Hi All
I was wondering what policy design some of you have followed for large
Oracle environments
I am busy doing a redesign of an Netbackup environment which backs up
about 70 DB's spread across about 30 servers (some of which are cross
sites)
I am doing my best not to end up with a solution
Until you have an [x]inetd process running the system wont be able to
listen for bpcd connections
You can start bpcd in standalone mode (bpcd -standalone) on the client
for the time being
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
rizalman_almasi
PROTECTED]
Sent: 25 July 2007 13:30
To: Shekel, Tal; veritas-bu@mailman.eng.auburn.edu
Subject: Re: [Veritas-bu] Cross Site Clustering
Hi Tal
As long as the DNS (and it's reverse lookup
configuration) is working or a good host files exists,
you also can backup the physical names, although
Doesn't that depend on whether you use the -i option?
If you want to pass the policy a file list you run bpbackup without the
-i
-Original Message-
From: Greenberg, Katherine (ISD, IT)
[mailto:[EMAIL PROTECTED]
Sent: 25 July 2007 20:35
To: Shekel, Tal; Justin Piszcz
Cc: veritas-bu
Looks like it could be a reverse lookup issue
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Dave
Markham
Sent: 26 July 2007 17:25
To: veritas-bu@mailman.eng.auburn.edu
Subject: Re: [Veritas-bu] Strange Status 25 on connection
Thanks ill give it a go.
Yes -
It will still work
The jobs will run as user jobs which are prioritised
Regards
Tal
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Justin
Piszcz
Sent: 25 July 2007 19:10
To: Greenberg, Katherine (ISD, IT)
Cc: veritas-bu@mailman.eng.auburn.edu
Looks like you have your answer
You should also try keep any db agents at the same MP as the master
I have seen issues from this with the Oracle agent
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of
malabelle
Sent: 25 July 2007 18:31
To:
I suppose its personal preference.
I prefer using agents because:
1. It avoids a two stage recovery if your backup is not on local disk.
2. It puts more control and understanding in the hands of the backup
administrator (and more work unfortunately)
3. It avoids scheduling issues - so you
I haven't followed the whole thread so excuse me if my message isn't
relevant but I read in a performance tuning guide you should set your
NET_BUFFER_SZ to 4 times the size of your SIZE_DATA_BUFFERS
Also - not mandatory but recommended to match the client and media
server network buffers
The
I have also been looking at this as an option.
One of the considerations is the effect encryption has on compression.
I have heard you loose the ability to compress effectively when using
compression.
Have you noticed an effect on your tape drive compression when using
tape drive based
requirement.
Regards,
Tal
-Original Message-
From: judy_hinchcli...@administaff.com
[mailto:judy_hinchcli...@administaff.com]
Sent: 02 March 2010 14:42
To: Shekel Tal; jpis...@lucidpixels.com; kc.on.the@gmail.com
Cc: veritas-bu@mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] Sun
Tal
Cc: Kevin C; veritas-bu@mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] Sun/StorageTek based LTO4 encryption
On Tue, 2 Mar 2010, Shekel Tal wrote:
I have also been looking at this as an option.
One of the considerations is the effect encryption has on compression.
I have heard you
Sorry, I just noticed now the one data set is encrypted and the other is
not.
Quite a noticeable difference
-Original Message-
From: Justin Piszcz [mailto:jpis...@lucidpixels.com]
Sent: 02 March 2010 14:49
To: Shekel Tal
Cc: Kevin C; veritas-bu@mailman.eng.auburn.edu
Subject: RE
MPX 03/20/2010 11:40N/A FULL
-Original Message-
From: veritas-bu-boun...@mailman.eng.auburn.edu
[mailto:veritas-bu-boun...@mailman.eng.auburn.edu] On Behalf Of Len
Boyle
Sent: 02 March 2010 13:53
To: Shekel Tal; Justin Piszcz; Kevin C
Cc: veritas-bu
Thanks for the info Justin
Definitely would like to test in our environment and compare . . . very
interesting
-Original Message-
From: Justin Piszcz [mailto:jpis...@lucidpixels.com]
Sent: 02 March 2010 14:56
To: Shekel Tal
Cc: Kevin C; veritas-bu@mailman.eng.auburn.edu
Subject: RE
throughput requirements
Regards,
Tal
-Original Message-
From: judy_hinchcli...@administaff.com
[mailto:judy_hinchcli...@administaff.com]
Sent: 02 March 2010 15:20
To: jpis...@lucidpixels.com; Shekel Tal
Cc: kc.on.the@gmail.com; veritas-bu@mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] Sun
Have you checked if you are running out of memory or swap space?
-Original Message-
From: veritas-bu-boun...@mailman.eng.auburn.edu
[mailto:veritas-bu-boun...@mailman.eng.auburn.edu] On Behalf Of WEAVER,
Simon (external)
Sent: 29 March 2010 15:41
To: VERITAS-BU@mailman.eng.auburn.edu
When you sent your original email you said your jobs were frequently
being killed with error 50 so it sounded like they would fail with error
50 while the server was running. Is it just one or two or do a whole
collection of jobs seem to die together during the backup window? Or do
you just come
Also - is this a multi-streamed job?
There may be more info in one of the other jobs associated with it
From: veritas-bu-boun...@mailman.eng.auburn.edu
[mailto:veritas-bu-boun...@mailman.eng.auburn.edu] On Behalf Of Jonathan Dyck
Sent: 06 April 2010 14:50
Have you checked for any entries in the admin log after firing off the
command?
-Original Message-
From: veritas-bu-boun...@mailman.eng.auburn.edu
[mailto:veritas-bu-boun...@mailman.eng.auburn.edu] On Behalf Of
pwhelan0610
Sent: 08 April 2010 17:05
To: VERITAS-BU@mailman.eng.auburn.edu
Hi Heather
If you have Cisco switches you can use EtherChannel (PAgP) which is very
similar to LACP for port aggregation - you would want to load balance by
source MAC. This should benefit your media servers depending on what
backup devices they are writing to.
You will also the need some kind
What hardware did you change?
Error 13/4 can be a real pain but they are often related to network
issues or reading data off a file system.
As this is happening to all your jobs it sounds network related
Have you checked some of your system/NetBackup legacy logs?
Also run a all log entries report
I think to make an informed decision you would need to look at the
bigger picture. Your original goal was to increase your backup
performance and shrink you backup window
Disk can be great but don't expect your backup times to increase just
because you are using it.
Is your disk being provided
Have you tuned you NetBackup buffers?
Have a look at the following technote and setup the size and number data
buffers - http://support.veritas.com/docs/183702
http://support.veritas.com/docs/183702
Are you using agents and mounting the image up after the split or just
performing a cold flat
the copy?
Have you checked the server hosting the disks system resource
utilisation?
From: hemant.kale...@wipro.com [mailto:hemant.kale...@wipro.com]
Sent: 10 May 2010 11:49
To: Shekel Tal; veritas-bu@mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] Slow backup
[mailto:hemant.kale...@wipro.com]
Sent: 10 May 2010 13:24
To: Shekel Tal; veritas-bu@mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] Slow backup speed
hi
pls find the output
# more NUMBER_DATA_BUFFERS
32
# more SIZE_DATA_BUFFERS
262144
The server hosting the disk system resource utilisation
[mailto:hemant.kale...@wipro.com]
Sent: 11 May 2010 06:10
To: Shekel Tal; veritas-bu@mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] Slow backup speed
Had done the same but still the issue persists
Thanks and Regards
Hemant S Kalekar
09819949422
From
This may go without saying but I think an important point is to
standardise as much as possible.
This will make future troubleshooting and stability easier to control.
Very often your library vendor will dictate the level of firmware for
tape devices but more importantly they will recommend
Hi Guys
I have quite an interesting one here
NetBackup 6.5.5 - windows 2003 Master
NetBackup 6.5.5 Windows 2003 Media Server behind a firewall backing
itself up
All comms going through vnetd
I am using a file list such as:
E:\Folder1\*
E:\Folder2\*
I am doing this so that I can
Thanks, will try that
I can see there is a stream file for each policy which ran with the file
structure inside - makes sense
From: Marianne Van Den Berg [mailto:mvdb...@stortech.co.za]
Sent: 26 May 2010 15:44
To: Shekel Tal; veritas-bu
Depends on how much data you are backing up and what your retention
periods and backup window are?
I am a big fan of storagetek libraries and with LTO5 out now you can get
3TB on a single cart.
For you regions you may want to look at puredisk clients or perhaps a
NetApp filer in Manchester with
Hi Kevin
If you use INSTANCE $ALL and DATABASE $ALL NetBackup will automatically
try and backup every DB and Instance on the server.
You can either logon to the server as has been described and use the SQL
client GUI to check the client progress log.
An easier method is to just remotely
Did you by any chance zone the devices into multiple HBA ports in the
new config but not the old?
Perhaps you had powerpath, DMP or some kind of multipath software
installed before?
By the way - there shouldn't be any problems with seeing multiple
devices and NB will only use the ones configured
The 64k limit was due to a tcp/ip stack limitation
I have always found larger buffers (256k) generally provide the best
performance.
Only if you had lots of small files would I recommend anything less
Flashbackup will definitely improve your performance as the data will
not need to be
2010 14:08
To: Shekel Tal
Cc: NetBackup Mailing List
Subject: Re: [Veritas-bu] Device Recognition in RHEL5/NBU7
As far as the zoning in the old environment, I've gone over both with
a fine tooth comb and it looks like I've zoned both environments
exactly alike.
As far as multipath software, I have
Doesn't sound like a netbackup issue
Sounds like a host configuration issue - maybe dns
Are you returning 127.0.0.1 even when you have any entry in your local hosts?
The system could be configured not to look at hosts
Can you resolve forward and reverse lookup when using nslookup?
Regards,
Tal
(Contractor) [mailto:jinfant...@intersil.com]
Sent: 16 March 2012 11:53
To: Shekel Tal; veritas-bu@mailman.eng.auburn.edu
Subject: RE: cold catalog recovery
That's what we thought also, but why then would it work just fine before the
catalog recovery?
I do have a hosts file entry for the system
: Infantino, Joe (Contractor) [mailto:jinfant...@intersil.com]
Sent: 16 March 2012 11:56
To: Shekel Tal; veritas-bu@mailman.eng.auburn.edu
Subject: RE: cold catalog recovery
It is a standard system build. I did the build and did not change anything to
do with DNS
Firewall is also turned off. I
Subject line so it is more specific
than Re: Contents of Veritas-bu digest...
Today's Topics:
1. cold catalog recovery (Infantino, Joe (Contractor))
2. Re: cold catalog recovery (Shekel Tal)
3. Re: cold catalog recovery (Infantino, Joe (Contractor))
4. Re: cold catalog recovery (Infantino, Joe
39 matches
Mail list logo