and straight. Sometimes when bulk loading, one can be knocked loose and/or fall out, or just become out of alignment. Each of these clear pieces should be in line with the black plastic on either side of them. If any are loose, they can normally be carefully snapped back into place. If any are missing or broken, you may need a magazine replacement.

[iqtc]

Thanks Barry_Reider

We've checked thtat the magazine with the optional mailslot is in the corret location and followed your instructions to check one by one all magazines, tapes and the plastic pieces.

Everything seems to be ok,  straight and correctly fixed. Therefore we guess the magazines wouldn't be the problem and therefore we should replace the base chassis or the drive controller.

Thank you for your help,  if you think there is anything else we could check to determine if the problem is more related to the chasis or to the controller, it will be really helpful for us.

Thank you for your help

[Barry_Reider]

@iqtc, You are welcome. Before you start replacing parts, let's try two more things. On those little clear plastic pieces, wipe each of the rectangular surfaces off to make sure there is no debris, or even oil from fingers on them. Also make sure you don't leave any debris or lint from whatever you wipe them with. It would also be good to run the wellness test on the library to see if any further information is given. If the clear targets are all clean, straight, and aligned, and we are still seeing the 9B 3B failure, I would suspect the chassis over the controller. 

Since you can't post serial numbers in public forums, could you private message me with that? I'd like to understand how old this unit is. Also, if you have another library on site, I would swap magazines between the libraries and see if the error follows the magazines or stays with the library.

-Barry

[NarayanKK-SO · Accepted solution]

If the robot movement is not blocked and if there are no sensor issues. Then i am sure this could be the controller. But we would ike to know the Model# of this unit. Let us know whether this is a SCSI, SAS or FC model. On the barcode you will find the product ID. Share us that info (not the serial#). So we can provide you with the right part to replace.

[iqtc]

Thanks @Barry_Reider  and @NarayanKK-SO 

We cleaned the magazines and the clear plastic parts using a compressed air spray but the error sitll appears. This is the only tape library we own by now, so we can't test swaping magazines.

We've realized that the barcode label at the bottom of each magazine slot (the one that appears when the tape is removed from the magazines) seems to be a little bit dirty. But the dirt seems not to go away with the compressed air or anything else.  

We are currently running a Wellnes test again to check if the error continues appearing.

The product  number of our MSL4048 tape library is AJ036A bought around 2009, so it is quite old. We only need to replace its Drive some years ago (2016 or so) until now.

[NarayanKK-SO]

Please share us the results of the Wellness test. The part to replace in this case will be as follows

413509-002 - Chassis assembly

[iqtc]

The Wellness Test Kept ending with the same error. We finally bought a new chassis (keeping our drive ) and now everything seems to be working again. 

Thanks you all for your help!

[tapelibraryfixr]

im pretty sure the error comes from the sensor on the front of the gripper

[ColetteH]

The issue just happened to occur when the media is replace.   Pulling out the magazines to replace the media must have disloged something into the robots path.  I would pull out the magazines and check that the light pipes are still entact and one has not fallen out.  While magazines are still out I look for something that fell into the robots path.  Make sure the library is on rails and nothing is sitting on top of the library as it can bow down the top causing this failure.   If having a hard time seeing debris the drives could also be pulled out.