• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 529
  • Last Modified:

How to increate rate of backup on 1 server decrease?

I am running backup exec 11d. I have a total of 6 servers gettings backed up. Recently the backups starting timing outs because the jobs are taking longer than they are suppose to take. The only thing I noticed was that all servers have had similar job rates as far as (MB/Min) except 1 server, which has showed a decrease going from 1000 MB/MIN rate to 500 MB/MIN. The only thing that recently changed, was that we decided to Virtualize that same server a few weeks ago. It was our Exchange server. Any help would be really appreciated. Thanks.
0
fstinc
Asked:
fstinc
  • 10
  • 7
  • 6
4 Solutions
 
Justin OwensITIL Problem ManagerCommented:
For clarification, are you asking for help in increasing your Exchange server's max output or help in diagnosing/resolving your backup job's error?

Other helpful info:

OS of the Backup server and the Exchange Server
Version of Exchange
OS and application of the Host server for the VM (ESX, ESXi, Windows OS, VMWare, HyperV, etc.)

DrUltima
0
 
Ron MalmsteadInformation Services ManagerCommented:
2 questions...

1) are you doing "brick-level" backups of exchange ?
2) are you backing up directly to Tape drive or are you using "backup-disk" folders ?


Advice,
Brick level backups of exchange are much slower than simply backing up the store.
The time it takes to back them up grows as users mail items grow.  You gain the ability to restore individual e-mails with bricklevel however, so depending on how often you are asked to restore individual items, it may or may not be worth the extra weight on the BE job.

If you are using Tape, you can only backup with 1 job at a time.   This can bottleneck the whole solution when the amount of data / backup time window, exceeds the rate of transfer from servers > to the tape drive .  If you use B2d(backup to disk), you can run mutliple jobs simultaneously, thereby speeding up the entire process.   You can then dump your b2d folders to the tape during the workday, without interfereing with any other processes/data that may be in use on those servers.
0
 
fstincAuthor Commented:
Ok. Help diagnosing/resolving the issue.

OS: Windows Server 2003 R2
Exchange Server 2007 - 64 Bit Windows Server 2003
Running Vcenter 2.5 on Windows Server 2003


0
Concerto's Cloud Advisory Services

Want to avoid the missteps to gaining all the benefits of the cloud? Learn more about the different assessment options from our Cloud Advisory team.

 
fstincAuthor Commented:
We backup to tape. Our Exchange Server Storage groups are the first to get get backups up. The 1st storage  group gets backed up and then the 2nd storage starts after that backup. This was working fine up until roughly 2 weeks ago.
0
 
Justin OwensITIL Problem ManagerCommented:
Did it "work fine" after it was P2V?
0
 
fstincAuthor Commented:
In a sense. The rate has slowly been decreasing through time
0
 
Ron MalmsteadInformation Services ManagerCommented:
...dumb question...

Do you have anti-virus running on the server ?

Does shutting it down temporarily change the transfer rate ?
0
 
fstincAuthor Commented:
Actually, yes, we had virus issues and installed the virus scanner a few weeks ago also on our exchange server. We thought it might be the anti virus but did I did not think that it would affect the transfer rate.
0
 
Justin OwensITIL Problem ManagerCommented:
Antivirus on Exchange is very, very persnickety.  Have you verified, beyond a shadow of a doubt, that your exclusions are set up for both on access (real time) and scheduled tasks?

DrUltima
0
 
fstincAuthor Commented:
In what way would you suggest that I configure the antivirus scanner?
0
 
fstincAuthor Commented:
I was thinking about excluding scanning the 2 storage groups. Or will this not affect the job rate?
0
 
Ron MalmsteadInformation Services ManagerCommented:
You may either A) want to exlude certain directories from scanning... or B) shutdown realtime scanning (preferrably with a script), until the backup finishes.
0
 
Justin OwensITIL Problem ManagerCommented:
I answered a Question a while ago about AV exclusions for Exchange 2003.  Let me find it....  It will be similar.

DrUltima
0
 
Ron MalmsteadInformation Services ManagerCommented:
persnickety,.... now there's a word you don't see everyday.  lol
0
 
Justin OwensITIL Problem ManagerCommented:
Every once in a while I can pull out a good one...

In typical fashion (for me), I cannot find the Question I answered sometime in 2009. :)

Here is MS link to best practice for AV exclusions for Exchange 2007:

http://technet.microsoft.com/en-us/library/bb332342%28EXCHG.80%29.aspx

You really, really need to follow this (regardless of your AV vendor) or you will take a serious hit on performance.

DrUltima
0
 
fstincAuthor Commented:
I don't even know what "persnickety" means, I think I have an idea. LoL . Thanks I will try the exclusions, I appreciate the help.
0
 
fstincAuthor Commented:
Hi, Looks like the scanner was not the problem. The decrease in Backup occured when Exchange was virtualized. The date matches exactly. The ESX box connected to Gigabit switch. How would you recommend I resolve this. Thanks.
0
 
Justin OwensITIL Problem ManagerCommented:
Is it a single NIC box, or does the Exchange VM have a dedicated NIC?  If the max output you are seeing is half of expected, the probably cause is a VM host misconfiguration which disallows full usage of the (either physical or virtual) NIC.

DrUltima
0
 
Ron MalmsteadInformation Services ManagerCommented:
What else is going on on the box you have this virtual server on ?

Is it possible you don't have enough horse power in the server (with other stuff going on), to add another layer of abstraction which is virtualization ?
0
 
fstincAuthor Commented:
The VM has its own dedicated NIC. I says "VM DMZ" under network label. The other option is "VM Inside". I am currently looking at disk read rates and comparing. I am thinking Disk speeds?
0
 
Justin OwensITIL Problem ManagerCommented:
I don't think Disk speeds are going to provide an exact 50% reduction.  VMWare uses its internal NIC mechanism as well as the physical NIC.  I am not a VMWare expert and don't have it in my lab, but if memory is correct, there should be a way to look at the physical NIC assigned to your Exchange server and make sure it is set for 1000/Full.

DrUltima
0
 
Ron MalmsteadInformation Services ManagerCommented:
I'm talking more about the amount of CPU usage and memory usage on the Hosting server, when a back up is occurring.

Does the server have enough resources in general to host a VM running exchange, in addition to other services or servers it may be hosting.
0
 
fstincAuthor Commented:
ESXi Servers are running on Dell PowerEdge 1950.
SAS Drives of Hitachi: 73G, SAS, 15 K, HIT VIPER
DataStore: Sata Raid 5 1.36 TB capacity  
Free: 172.72 GB
xuserx2000: I did not get a chance to look at the performance last night when the backup was occuring. I will try again today.

DrUltima: I am still trying to verify that their is one NIC assigned to Exchange VM.
0

Featured Post

Efficient way to get backups off site to Azure

This user guide provides instructions on how to deploy and configure both a StoneFly Scale Out NAS Enterprise Cloud Drive virtual machine and Veeam Cloud Connect in the Microsoft Azure Cloud.

  • 10
  • 7
  • 6
Tackle projects and never again get stuck behind a technical roadblock.
Join Now