Forum Discussion

PD001's avatar
PD001
Level 5
11 years ago

Duplication to tape slow

Running a 5220 with 7.5.0.6. I'm using SLP's to replictate to a Master appliance at our DR site and to duplicate to tape, going to IBM.ULTRIUM-TD6 tape drives. Starting last week my dupliction jobs are running slow and showing records below in the job log. Not sure where else to look to troubleshoot. 

thanks

Pat

 

07/13/2014 21:36:17 - begin reading
07/13/2014 22:12:19 - end reading; read time: 0:36:02
07/13/2014 22:12:20 - begin reading
07/13/2014 22:50:45 - end reading; read time: 0:38:25
07/13/2014 22:50:46 - begin reading
07/13/2014 23:52:42 - end reading; read time: 1:01:56
07/13/2014 23:52:42 - begin reading
07/14/2014 00:41:53 - end reading; read time: 0:49:11
07/14/2014 00:41:54 - begin reading
07/14/2014 01:34:46 - end reading; read time: 0:52:52
07/14/2014 01:34:46 - begin reading
07/14/2014 02:24:04 - end reading; read time: 0:49:18

  • Have you checked you appliance web gui for errors?

    A failed disk or similar will slow it right down.

    Even worse if if the raid battery has failed .. being a 5220 it happens quite often and will slow the whole thing down to a crawl.

    In the CLISH go to the Support section and then the use:

    MegaCLi Run

    MegaCli64 -adpbbucmd -getbbustatus -a0

    Use the space bar or Enter to scroll down the list .. use q to quit the utility

    You can also do it in the O/S (Support - Maintenance):

    /opt/MegaRAID/MegaCli/MegaCli64 -adpbbucmd -getbbustatus -a0

    You are looking for the line like:

    Absolute state of charge: 64 %

    Anything less than about 97% means you could have an issue - is so run a DataCollect and get a case opened with Symantec

     

9 Replies

  • Hi,

     

    Those message are normal and happen as its starts and completes the duplication of an image. What are the transfer speeds if you look at the bptm logs? How full is your appliance? And what are your buffer setttings cofigured set to?

  • The following are from the BPTM log on my Master\Media server. Are these the tranfer speeds?

     

    write_backup: successfully wrote backup id sqlvs03_1405140678, copy 1, fragment 1, 5024 Kbytes at 758.225 Kbytes/sec

    write_backup: successfully wrote backup id sql01_1405131326, copy 1, fragment 1, 25600000 Kbytes at 2562.799 Kbytes/sec

    write_backup: successfully wrote backup id sqlvs01_1405141042, copy 1, fragment 1, 4832 Kbytes at 305.803 Kbytes/sec

    My Disk Pool is at 62% full

    How do I check my buffer sizes?

     

    Pat

  • Yes those are your transfer speeds, they are very low.

     

    You can verify the buffer settings using the below menu in the CLISH.

     

    Setting > NetBackup

     

    Have you tried a reboot recently of your appliance?

  • These are my settings and the unit was rebooted 32 days ago. 

     

     NetBackup DataBuffers Number show
    NUMBER_DATA_BUFFERS                               : 256
    NUMBER_DATA_BUFFERS_DISK                          : 512
    NUMBER_DATA_BUFFERS_FT                            : 16
    NUMBER_DATA_BUFFERS_RESTORE                       : 30

     

     NetBackup DataBuffers Size show
    SIZE_DATA_BUFFERS                                 : 262144
    SIZE_DATA_BUFFERS_DISK                            : 1048576
    SIZE_DATA_BUFFERS_FT                              : 262144

     

     Uptime
      8:05am  up 32 days 20:51,  5 users,  load average: 1.07, 1.00, 0.93

     

     

     

     

  • here is the output, 

    Thanks for the reply

     

     Uptime
      8:05am  up 32 days 20:51,  5 users,  load average: 1.07, 1.00, 0.93

     

    NetBackup DataBuffers Number show
    NUMBER_DATA_BUFFERS                               : 256
    NUMBER_DATA_BUFFERS_DISK                          : 512
    NUMBER_DATA_BUFFERS_FT                            : 16
    NUMBER_DATA_BUFFERS_RESTORE                       : 30

     

     NetBackup DataBuffers Size show
    SIZE_DATA_BUFFERS                                 : 262144
    SIZE_DATA_BUFFERS_DISK                            : 1048576
    SIZE_DATA_BUFFERS_FT                              : 262144

     

     

     

  • Hi,

     

    The buffers seem fine and are as recommended.

     

    If you can I would reboot the appliance if you can.

     

    Also take a look at this note http://www.symantec.com/docs/TECH181784

  • Thanks for all the replies, 

    have another question, maybe you could help me on. 

    I ran a backup job on a SQL Database yesterday with no other jobs running and had a KB/Sec of 100,000, today with 40 other Backups/Replication/Duplication jobs running, the same job has a KB/SEc of 10,000. Is there a log I can look at to see what might be causing the difference in speed between today and yesterday?

     

    Pat

  • Hi,

     

    The difference in speed is due to the other jobs using up bandwidth. What is the network card config in your appliance? 1gbps  or 10gbps? Have you configured link aggregation?

  • Have you checked you appliance web gui for errors?

    A failed disk or similar will slow it right down.

    Even worse if if the raid battery has failed .. being a 5220 it happens quite often and will slow the whole thing down to a crawl.

    In the CLISH go to the Support section and then the use:

    MegaCLi Run

    MegaCli64 -adpbbucmd -getbbustatus -a0

    Use the space bar or Enter to scroll down the list .. use q to quit the utility

    You can also do it in the O/S (Support - Maintenance):

    /opt/MegaRAID/MegaCli/MegaCli64 -adpbbucmd -getbbustatus -a0

    You are looking for the line like:

    Absolute state of charge: 64 %

    Anything less than about 97% means you could have an issue - is so run a DataCollect and get a case opened with Symantec