>and the longest frequency will win--the monthly will run.
Bob, that's what usually happens, but that's not why. The trump card is
the retention level, which usually is matched to a longer frequency, but
may not always be the case.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
As I'm sure many others can attest, SSO for NDMP works quite well.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected® ▪
http://availability.sungard.com/
P Thi
It's a good thing I didn't take that bet with Bob, though I'm quite sure
that one of the guides stated that retention level was the culprit. I went
and looked at the 5.1 and 6.0 guides and they are identical to what Darren
stated.
After reading that email from the Symantec guy, I'm more intrigue
Ok, you've rubbed my nose in it long enough ;).
BTW, I never agreed to your bet, thankfully!
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected® ▪
http://availa
Is there a way in 6.5.x to verify all the pieces of an image exist on a
disk storage unit? The bpverify command looks like it does not have
anything to do just this, only to read the entire image.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldri
For this level of effort, it's the same to create two policies to reach
your goal. You can get the streams and use the excludes to filter out what
you don't want.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
But the excludes are created from the Backup Archive Restore GUI on the
client.
You can create the policy and/or schedule granular excludes here as well.
Remember that NetBackup will only read one exclude list. What I'm getting
at is if you have a global (to the client) and policy and scheudule
For the policy you want to multi-stream, is the Allow Multiple Streams box
checked in the Policy Attributes?
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected®
We aren't sending that many tapes offsite, but I do have similar issues
with the ACS libraries hanging in eject mode until the cap is emptied and
then closed in an empty state. I really hate this operation, btw, but we
utilize the eject notification to email the operators to inform them they
ne
13 is a fun one. If you haven't resolved it yet, check the following:
Does the backup fail on incrementals as well as fulls, or only one?
Check the event log for clues to other issues going on with that server
Can you transfer a large file over your backup network successfully?
Check switch/port lo
Donald (haha, just kidding Mark, I couldn't resist),
You can also utilize multiple data streams on an Exchange backup just like
any other policy. If you have this option set, you can let NBU break all
the Storage Groups and Logs into their own stream, or you can manage this
yourself. I find I p
The Symantec Certification site has a sample test and guidelines for what
you need to know.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected® ▪
http://availa
To use the drives for regular backups, you have to present the drives to
the NetApp and the Media Server(s). The SSO Option also has to be
installed for this drive sharing to be enabled.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite
If the other suggestions don't pan out, check that the GUI matches the
master version. If you aren't using the correctly patched version of the
GUI you can have issues with Storage Units, Volume Pool, etc. selection.
Use the command bppllist to print out the configuration for that policy to
see
I'm trying to get Vault to generate emails on a 6.5.3 Windows
configuration and it isn't working. I used this technote to configure the
NetBackup side of things:
http://seer.entsupport.symantec.com/docs/254809.htm
I am able to get the server to send an email using blat from the command
line, s
Dave,
I would recommend installing your new server at 5.0MP7 and switching over
to it. Our upgrade path for new hardware has been to install NBU on the
new server, then copy a tarball of /usr/openv over and untar it on the new
server. Then swap IPs, hostnames, cabling, etc. and you're done. It
Just wanted to follow up with the solution on this. Blat was having
problems parsing the information passed to it and using the quotes already
in place around the subject variable, and thus failed. Removing the quotes
resolved the issue and mail is working. Blat version is 2.6.2, build Feb
25 2
I use iostat -xzn #, where -x shows disk, -z omits devices doing nothing
(having all 0's) and -n shows names of the devices.
Tapes are included in this output, but this is for Solaris.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200
Michael,
Thanks for the reply. I actually got it working and forgot to update the
list.
My solution was to install Cygwin, configure blat as the email provider,
edit the nbmail.cmd script, and make changes to our standard Unix scratch
script to work in Cygwin.
Rusty Major, MCSE, BCFP, VCS ▪ S
I'm trying to utilize a Java GUI to manage a Windows 6.5.3 environment.
When the auth.conf file is present, and having proper name entries, I get
an error that the username is not listed. If I remove/rename the file, the
Java Console loads everything and seems just fine, as long as I'm using a
Dave,
I can't tell you how glad I am to see someone else with this issue.
Since 6.x, we've had some really strange issues with the Java GUI that
we've chalked up mostly to version differences througout our environment
(we managed quite a few different NBU domains geographically dispersed).
We
Steven,
This worked!!! Thank you VERY much for the suggestion!
I need to check my other environments much closer, but so far it seems ok.
Dave, give this a try! Make sure you have both shortname and FQDN
resolving on the Java GUI workstation.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer
We've checked out everything and all the 6.5.3 environments work with
6.5.4. A few more days of administration and I'll officially support this
'fix' but I think it's going to work.
Note that the 6.5.4 install will delete your 5.1 (and possibly older) Java
GUI shortcuts if you have them - you'l
Steve,
We have several libries, including an SL3000, a few SL8500s and i2000s,
among others.
The SL3000 and i2000 have been rock solid. The SL3000 offers an additional
longbot for redundancy, which is nice if you can afford it. The i2000 does
not have this option.
Either way, I think you'll be
Dave,
You could try using WinZip, WinRar, 7zip, or even NTBackup to create a
flat file of all of these files for you. At a previous job we had a
similarly designed application that created tons of directories for each
client entry. Of course this was pretty static data, but it took a few
hours
Scalar 1000's were/are notorious for this. Also the dreaded "Dropped
Cartridge" error where the entire library stops functioning because it
thought it dropped a cart when nothing is wrong.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suit
You are losing connectivity to your robot. Check how this is connected,
SCSI, FC, ACSLS/IP and verify that component's operation.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and I
For that specific issue you can monitor messages on your robot control
host for something about the robot going offline or losing connection. If
you look you should be able to still see it, assuming it hasn't rolled
out.
When it comes back online you'll have something like Robot Going to UP
sta
I just completed another upgrade from 5.1 to 6.5 and ran into another
issue with getting the Java GUI Console to operate. I ended up adding the
FQDN as an alias to the shortname in my hosts file and the problem went
away. I would also recommend this as a first step for any issues you have
with
If your operators do not tell Iron Mountain when to bring them back, or
uses the "Infinite" date, IM just keeps them and continues to charge you
for it.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4
Simon,
Here's a link to an archive of a post that Tanuja K. (most awesome SAN
client engineer ever!) gave about 4 months ago. I would also recommend
checking the 6.5.4 updates as well.
http://www.mail-archive.com/veritas-bu@mailman.eng.auburn.edu/msg22839.html
Rusty Major, MCSE, BCFP, VCS ▪ Sr
I would not advise using an agent to backup to a different version Media
server, especially one so far downrev as you have described.
Your problem sounds like you have something else going on in your
environment. A 54 means it cannot reach the client at all and it seems
strange that it worked a
You also need to ensure that the filer can resolve your backup server, and
if I recall correctly, NDMP control traffic goes over ports 1 and
10001 so these need to be open.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houst
Licensing, for one. The reduced administration would be another big plus
for me to look at SAN clients.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected® ▪
h
While we do a mix of both, I personally like the "backup to a file on disk
then backup that file" method. If you've got the space to keep a file or
three on disk, I find this works best as I usually don't have problems
with having to support the agent and dealing with the finger pointing
betwee
You could install and use cygwin to run Nix scripts on windows.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected® ▪
http://availability.sungard.com/
P Think
I'm seeing the same issue at one site and I'm curious if your drive had a
stuck tape, or for some other reason couldn't move the tape from the drive
to the home cell?
The problem I see is the tape gets stuck in the drive, cannot be unloaded,
which causes NBU to constantly request another mount
I don't think you would like the results of this solution. As you know, a
snapshot of the vm is copied to the backup proxy's holding tank and it
isn't released on the ESX server until this is completed. I'm pretty sure
you would end up with snapshots being released before the copy process is
do
Mark, you might consider Synthetic backups if you can't get around the
bandwidth limitation. The first one would take a long, but after that,
it's incrementals forever.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77
That's crazy. It sounds like they are digging for more info or trying to
get you to convert to VBR.
We are on this licensing model and the data we send them is via Aptare. I
don't understand why they are making you do that.
Sorry, that was of no help to you. We installed an older version of VBR
Boris,
VBR will not be included in 7.0. There will be some reporting included in
Ops Center which may be new reports to the base NetBackup product, but
there will also be another option for additional, advanced, reporting -
this is probably the stuff most of us would want.
My XSP licensing als
I understand that, Boris. My point was that OpsCenter will not have the
advanced reporting (I forget what Symantec is calling the additional
module) that you are used to today with Aptare. I do not know if the
freebie VBR license will allow you to utilize this module or not once 7.0
comes about
Mark,
First, all I could think of was the opening to the Electric Company where
a lady yells "HEY YOU GUYS"
For those lost in the 80's like me:
http://www.youtube.com/watch?v=TFYMijdQ_sA
Anyway, I have not tried this in a while, but on the BAR client side GUI,
select File>Specifcy NetBacku
Did you update the Oracle server agents to the same major and minor level
as the master/media servers? You didn't say if you went to 6.5.0 or
applied any patches? I would at least patch to 6.5.3.1 or higher.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 7
I agree, lifecycle is better as it will guarantee what you have requested
to happen will happen before the images will expire. If you have a vault
process fail, you may not get the images duplicated if you don't rerun it
manually. All you'll have to do in Vault is setup the eject.
Rusty Major,
The PBX binaries are on both DVD1 and DVD2 of your install media (pick
one). I believe 6.5.4 patches PBX, but if you have an older version that
isn't compatible with the 6.5.4 patch, you need to upgrade it from the
install media.
To install or upgrade your existing PBX, go to the appropriate NB_
The binaries for Exchange and MSSQL are also included in the Windows
client, but you have to license them before using, so this model is not
new. It would be nice if Symantec clarified what is and is not included,
though. It would be nice if they cleaned up the licensing model, too,
while they
Yep, 6.5 is very picky about this. So is the Java GUI. Adding the FQDN to
the hosts file has resolved a number of connection problems with the Java
GUI.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4
I agree, Ed. The same rep did tell me that the Java GUI will eventually go
away. And since Ops Center will be/is web based, I would assume that would
make the Windows Admin Console go away as well. I was glad to hear that
the GUIs will be sticking around, as I had thought we were getting forced
I'm trying to get bplist to give me output on some images made for an SAP
backup. All I keep getting is EXIT STATUS 227: no entity was found.
These backups run daily and I know I have images, I just want to know the
filelist for each one. I have used the following syntax:
bplist -C -R 999 /
An
Disregard. I had to put the policy type in. I now see that the default is
Windows (13)...which makes sense, these aren't windows backups!
bplist -C clientname -t 17 -R 999 /
The above does work for me. I thought I had tried it, but apparently not.
Thanks to those who responded.
Rusty Major, M
This is a good idea, I like it.
While not applicable to this request, I would also document all of the
images being expired so that in case you do mess up, you have the
bpimagelist output that you can use to go back and unexpire, import, or
have on hand to know which tapes need to be held out o
I believe most people on the list are using a backup reporting tool. I use
Aptare, but there are others and you can also script it to provide you
success rates.
Those who are watching their success rates are in the 90's. Those who are
not think their rate is better than what it really is.
Rust
You can import the LTO1 media (using the LTO3 drives, obviously) into the
new catalogs. Hopefully you know which tapes the data is on, otherwise you
have to import each tape until you find what you need. Before you insert
these tapes into the drive, make sure you write protect them, and have
th
Bart,
We have i2000s, SL8500s and SL3000s, all over 700 slots.
All are good libraries, but the SL libraries have capability of redundant
robots. The SL8500 is a huge beast of a library, but that's nothing new to
you with a Powerderhorn. The SL libraries can also be hooked up with ACS
(the 8500
You can run bptestbpcd from the master and media servers to verify
connectivity.
bptestbpcd -host -debug
In place of -debug, you can use -verbose, or no option at all, but they
get gradually less detailed respectively.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availabilit
I just exclude the images directory as that's the bulk of the size. There
are also other configuration files and whatnot scattered all over that I'm
not sure are captured in the hot backup. The size of the rest of the data
is inconsequential to me.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Eng
This is more of a limitation on the OS and the wildcards it supports than
NBU.
You can try *:\*\test and see if it works.
Alternatively, if the local drives and dir structure is static, you can
list the paths as such:
C:\*\test
E:\*\test
in the policy file list. Even though only one path is vali
If all you want is the listing of the clients, you just need the directory
structure for the clients on the Master, you don't actually have to
install all the client binaries if you don't want to.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldri
If you're on Windows, you can install Cygwin and then use grep/awk if you
want to go that route, but the scripts others have provided are probably
better. If this is something you need on a regular basis, you might
consider a third party reporting tool (Aptare, VBR, etc.).
NOM, which I think is
I'm trying to get NBU 7 installed on my test environment and am having
trouble getting the downloaded file(s) prepped. The install guide for Unix
claims "A Readme file on FileConnect describes the following points:",
however I see no readme file. I've tried to uncompress and extract them,
but t
Thanks, all, for the Windows document (Which I do see on FileConnect), but
I was looking for specific Unix instructions. I ended up getting it worked
out, but asked Symantec about the entry in the guide whice alludes to
another Readme for Unix:
The ESDimages forNetBackup 7.0 are available for d
In 6.x, all you should need is 13724 (vnetd), but have a look at the "Port
Usage Guide for Windows and UNIX Platform ":
http://seer.entsupport.symantec.com/docs/281623.htm.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston T
Heading off topic, but for the benefit of those who had a heart-attack
thinking they were out of support with 5.1 clients as they ARE supported
with NBU 6.x Master/Media servers. NBU has supported two major versions
back for as long as I can remember, perhaps that's the unspoken rule as I
don't
We have a NetApp filer that has a few TB of data made up largely of
millions of small files (about 30 million or so) and we are using several
NDMP policies to back up this data. The two main problems are length of
time it takes to backup (we usually have 2-3 backups running all day every
day) a
If the devices are scsi attached, you can take the system down to the OK
prompt and run probe-scsi-all to verify that at the OS level, the drives
are being presented. This page tells you how to do it:
http://www.tech-recipes.com/rx/114/troubleshoot-scsi-devices-with-probe-scsi-all/
Alternatively
I've just upgraded to 6.5.6 and while everything said it was successful, I
can no longer log in with the Windows Java GUI. It sits at the login
screen with "Login in progress" in the status bar. I have re-applied the
patch (on top of 6.5.6), removed the 6.5.6 JAV patch (so it's at 6.5.5)
and it
I've never heard of that switch. Anytime you try to connect to one other
the other, you get an error, therefore we had the other java version
installed as well for when we needed it.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200
Hello everyone,
I'm new to VMware and I'm setting up an environment that consists of
virtualized NBU 7 Master (Linux) and Media (Win2k8) servers that backup a
fully virtualized environment using the new vStorage APIs. There are
currently only about 20 VMs I'm backing up currently, but every nig
There is or was an 'issue' with NDMP where the dump levels for
incrementals rolled past 9 and went back to 0, causing a full and it would
stay at dump level 0 until the next scheduled full ran. I cannot find a
technote or any documentation on it, but I believe it was related to
cumulative incs
Not a cooler way, but if you want to run multiple Java GUIs on the same
workstation, just do the base install into different directories. I
typically make them \Java\60, \Java\65, etc.
You woun't have any issues and the patches can find the appropriate
version and install without issue as well.
I can confirm the same command worked for me in a NBU 7.0 environment on
Solaris 10 Master. I also checked the same command on a Windows 7.0 media
server and it worked as well.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houst
You can use SAN if you wish. You can use it if the client is setup to use
FT, or if using VCB/VADP you can select the SAN Transfer Type option, or
it *should* automatically select it if available when All Types is
selected.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availabi
After messing with it for awhile last week, it appears that it was two
things:
-Name resolution (VADP likes FQDNs)
-Botched client install
This particular VM was migrated from a 6.5.x environment and the
implementation team installed NBU 7 client on top of it and apparently it
did not take, whi
DSSU is a relocation (copy, if you will), which is why you don't have
those advanced options. The bpduplicate command is what you want to use
outside of DSSU relocation or Vault (or SLP if you go that way). I would
recommend using Vault to do the additional copies and retention changes.
It's au
This is a basic list of what I've done this in the past:
1) Configure new server to match your old server as much as possible and
where applicable (interface configuration, filesystems, NBU patch level,
tape config, etc.).
2) Shut down NBU on original
3) Tar up the entire /usr/openv directory
4)
Has anyone seen intermittent 219 errors (storage unit is not defined in
EMM) on NBU 7 due to storage units changing from their defined settings?
When viewing the overview of the policy (either at CLI or Java GUI), the
storage unit is set properly in the schedule (overriding the Attributes
setti
In the Windows BAR GUI the default view is the colored dots timeline. I've
noticed in this view sometimes that it does not show all the images. I
switch it to the 'listing' view by deselecting the timeline option under
View>Show NBU History as a Timeline and I get everything I'm expecting.
Rust
I have a NBU 7.0 domain configured backing up about 60 VMs using VADP
going to Data Domain. It's been running for about 3 weeks now and I'm
seeing dedupe rates just over 25x (~95+%). We are currently running daily
incrementals (2wk retention) and weekly fulls (1month retention) which are
mostly
It sounds like you have some jobs that are not destaging properly. I saw
this whith 5.1 jobs after I upgraded from 5.1 to 6.5 and had to manually
expire quite a few images from disk (after verifiying we had a copy on
tape, of course).
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGar
I first did an ls -l and grabbed the image ID of the images I thought were
too old and should have been destaged and expired off of the DSSU.
I then took those images and ran them through bpimagelist to see if they
had a copy on tape (or if it should have been expired).
Those that had a copy on t
I would agree, try bonding/teaming the NICs together to get more
bandwidth. Don't forget to investigate your network to make sure that you
don't have any other bottlenecks inline and that the clients are all set
to 100/full (assuming they are 100Mbit).
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage
Unfortunately I don't have much experience with buffer settings on
Windows, but I have heard that Windows drivers limits the size of the
buffer. I also heard that had been resolved/increased but I can't say for
sure. I'm really surprised you can't get above the default (which I think
is 64k?),
Mark,
No, SLPs are only aware of the backups they make, so you can't do what
you're looking for unless it does the initial backup. I would find that
very handy as well, so I hope Symantec is listening.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N
I don't have a script, but for those that might, it would be helpful to
know what platform you're on.
Unix can use iostat -xzn # (where # is the interval).
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-58
Firewall? Try to telnet vnetd from the client to the Master and
Media servers.
You can also try using this command:
bptestbpcd -client -debug
Run this from Master and Media servers. Debug gives you the most output
and isn't documented, -verbose gives you a little less output, and no
switch gi
I would agree to try to patch as there were some issues with the early
releases and calendar based scheds. Calendar scheduling is the bane of
NetBackup engineering.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079
If you have all the tapes in the library, or at least all the tapes that
the images span, then yes, NBU will work it all out for you. If not you'll
have to get tapes in quickly. This is what Ed was alluding to with "a lot
of groundwork".
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ Sun
Which lockfile, there is more than one. What is your specific error or
circumstance that you're seeing this.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected®
I would recommend moving away from the Virtual Tape Library and utilizing
Advanced Disk, Disk Pools and Storage LifeCycle Policies. Most "VTLs"
today can be utilized as a giant pool of disks, instead of a bunch of fake
tape drives, as well. But, at that point, you could remove the
complication
All of our clients are set for 30 minutes.
There is very little performance impact on the backup, I don't even think
you could measure it, as the checkpoint is a log entry on the client side.
It's not kept on tape as someone thought.
FWIW, for all of our jobs that we have to break up, we create
We have several D2D2T system currently, running Advanced Disk and SLPs. I
would never go back if I had the choice. There's nothing wrong with disk,
in fact it cures many of the ills of a tape system, like slow backups
which kill tape drives, speed to restore, etc. Even with dedupe, tape is
stil
I'm having issues with Data Domain optimized duplication and wondered if
anyone here had seen any issues like this. I've got a case with both
Symantec and DataDomain but thought I'd check here too.
Environment:
NBU 7.0 on Linux RHEL5, Nine Windows 2008 R2 Media servers (Master and
Medias are al
We have appx 1.2PB spread out across 9 NBU domains that we use just for a
backup target. The goal is to have 7-10 days of all backup images
available on disk at these larger sites. These arrays are all mid-range,
big vendor names, all SATA and can handle running 150 or so jobs
simultaneously. W
Thanks, Wayne. I didn't mean to exclude any vendors.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
Keeping People and Information Connected® ▪
http://availability.sungard.com/
P Think before you
You're probably running into disk IO issues with all of the other data.
Are you reading the data off the same EVA array that you are writing it
to?
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
281-584-4693
K
I don't have any personal experience on this, but have been told that SATA
will work fine for dedupe. Data Domain and the new NetBackup appliances,
and I'm sure all of the other dedupe hardware vendors, use SATA drives.
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability
The API that makes it happen is OST and the 'scheduler', if you will, is a
Storage Lifecycle Policy.
The high level configuration is to setup two storage servers, two disk
pools, and two storage units (one for each dedup device). You then
configure a Storage Lifecycle Policy to do the initial b
Ahh yes, good one Nic, I had forgot about that. Here is a blog entry with
more details:
http://netbackup.wordpress.com/2009/03/22/media-in-use-when-it-really-isnt/
Rusty Major, MCSE, BCFP, VCS ▪ Sr. Storage Engineer ▪ SunGard
Availability Services ▪ 757 N. Eldridge Suite 200, Houston TX 77079 ▪
This is not happening to me, but I am at 7.0.0 (with an EEB for VM backup
issues).
Have you enabled the Block Level Incremental Backup checkbox? If this is
not enabled, it will always do a full backup. If it's not enabled now, you
will have to do two backups to start seeing the benefit of BLIB.
1 - 100 of 148 matches
Mail list logo
