I don't know if anyone can confirm or debunk this, but I was talkingwith a
couple VAR dudes a couple weeks ago, and one of the techs wastelling me that
the communications have changed with 6.0 as well,specificially with respect to
firewalls
He said they had recently done a couple installs where
-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL
PROTECTED] On Behalf Of WEAVER, Simon Sent: December 12, 2006 1:11 AM To:
'[EMAIL PROTECTED]'; veritas-bu@mailman.eng.auburn.edu; [EMAIL PROTECTED]
Subject: Re: [Veritas-bu] Checkpoint on backups. If using VSP - must
On Tue, 12 Dec 2006, bob944 wrote:
I don't have a dog in this fight--I have as little to do with Microsoft
products as possible.
I have also found our backups
can run 10 mins slower with VSP on, compared to turning it off.
Your bpbkar log will show you how much of that is due to the
Possible network issues or congestion? Run a ping during the entire
period of the backup and watch for lost packets also check netstat -i and
check for network errors?
On Tue, 12 Dec 2006, Whelan, Patrick wrote:
Environment:
Solaris master server NBU5.1MP5
HP-UX Media Server NBU5.1MP5
The 2.4 kernel itself can handle well over 2.0GB, it depends on what
version of glibc you have loaded. Solaris 8 was 1TB, Solaris 9 depends on
whether its 32bit/64bit and what FS you are using, QFS/UFS, etc, but
definitely over 2.0GB.
Error 96 means you ran out of tapes, (cannot allocate new
True - but when you have LARGE TB's of Data, its horrible!
The example was for a file system 1.7GB in size !
Regards
Simon Weaver
3rd Line Technical Support
Windows Domain Administrator
EADS Astrium Limited, B23AA IM (DCS)
Anchorage Road, Portsmouth, PO3 5PU
Email: [EMAIL PROTECTED]
Ah, wow.
Also, if NetBackup does not use fopen64() then the log size will be
limited to 2.0GB.
Justin.
On Tue, 12 Dec 2006, Hindle, Greg wrote:
What was happening is the file size limit was being reached on the Linux
server and the log could no longer be written to. The ndmp backup
When I have had this issue consistently on a client I resolved it by hard
coding the NIC's link speed instead of using Auto-negotiate. A single stream
issue though wouldn't seem to be the case with a NIC issue. Try disabling
Tracker on the client, this resolved many of my pesky random
May run faster if the wind goes south ;-)
Regards
Simon Weaver
3rd Line Technical Support
Windows Domain Administrator
EADS Astrium Limited, B23AA IM (DCS)
Anchorage Road, Portsmouth, PO3 5PU
Email: [EMAIL PROTECTED]
-Original Message-
From: Paul Keating [mailto:[EMAIL PROTECTED]
Also try turning off VSP/VSS!
On Tue, 12 Dec 2006, Haskins, Steve wrote:
When I have had this issue consistently on a client I resolved it by hard
coding the NIC's link speed instead of using Auto-negotiate. A single stream
issue though wouldn't seem to be the case with a NIC issue. Try
What was happening is the file size limit was being reached on the Linux
server and the log could no longer be written to. The ndmp backup would
go through all the tapes and then error our with a 96 even though the
non-ndmp would work fine. Had me and the tech support guy stumped until
he came
Justin
Not had a real issue with VSS for Win2k3 - its been VSP which to me, has
always been a slightly revised revision of OTM.
Still, each to their own! And in most cases, our large filesystems get
backed up 100% without this feature turned on :-)
Regards
Simon Weaver
3rd Line Technical
It was true of 32 bit. The limit was actually less than 2 GB. Even
without that if the application itself isn't compiled with largefiles
support it would hit this limit. Since most Linux is still 32 bit the
NBU for it may not have this support.
Of course the real issue here is that
Yes. It's the NDMP log file. And its more of a Linux thing then
netbackup as it only happens on Linux or at least they (Symantec) are
only seeing that on Linux. The tech not is not a public one that I can
share but it does exist.
Greg
-Original Message-
From: Peter DrakeUnderkoffler
I'm implementing 6.0 here and I'm wondering how critical it is that we
upgrade *every last client* to 4.5+ !? We only purchased a limited
number of 6.0 clients, most of our machines are 4.5 or 5.1. But what
about the 3.4s? I know they aren't supported, but will it work? I
don't mind spending
A!! It was you Chris! Wasn't it yourself that sent out an email a few
weeks back with problems?
Regards
Simon Weaver
3rd Line Technical Support
Windows Domain Administrator
EADS Astrium Limited, B23AA IM (DCS)
Anchorage Road, Portsmouth, PO3 5PU
Email: [EMAIL PROTECTED]
-Original
We have had nightmares. And, that might even should be in CAPS,
underlined, and in bold.
Best to upgrade them all.
Most 3.4 clients appear to fail. You can't even use tools like 'bpdir'
on them after the upgrade.
Interestingly, some Mac 3.4 clients do seem to work.
But, at least 90% will
I've been told by Veritas support to update clients to 5.x prior to
updating 6.0 because the rewrite of the system won't support 4.5 and
earlier. My experience after any update is if there are any issues with
a client after an update that updating the client resolves the issue
(then I only have to
I think this might help you out?
http://www.microsoft.com/windowsserversystem/wss2003/techinfo/plandeploy
/storportwp.mspx
Cheers,
Jon
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of mkiles
Sent: Monday, December 11, 2006 12:20 PM
To:
I have also found our backups
can run 10 mins slower with VSP on, compared to turning it off.
Your bpbkar log will show you how much of that is due to the
initial wait for n seconds of disk-quiet time.
True - but when you have LARGE TB's of Data, its horrible!
The example was
I should have thought of it sooner, but it happens so infrequently. There is a
problem with one of the mount points within the stream that failed. A find
/ hangs. Since this is basically what NetBackup does, the backup hangs and
eventually times out. If I had logged in and done the find
Hi Simon,
It was indeed me. :-)
I have been having just loads of fun with this upgrade, since here we
have only limited access to client machines.
A 'drag' might even summarize the whole 6.0 experience.
EMM did not upgrade nicely, so all of our NDMP backups have been hosed
up, plus Vault
Enviroment Information: Netbackup Master Server - W2K3SP1/NBU 6.0
MP3/LTO 3
We are currently deploying Netbackup into our datacenter enviroment (we
are currently using Storage Data Protector). I have one backup that
requires a script to be run before the backup kicks off, and one that
needs to
Thanks Bob :-)
Yep - been down the book, and various technotes too :-)
Regards
Simon Weaver
3rd Line Technical Support
Windows Domain Administrator
EADS Astrium Limited, B23AA IM (DCS)
Anchorage Road, Portsmouth, PO3 5PU
Email: [EMAIL PROTECTED]
-Original Message-
From: bob944
If it is not NFS mounted, what causes it to hang?
Justin.
On Tue, 12 Dec 2006, Whelan, Patrick wrote:
I should have thought of it sooner, but it happens so infrequently. There is
a problem with one of the mount points within the stream that failed. A find
/ hangs. Since this is
Search Informations int he administrator's guide about bpstart_notify
and bpend_notify scripts.
T+
Carlos Alberto Lima dos Santos
Eng. Computação Jundiaí SP-BR
* mailto:[EMAIL PROTECTED] [EMAIL PROTECTED]
H https://www.linkedin.com/in/carlostoca
skype carlostoca
msn only [EMAIL
I believe in Windows, it is located in C:\Program
Files\VERITAS\NetBackup\bin\goodies.
There are bpstart_notify.bat and bpend_notify.bat scripts. You can modify
them to run other scripts, at the start and end of the backup. But BE
CAUTIONED! These start script runs prior to each stream, and
Thanks for the caution about the streaming issue - we are currently only
backing up one directory, so multiple streams should not be an issue.
However, this presents a small issue for our UNIX backups.
In our Data Protector configuration, we have all our unix clients in one
backup policy, with
On Tue, Dec 12, 2006 at 09:49:52AM -0500, Jeff Lightner wrote:
It was true of 32 bit. The limit was actually less than 2 GB. Even
without that if the application itself isn't compiled with largefiles
support it would hit this limit. Since most Linux is still 32 bit the
NBU for it may not
CJ,
On the client there should be in the
install_path\veritas\netbackup\bin\goodies directory two scripts:
bpstart_notify.cmd bpend_notify.cmd
bpstart_notify.cmd is called by the bpbkar32.exe process before the
backup starts and is given by default 300 seconds (5 minutes) to
complete.
What's the force streams option?
Steve
_
From: Jones, Courtenay [mailto:[EMAIL PROTECTED]
Sent: Tuesday, December 12, 2006 3:44 PM
To: Steve Fogarty; veritas-bu@mailman.eng.auburn.edu
Subject: RE: [Veritas-bu] Netbackup Scripting Question (NBU 6.0 MP3)
Thanks for the caution about
On 12/12/2006 9:50 AM, Hindle, Greg wrote:
Yes. It's the NDMP log file. And its more of a Linux thing then
netbackup as it only happens on Linux or at least they (Symantec) are
only seeing that on Linux.
It's not a Linux thing - it's an application thing. It's just that
Symantec's code is
* We had an occurrence where the backup of a highly-used
production server was running at 7:00 AM and beyond.
* Performance degradation occurred and users complained.
* The backup job started in the backup window, but is there a way
to suspend or cancel a backup job if it
33 matches
Mail list logo
