06-12-2015 05:02 AM
Hello,
We have a slow backup situation while attempting to backup a drive a VM in a flashbackup policy. The VM drive is 4 TB (1.8 TB used) RDM LUN.The backup is running extremely slow 2-3 MB/Sec having backed up just 220 GB is over 20hrs, at this speed the backup is never going to complete in time. I have adjusted the read buffer ( followed the technote mentioned below) changing it from 32 to 1024 and then 2048 but it has made no difference. Please suggest if there's anything that I can do to improve the speed.
Netbackup Master & Media Server : 7.6.0.3 Version on Windows 2008 R2
Client: 7.6.0.3 on Windows 2008 R2
https://support.symantec.com/en_US/article.HOWTO99868.html
Thank You
Solved! Go to Solution.
06-14-2015 07:08 AM
Your bottleneck could be absolutely anywhere.
.
The backup source:
Where is this 4TB LUN?
Is it from DAS, or from NAS iSCSI, or from SAN FC?
Have you checked the storage array and/or LAN network, and/or SAN fabric links for port saturation?
If the LUN is from a storage array, have you checked all characteristics of the array?
Some storage arrays require that the NTFS volume/partition be block aligned to the underlying storage array - e.g. if you let Windows 2003 create an NTFS partition itself, and you are using NetApp iSCSI or FC LUNs (on WAFL) - then it is very very likely that your NTFS partition inside your RDM LUN is not block aligned - which can have an adverse affect on performance.
.
The backup target:
Same questions as the source - pls provide details.
06-12-2015 05:03 AM
I'm see a lot of these in the job details.
11/06/2015 05:21:50 PM - Info bpbkar32(pid=4268) 17860000 entries sent to bpdbm
11/06/2015 05:22:03 PM - Info bpbkar32(pid=4268) 17955000 entries sent to bpdbm
11/06/2015 05:22:16 PM - Info bpbkar32(pid=4268) 18050000 entries sent to bpdbm
11/06/2015 05:22:30 PM - Info bpbkar32(pid=4268) 18145000 entries sent to bpdbm
11/06/2015 05:22:44 PM - Info bpbkar32(pid=4268) 18240000 entries sent to bpdbm
11/06/2015 05:22:57 PM - Info bpbkar32(pid=4268) 18335000 entries sent to bpdbm
11/06/2015 05:23:11 PM - Info bpbkar32(pid=4268) 18430000 entries sent to bpdbm
06-14-2015 07:08 AM
Your bottleneck could be absolutely anywhere.
.
The backup source:
Where is this 4TB LUN?
Is it from DAS, or from NAS iSCSI, or from SAN FC?
Have you checked the storage array and/or LAN network, and/or SAN fabric links for port saturation?
If the LUN is from a storage array, have you checked all characteristics of the array?
Some storage arrays require that the NTFS volume/partition be block aligned to the underlying storage array - e.g. if you let Windows 2003 create an NTFS partition itself, and you are using NetApp iSCSI or FC LUNs (on WAFL) - then it is very very likely that your NTFS partition inside your RDM LUN is not block aligned - which can have an adverse affect on performance.
.
The backup target:
Same questions as the source - pls provide details.
07-22-2015 12:42 AM
If this still is an issue, the MAX_ADD_FILES parameter might be worth trying.