Jim,
I've been testing it myself. I think you do need to keep that full backup
around. The "synthetic" part of a synthetic backup is the consolidation of the
full backup and the incrementals into a single image. This is done via
duplication.
The test scenario I'm working with is the follow
I'm starting to mess around myself. I've not made any changes to the default
parameters for the LIFECYCLE_PARAMETERS. Here's the defaults as defined in the
manual:
MIN_KB_SIZE_PER_DUPLICATION_JOB 8192
MAX_KB_SIZE_PER_DUPLICATION_JOB=25600
MAX_MINUTES_TIL_FORCE_SMALL_DUPLICATION_JOB 30
The cl
One thing to keep in mind that it is critical to go through the steps of
running catalog consistency tests. I had tons of problems with this (100+ GB
catalog space). And never got the catalog into a state that we could actually
upgrade our 5.1 MP7 in place. It would have taken us an estimate
It will reset the expiration to the future. Think of if like the images on
that tape were all backed up at the time of the import. The expiration date is
then recalculated based upon that import date. It's a pain but if you are
aware of if you can work around it.
I've frozen the media (susp
Jeff,
I've had some experience with the Celera NAS devices. There is a parameter you
need to set that must match the SIZE_DATA_BUFFERS_NDMP. I changed the
SIZE_DATA_BUFFERS_NDMP without configuring the Celera and all my NDMP backups
started failing. I'm sorry, but I don't remember the param
Simon,
Yes, I did choose a completely new name for the master server.
Tom
+--
|This was sent by t...@tjsimerson.org via Backup Central.
|Forward SPAM to ab...@backupcentral.com.
+
I've been doing NDMP backups to our EMC NS700 without this configuration file.
The defaults are assumed if this file does not exist.
On that same subject, anyone have any suggestions for tuning an older EMC
Celerra environment? I would like to get my tape buffers up to 256K but can't
make he
Thanks to everyone's responses. The readWriteBlockSizeInKB was the parameter I
was going to focus on. Just wanted to make sure that I got the right one the
first time. My SAN admin guys get a little snotty when we don't get it right
the first time javascript:emoticon(':)')
+
I've been doing the same thing for years using various incarnations of Perl
scripts building a bid file as Jonathan suggested. But I recently discovered
the Storage Lifecycle Policies. These are fantastic! Once I worked out the
details, the SLP's will create the primary copy to VTL and then
Mark,
Don't know anything about clustering master servers, I'll put in my two cents
worth on the master server migration.
Since you're going to a *nix to a *nix environment, you should be able to
simply do a disaster recovery type restore of NetBackup. The process is pretty
much like this:
Take a look at the following link that did some actual testing against the big
4 LTO media vendors.
http://searchstorage.techtarget.com/magPrintFriendly/0,293813,sid5_gci1258158,00.html
I've almost always used Fuji or Maxell and had very few problems with the
media. This was true even at the
I remember running into this and beating my head soft :(
I suspect that you downloaded a copy of the installation files from the
Symantec web site. For whatever reason, the installatin "image" from there
does not have all the permissions set correctly for *nix systems. It looked
like all t
I've seen this also. As much as people want to make Windows and UNIX play
nicely together, there are just somethings that don't translate between the
systems. The file system and file structure is one huge difference.
I would see similar issues when logging into our AIX master server using th
I had setup the exact same type of backup policy (backup to VTL then use SLP to
move the data to physical tape) and it worked great. I was backing up 2
servers from remote environments (about 80). The VTL would suck all the data
in (I had about 20 tape drives) through the course of the night.
I think we're going to need some more information about your environment to
help:
1. What operating system and version is your master/media server?
2. How are the tape drives attached (SCSI, fibre channel, etc.)?
The 800 code is a fairly generic message stating the NBU cannot talk to the
dri
I don't think this problem is a NBU issue. It looks like Solaris has lost the
drives. You need to rescan the devices to make sure that Solaris can see the
drives. I believe you can run "devfsadm -c tape" to have the system rescan for
devices or issue a "touch /reconfigure" then reboot the sy
NetBackup does not differentiate between images on disk versus images on tape.
They are all just images to it. If the data you request resides on tape, NBU
will generate a tape mount request automatically.
As for the date thing, NBU will, by default, go back to the most recent full
backup dis
What version of NBU are you running and what is the operating system of the
media server?
+--
|This was sent by t...@tjsimerson.org via Backup Central.
|Forward SPAM to ab...@backupcentral.com.
+--
I've not actually done one but that is supposed to be a hairy process full of
pit falls. This is usually done as a professional services type job.
It sounds like you have a master/media server combination. How about simply
creating a new Window media server to add to the NBU domain? That way
This is actually not unusual. Because of fragmentation and multiplexing, an
image might "shrink" just a little bit after duplication. You should be okay
but Darren is right for testing. Do some restores from that copy to just make
sure it is okay.
You can use the bpchangeprimary command to c
Since your media server will be moving from one master server to another, you
want to do a clean install. Migrate your mediadb like you were planning,
remove the old 5.1 software and then install the 6.5 software. I went through
this exact same process when upgrading my environment from 5.1 to
Kind of depends on your DBA's [Wink] At a previous position, all of the DBA's
(Oracle, SQL and Informix) hated the idea of anything communicating with their
database that they had no control over (that and they were cheap).
We worked with the SQL DBA to make use of user directed backups. He ba
As smpt noted, the files are encrypted. To veiw the actual details, use the
get_license_key command (stored under /usr/openv/netbackup/bin/admincmd) to
view the actual keys and what features the keys activate.
+--
|This was sen
pwhelan0610 is correct: you will get the equivalent of a full backup (because
there is no baseline) but the retention will be based upon the incremental
schedule. And you will continue to get a full data backup with each
incremental until you the full backup schedule successfully runs. I ran
Spot on, Justin, with the logging. Best diagnostic tool. I would guess that
the firewall rules on the master server are blocking the connection (just
because it is Windows 2008 R2).
There is no longer any kind of driver that Symantec supplies. They chose (IMHO
the right choice) to leverage t
You can go to the Symantec licensing portal (licensing.symantec.com) to
register your licenses and process an upgrade on them to convert them from a
6.X license to a 7.X license. This will activate all the features in NBU 7.X.
+---
- Should I look at Symantecs solution? Is there something better out there
(VEEM?)
The VMware backup mechanism in NetBackup is very good. The niche products like
Veeam and vRanger are also good products but you have one major disadvantage:
they support only VMware backups. And, most impleme
NBU 7.1.0.4 is probably the most stable of the 7.1 patches. I have had several
customers that were/are on earlier releases which required several EEB fixes to
be applied. If you are doing any kind of deduplication (MSDP or OST
appliances) or using SLPs, then you will want to be on 7.1.0.4.
+-
You might be stuck. Clearing the archive bit requires accessing the data at
the file system level. NDMP is a lower level protocol that is accessing at a
block level.
Out of curiosity, why do you need to clear the archive bit? My experience is
the archive bit is not very reliable for determini
The key thing is to get the Windows system to see the drives first. Symantec
stopped shipping device drivers for Windows several versions ago and is relying
upon the underlying operating system to properly configure the drives. Quantum
should have on their web site some Windows drivers for the
The reason for the professional services is the nature of the NetBackup catalog
space. The master server hostname is embedded throughout the EMM and images
databases. Symantec or authorized VARs have the proper tools to make these
changes to the NetBackup catalog. As Dan indicated, it may loo
I'm curious as to why your DNS admin thought removing the reverse lookup zone
was a good idea. The only reasons I've ever gotten at customer sites or within
my own infrastructure has been "I thought nothing used them" or "It is a
security risk" (which no one has provided valid reasons for this
There are several factors here that will help determine what technologies you
should look at:
1. Do you need to do file level restores of this VM that has these huge number
of files?
2. About how many and what kind of files (.avi, .mkv, .jpg, etc.) comprise the
"huge number of files"?
3. Wha
The three error codes you list all indicate something on the client side
causing the issue. I will suggest that you download and look through the
Symantec NetBackup Status Code Reference Guide
(http://www.symantec.com/business/support/index?page=content&id=DOC5181).
There are some very good n
You don't need the full size of your images database for free disk space. Only
the metadata of the images database is imported into the EMM database. This
usually accounts for about 1% - 3% of the total images database size. In your
case, the 20 GB of free space should be fine. This is only
Take all look on your master and media servers to see if you have the entry
"AUTO_PATH_CORRECTION=YES" in the /usr/openv/volmgr/vm.conf file. This will
tell NetBackup to rescan the system devices and update the NetBackup device
configuration with the new/changed node information. The rescan is
Make sure that you are adding the license keys to the master server also. That
is where the issues are actually occurring. The media server attempts to run
the configuration tools but the tools are actually making changes on the master
server.
You don't have to add every license key on the ma
You can also use a "bpimagelist -L" (that's a capital L) which will be an more
human readable format. If this is a big backup image with multiple copies, it
can have lots of fragments that can be hard to work through.
+--
|This
According to the documentation, that should have waited for 12 hours. To wait
an infinite time, use the option "-w 0". You never know how long a job will
sit in a queued state due to storage contention.
+--
|This was sent by t
NetBackup does not allocate new media based upon policies themselves but upon
volume pools and retention levels. If you had different volume pools being
used by each policy, then NBU behaved properly. If you have different
retention levels in play in each schedule in each policy, then that wil
Prasanna is correct that you have to patch to 6.5.6 first. But, you can
upgrade directly to 7.5 from 6.5.6. I just checked the latest installation
release notes on this.
Be prepared for lots of changes to the NetBackup environment. The GUIs are
pretty much the same but there are all sorts of
Actually, there are two significant database changes. The EMM database is
being upgraded from Sybase 11 to Sybase 12. This is one of the causes for
lengthy upgrades. This process (I believe) is to actually dump the database,
upgrade Sybase and then reload the database. This is where if you h
Finally able to get to Symantec web sites. The issue with the TIR pruning not
working correctly is isolated to 7.5.0.6. There is a specific EEB that you can
get to patch either NBU 7.5.0.6 or appliance version 2.5.3. The other solution
is to patch up to NBU 7.5.0.7 or appliance version 2.5.4.
With NBU 7.1, the Java console and native Windows console are pretty much the
same. Prior to that, there was a huge difference in functionality. The only
real difference between the two is that when you are accessing client
properties through the native Windows console, you can actually browse
One thing to keep in mind with the major difference between a NetBackup
appliance and a dedupe storage appliance (Data Domain, DXi, Exagrid, etc.): a
NetBackup appliance is a media server that can do dedupe storage but the other
appliances require a media server in front of it.
When all is sai
45 matches
Mail list logo