cancel
Showing results for 
Search instead for 
Did you mean: 

error code 50

Home_224
Level 6

Hi,

I run the catalogue backup on schedule, it returns error code 50 from time to time. May I know what is the problem  error 50.jpg

1 ACCEPTED SOLUTION

Accepted Solutions

Hi Marianne,

I swap it back to the production Library, and find out the barcode start with CPxxxx , to normal status.  

Now i wish that I can get someone to assist with me the reason for run backup with error code 96 when all the STU and Volume pool already configuration. I only run the backup itself but it still return the error code 96. 

I wish that someone can advice me the situtation.  

Enjoy your holiday ! 

Thanks, 

View solution in original post

22 REPLIES 22

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

You need to look at the rest of the Job details - the actual error is the I/O error 3 minutes earlier to media id CPC004.
You need to check bptm log and /var/adm/messages for hardware errors. 

We see in your other post  that you seem to be battling with aging infrastructure - old, unsupported NBU version and possibly same as far your tape devices and media is concerned? 

Hi Marianne,

Thank you for your comment.

I attach the log for your review 

 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

I cannot open files with unknown extentions on my mobile device.
That is why I asked that you copy the files to .txt.

You may want to have a look at the logs yourselves.

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Just a hint about reading logs - we see that the error occurred on 15 March at about 11:17 and that the bptm PID was 10761.
So the bptm log file that you need to look into will be log.031518. 
The timestamp will be the 1st column followed by the PID in square brackets.

Look for entries starting with :

11:17:##:##  [10761]
The numbers between < > represents severity level:
<2>  -  debug 
<4> -  info
<8>  - warning
<16>  - error
<32>  -  severe error

Look for errors in /var/adm/messages for the same period.

Nicolai
Moderator
Moderator
Partner    VIP   

Simple media write error. Freeze the tape and re-try the operation.

# bpmedia -m CPC004 -freeze

 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Looking at @Home_224's recent posts, it seems that there are a bunch of tapes giving I/O errors, e.g 
CF0128
CF0069
CP0063
CPC004
etc.

Either bad tapes or drives.

Thank you information.

As you mention , the media error, but we use different media in different day, the problem still happen so I use another UAT library for testing if there is hardware or media problem.  But there is some problem I cannot fix it, as i follow the article ID 100021294 to replace the tape in netbackup configuration, every time i do inventory , there is some media id not in db error message, then I delete the media or bpexpdate those media , inventory again , it never come out the error message.  May I know if there is normal error message when re-connect new tape library ?

yes, it makes me crazy to handle this problem, as the tape library and media for over 4 years or over , the hardware should be second-hand to make this trouble , i call to vendor to follow up on hardware issue, but they don't know NBU operation so I gratefull to help for all of you .

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

The tape vendor does not need to know NBU. 

In your very 1st screenshot, there was an unclear photo of messages file. 
It looks like there was a hardware sense key error. Something that looks like :
  key = 0xXXasc = 0xXX, ascq = 0xXX 
Those are hardware errors that they can look up without needing to know anything about NBU.

There are other ways to test hardware as well - manually load a tape with no valid NBU images in a tape drive (you can use robtest for that). Use OS backup commands (tar / cpio / ufsdump) to write backup of files/folders to the drive. 

To replace a tape library, there are more steps than simply replacing a tape drive. 

I have on a previous occation posted these steps for library replacement:

Delete drives and robot in NBU. (This will change all tapes in robot to Standalone.)
Physically disconnect robot and drives and ensure devices are removed from OS view.
Connect new devices and ensure OS can see devices.
Run device config wizard.
If you will re-use old tapes, put tapes removed from old robot in the newly attached robot.
Run Inventory but do not update yet - Do 'Preview recommended changes'. Ensure new robot reads tape labels the SAME way as old robot (if old robot used 1st 6 chars of label, ensure that new robot does the same.) If not the same, add Media ID Generation rule.
Complete Inventory.

Hi Marianne,

I swap to the same model of library, and check the OS and NBU able to see the tape library but the barcode ID with the L3CP0012, the old barcode id sequence of CP0012L3, I check the media id generation is 1:2:3:4:5:6 , and I change it 3:4:5:6:1:2 and do inventory, the barcode still the same no change any more,  I try to run the Testing backup , it returns the code error of 96 , i check the STU and volume pool that is correct.  I have no idea if it impact for backup error.   

I try to test the moving tape media to another slot, it works on the NBU console.  Then I try to use tar -cvf to test the library, it works but tar -tvf  to view the content , it returns the blockzero.  

Last time I use the library which is working fine without problem, but now it failed to run testing backup.

 

 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
This type of media id display L3CP0012 is a library setting. You need to check documentation for your hardware to change it to CP0012L3 since this is how the media has previously been added to NBU volume database. Media Generation rule will not help here.
Otherwise you will need to use vmchange to update NBU barcode field for each media-id in the volume db.

The problem is quite slight difference.

The Media ID show CP0012 , barcode show L3CP0012 , I am not sure if the barcode dismatch to impact for run backup job . 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

The problem is EXACLTY the same. 

You are not going to resolve your issues unless you can change the robot config to read the barcode correctly with L3 at the end instead of the beginning. 

There are 2 fields in the volume database : 
Media ID - NBU uses 6 characters of the barcode label and is the main volume identifier in NBU.
Barcode - this is determined by the physical label on the media and the robot config.

So, you need to ensure that barcodes communicated by the robot remains consistent or else change them one-by-one using vmchange. e.g. 
vmchange -m CP0012 -barcode L3CP0012

As long as barcodes are reported as L3... but NBU config is ....L3, running an Inventory is going to give errors, media will not be correctly updated in NBU, and your backups will fail with status 96.

So, I am 100% sure that the barcode mismatch will impact your backup jobs.

 

Thanks Marianne, 

I already use vmchange to make sure media id and barcode are the same, and try to run backup , but it returns error code 96 

Now I try to use UAT library to run tar or ufsdump , it failed, it may be hardware on the UAT library, then i swap it back to production library, to run the same command , it works, so I re-configure the netbackup, do inventory and all the necessary task to see the NBU to know the TAPE Library, the barcode ID and media ID are normal,  I check run test backup for testing, it come up very fast to return error code 96, I have try to use different volume pool to test backup , it returns the same code 96.  I feel very sad to work over time to try my best to fix it , but it fails in final.  

I check the bptm log, it always show the message 

16:42:38.442 [7593] <4> delete_expired_media: (-) Translating EMM_ERROR_Success(0) to 0 in the NetBackup context. 

I run the testing backup 16:53 -

requesting resource Any

requesting resource  Imduarbak01. NBU_Client _MAXJOBS.IMDUARBAK01

requesting resource  Imduarbak01. NBU_Client _MAXJOBS.TESTING2

ERROR BPTM pid=9379  NBU Status:96, EMM Status: No media is available 

status: 96: unable to allocate new media for backup, storage unit has none available

I check all the STU and volume pool available , so i have no idea what happen.  

Marianne, I grateful for your help , and hope if you have see this happen. 

If there is any member who know the root cause, please kindly to assist me. 

Thank you very much 

 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

bptm logs are NOT going to help here.

You need the Inventory step to add media correctly.

PLEASE run the vmquery and vmcheckxxx commands that I have asked for yesterday.

Help us to help you. 

vmquery -m CP0012

vmcheckxxx -rn 0 -rt tld
(assuming robot number is 0)

 

Now I connect to Production library to check the backup

Except for this media id CP0012 , do i need to check others media .  It is because I have to find the end user to logon the machine , he doesn't always free time to help me logon to server to run the command.   

May I know if the vmcheckxxx  to get the result , could you please what is the next action ?

vmcheckxxx to check the library content, it is the same as robtest, I run the robtest  > s s > to show the slot which same as the NBU console slot number media   Please correct me if I wrong .

I try to arrange for end user to help logon account to get the results tomorrow morning

 

 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
We need vmquery output for any of current media-id's to confirm media-id and barcode fields.
vmcheckxxx will show us equivalent of robot inventory 'check and compare'.
Next step will depend on output.
If barcode is still reported as L3.... , you need to do vmchange command for ALL known media-ids in voldb before you run robot inventory again.
Media ID generation will need to be changed to use last 6 characters - 3:4:5:6:7:8.

Hopefully someone else can assist you from here onwards. I will be on leave until next Monday.

Hi Marianne,

I swap it back to the production Library, and find out the barcode start with CPxxxx , to normal status.  

Now i wish that I can get someone to assist with me the reason for run backup with error code 96 when all the STU and Volume pool already configuration. I only run the backup itself but it still return the error code 96. 

I wish that someone can advice me the situtation.  

Enjoy your holiday ! 

Thanks,