cancel
Showing results for 
Search instead for 
Did you mean: 

Error 84

Hendrik_Bezuide
Level 4

I am getting this error, now when running netbuakcup 7.6.0.2 (Windows 8)

I see out of the 34.48TB, 32.32Tb (93.70%) is been used, there are more than 500 slp that still must run,

How can I get space on ths dirve.

09/12/2016 20:12:20 - Info nbjm (pid=4368) starting backup job (jobid=406172) for client PRASAGPPSCCM01.prasa.com, policy PRASA_VMWARE_13, schedule DAILY
09/12/2016 20:12:20 - estimated 90648451 kbytes needed
09/12/2016 20:12:20 - Info nbjm (pid=4368) started backup (backupid=PRASAGPPSCCM01.prasa.com_1473703940) job for client PRASAGPPSCCM01.prasa.com, policy PRASA_VMWARE_13, schedule DAILY on storage unit prasagppbck06_disk_stu
09/12/2016 20:12:22 - started process bpbrm (pid=976)
09/12/2016 20:12:23 - Info bpbrm (pid=976) PRASAGPPSCCM01.prasa.com is the host to backup data from
09/12/2016 20:12:23 - Info bpbrm (pid=976) reading file list for client
09/12/2016 20:12:23 - Info bpbrm (pid=976) starting bpbkar32 on client
09/12/2016 20:12:23 - connecting
09/12/2016 20:12:23 - connected; connect time: 0:00:00
09/12/2016 20:12:24 - Info bpbkar32 (pid=11420) Backup started
09/12/2016 20:12:24 - Info bpbkar32 (pid=11420) archive bit processing:<enabled>
09/12/2016 20:12:24 - Info bptm (pid=7196) start
09/12/2016 20:12:24 - Info bptm (pid=7196) using 262144 data buffer size
09/12/2016 20:12:24 - Info bptm (pid=7196) setting receive network buffer to 1049600 bytes
09/12/2016 20:12:24 - Info bptm (pid=7196) using 30 data buffers
09/12/2016 20:12:25 - Info bptm (pid=7196) start backup
09/12/2016 20:14:58 - begin writing
09/12/2016 20:18:03 - Info bpbkar32 (pid=11420) INF - Transport Type =  nbd
09/12/2016 20:19:04 - Warning bptm (pid=7196) disk pool prasagppbck06_hp_disk_pool volume NBU_Store is full: processing disk full condition
09/12/2016 20:19:36 - Critical bptm (pid=7196) image write failed: error 2060022: software error
09/12/2016 20:22:11 - Error bptm (pid=7196) cannot write image to disk, Invalid argument
09/12/2016 20:22:11 - Error bpbrm (pid=976) from client PRASAGPPSCCM01.prasa.com: ERR - tar file write error (14)
09/12/2016 20:22:11 - Info bptm (pid=7196) EXITING with status 84 <----------
09/12/2016 20:22:12 - Info bpbkar32 (pid=11420) bpbkar waited 62 times for empty buffer, delayed 6002 times.
09/12/2016 20:22:14 - Info prasagppbck06.prasa.com (pid=7196) StorageServer=hp-StoreOnceCatalyst:prasagppd2d01.prasa.com; Report=scanned: 1024 KB, CR sent: 1024 KB, dedup: 14.45%
09/12/2016 20:22:14 - Error bpbrm (pid=976) could not send server status message
09/12/2016 20:22:14 - Critical bpbrm (pid=976) unexpected termination of client PRASAGPPSCCM01.prasa.com
09/12/2016 20:22:15 - Error bpbrm (pid=976) cannot send mail to hendrik@apx.co.za
09/12/2016 20:22:15 - Info bpbkar32 (pid=0) done. status: 84: media write error
09/12/2016 20:22:15 - end writing; write time: 0:07:17
media write error  (84)

1 ACCEPTED SOLUTION

Accepted Solutions

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
You need to check SLP config and settings to see why duplications are not completing fast enough.

Disk copies cannot expire and cleanup as long as duplication has not completed.

Without more space on your StoreOnce and duplications not going through, there is nothing to do for this backup. Other than writing it straight to tape... but that will then content for resources when you don't have enough tape drives for duplication....

View solution in original post

3 REPLIES 3

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified
You need to check SLP config and settings to see why duplications are not completing fast enough.

Disk copies cannot expire and cleanup as long as duplication has not completed.

Without more space on your StoreOnce and duplications not going through, there is nothing to do for this backup. Other than writing it straight to tape... but that will then content for resources when you don't have enough tape drives for duplication....

Thank You Mariane

My tape drivers is on 1GB and my Backup to disk is on 10GB, did ask my Manager that we need to buy two extra tape drives, but so far they don't want to, hopefully now they will

 

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Hopefully your tape drive is attached to 4 or 8 Gb FC.... 

Nevertheless - 1 tape drive is probably not sufficient.

If you need to convince your management that another tape drive is needed, you will need to start collecting stats.

Download the SLP Best Practice Guide: http://www.veritas.com/docs/000018455 

Read through the doc and note how to check for backlogs. 
Read about tuning SLP parameters.
See if making the SLP bactch size helps with performance.
See if changed parameters has an effect on backlog.
Record each change and effect on backlog. 

If backlog is increasing instead of decreasing and you can have proof of backlog (from everyday nbstlutil output) and failed backups due to disk full condition, you may be able to justify that another tape drive is needed. 

PS: 
If your Veritas reseller is also a reseller for tape, they may be able to assist you with this justification...