Veritas-bu

[Veritas-bu] NT OTM Questions

2003-02-20 12:06:37
Subject: [Veritas-bu] NT OTM Questions
From: scott.kendall AT abbott DOT com (scott.kendall AT abbott DOT com)
Date: Thu, 20 Feb 2003 11:06:37 -0600
This is a multipart message in MIME format.
--=_alternative 005DF94B86256CD3_=
Content-Type: text/plain; charset="us-ascii"

You can get rid of the 11 errors without disabling OTM if you tune it 
properly.  Then you'll still be using OTM and hopefully have a better 
chance of skipping less files.  When the OTM cache fills, it will do one 
of three things.  One of them is error out, which causes the 11 errors. 
Change the OTM Error Control to "Retry OTM" or "Disable OTM".  Retry works 
fine.

The cache temp files could cause system problems if they are filling the 
entire partition that the system files are on.  One thing to do is move 
the cache file to another location, however, this would not be my 
recommendation.

The cache temp files are often left uncleaned (it is supposed to clean 
them up and this can be seen in bpbkar logs at the end of the job) if the 
backup fails (system hangs, network issues, reboot while backup still 
running, etc.).  We have also seen problems leaving cache files if using 
too large of a max cache size (or 0, which equals 10% of disk) space.  OTM 
uses memory space that has limits.  If you exceed this, OTM will error out 
behind the scenes and the files can be left uncleaned.  See caveats 
section of this for more info.

http://seer.support.veritas.com/docs/233254.htm


For those of you that don't like OTM, do the following and I think you'll 
agree that, while not perfect, OTM can be removed from your 
"thorn-in-my-side" list.
- change Error Control to something other than "abort on error"
- decrease your busy file wait a few seconds
- increase your busy file timeout to a few minutes
- change your initial cache size to something like 100 MB
- change your max cache size to something other than 0, higher than the 
initial size, and lower than the allowed limit for your OS per technote 
article above


- Scott





"Steven L. Sesar" <ssesar AT mitre DOT org>
Sent by: veritas-bu-admin AT mailman.eng.auburn DOT edu
02/20/2003 10:17 AM

 
        To:     "Nash, Ebon" <Ebon_Nash AT compuware DOT com>
        cc:     "'Weber, Philip'" <Philip.Weber AT egg DOT com>, 
veritas-bu AT mailman.eng.auburn DOT edu
        Subject:        Re: [Veritas-bu] NT OTM Questions


I don't recall seeing this option in 3.4, but on a 4.5 master, you can 
disable OTM at the client level. It's just a checkbox, and it rid me of 
those pesky 11 system call failure errors.

Here's couple of links to some good info on NBU and OTM:

http://www.backupcentral.com/foms/netbackup-serve/cache/151.html

--Steve


Nash, Ebon wrote:
> I would recommend either increasing the size of the system drive, which 
is
> most likely your C: drive.  If that is not possible, I believe there is 
a
> way to tell NetBackup to write OTM files to a different location 
(anything
> other than c:).  I believe their web site has information on how to do 
this.
> One other thing you can try is disabling OTM to see if that fixes the
> problem.  That should be documented on Veritas' web site as well.
> 
> 
> Ebon Nash
> Compuware Corp.
> 
> -----Original Message-----
> From: Weber, Philip [mailto:Philip.Weber AT egg DOT com]
> Sent: Thursday, February 20, 2003 6:17 AM
> To: veritas-bu AT mailman.eng.auburn DOT edu
> Subject: [Veritas-bu] NT OTM Questions
> 
> 
> Dear all,
> 
> We are experiencing system problems with a large NT4 fileserver (500 Gb
> storage, 2 * P3 Xeon 550, 512 Gb RAM), the system has been found hanging 
the
> last three mornings & had to be cold restarted.  We are running 
Netbackup
> 3.4 patch level 3 on the Solaris servers & patch level 1 on the NT 
clients.
> 
> Fingers are being pointed at OTM (because yesterday when the problem
> occurred the winnt system drive was found to have filled up, & an old 
OTM
> cache file was found on it).
> 
> I have seen on this list that other people have had issues with OTM.  I 
know
> it is a pain to configure and doesn't seem to work very well, can it 
cause
> system problems?
> 
> I have also seen that people have turned OTM off.  What are the effects 
of
> this - I would assume that all open files would therefore not be backed 
up
> and we would get lots of 1 return codes.
> 
> When OTM is working correctly I does it create cache files & delete 
after
> use?  Can it create multiple?  Why would it leave them lying around?
> 
> Any help appreciated ... thanks, Phil
> 
> Phil Weber
> Egg Distributed Hosts - UNIX Systems Engineer
> Phone: 01384 26 4136
> Mobile: 
> 
> 
> -----Original Message-----
> From: Liddle, Stuart [mailto:liddles AT amgen DOT com]
> Sent: 19 February 2003 22:48
> To: veritas-bu AT mailman.eng.auburn DOT edu
> Subject: [Veritas-bu] Catalog backup question (NetBackup Datacenter 3.4
> UNIX)
> 
> 
> I need a way to find out how much data is actually being written to our
> catalog tape.  I used the following:
> 
> bpbackupdb -v
> 
> and looked at the /usr/openv/netbackup/logs/admin/log.<date> file
> 
> It did not have the total number of bytes written to the tape anywhere 
in
> the output.  I'd like to know how close I'm getting to filling my 
catalog
> tape.
> 
> Is there an easy way to determine this information, or do I have to just
> guess that since the data in the 
> 
> /usr/openv/netbackup/db 
> 
> directory is mostly text that it will get at least a 2:1 compression
> (probably closer to 4:1) and use that as a guide for how much will fit 
on my
> DLT7000 tapes?  Since the hardware compression will get about 70GB of
> compressed data on a tape, what's the realistic amount to expect that 
will
> fit on the tape?  Would I have to worry if my db directory is about 
50GB?
> 100GB?
> 
> thanks,
> 
> --stuart
> _________________________
> Stuart W. Liddle
> Amgen Corp.
> liddles AT amgen DOT com
> 
> 
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
> 
> 
> This private and confidential e-mail has been sent to you by Egg.
> The Egg group of companies includes Egg Banking plc
> (registered no. 2999842), Egg Financial Products Ltd (registered
> no. 3319027) and Egg Investments Ltd (registered no. 3403963) which
> carries out investment business on behalf of Egg and is regulated
> by the Financial Services Authority. 
> Registered in England and Wales. Registered offices: 1 Waterhouse 
Square,
> 138-142 Holborn, London EC1N 2NA.
> If you are not the intended recipient of this e-mail and have
> received it in error, please notify the sender by replying with
> 'received in error' as the subject and then delete it from your
> mailbox.
> 
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
> 
> 
> 
> The contents of this e-mail are intended for the named addressee only. 
It
> contains information that may be confidential. Unless you are the named
> addressee or an authorized designee, you may not copy or use it, or 
disclose
> it to anyone else. If you received it in error please notify us 
immediately
> and then destroy it. 
> 
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
> 


-- 
===================================

Steven L. Sesar
Ops. Sys. Programmer/Analyst, Sr.
Application Operations R10A
The MITRE Corporation
202 Burlington Road - R101
Bedford, MA 01730
tel: (781) 271-7702
fax: (781) 271-2600
email: ssesar AT mitre DOT org
mobile: (617) 893-9635

===================================


_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu



--=_alternative 005DF94B86256CD3_=
Content-Type: text/html; charset="us-ascii"


<br><font size=2 face="Arial">You can get rid of the 11 errors without 
disabling OTM if you tune it properly. &nbsp;Then you'll still be using OTM and 
hopefully have a better chance of skipping less files. &nbsp;When the OTM cache 
fills, it will do one of three things. &nbsp;One of them is error out, which 
causes the 11 errors. &nbsp;Change the OTM Error Control to &quot;Retry 
OTM&quot; or &quot;Disable OTM&quot;. &nbsp;Retry works fine.</font>
<br>
<br><font size=2 face="Arial">The cache temp files could cause system problems 
if they are filling the entire partition that the system files are on. 
&nbsp;One thing to do is move the cache file to another location, however, this 
would not be my recommendation.</font>
<br>
<br><font size=2 face="Arial">The cache temp files are often left uncleaned (it 
is supposed to clean them up and this can be seen in bpbkar logs at the end of 
the job) if the backup fails (system hangs, network issues, reboot while backup 
still running, etc.). &nbsp;We have also seen problems leaving cache files if 
using too large of a max cache size (or 0, which equals 10% of disk) space. 
&nbsp;OTM uses memory space that has limits. &nbsp;If you exceed this, OTM will 
error out behind the scenes and the files can be left uncleaned. &nbsp;See 
caveats section of this for more info.</font>
<br>
<br><font size=2 
face="Arial">http://seer.support.veritas.com/docs/233254.htm</font>
<br>
<br>
<br><font size=2 face="Arial">For those of you that don't like OTM, do the 
following and I think you'll agree that, while not perfect, OTM can be removed 
from your &quot;thorn-in-my-side&quot; list.</font>
<br><font size=2 face="Arial">- change Error Control to something other than 
&quot;abort on error&quot;</font>
<br><font size=2 face="Arial">- decrease your busy file wait a few 
seconds</font>
<br><font size=2 face="Arial">- increase your busy file timeout to a few 
minutes</font>
<br><font size=2 face="Arial">- change your initial cache size to something 
like 100 MB</font>
<br><font size=2 face="Arial">- change your max cache size to something other 
than 0, higher than the initial size, and lower than the allowed limit for your 
OS per technote article above</font>
<br>
<br>
<br><font size=2 face="Arial">- Scott</font>
<br>
<br>
<br>
<br>
<table width=100%>
<tr valign=top>
<td>
<td><font size=1 face="sans-serif"><b>&quot;Steven L. Sesar&quot; &lt;ssesar AT 
mitre DOT org&gt;</b></font>
<br><font size=1 face="sans-serif">Sent by: veritas-bu-admin AT 
mailman.eng.auburn DOT edu</font>
<p><font size=1 face="sans-serif">02/20/2003 10:17 AM</font>
<br>
<td><font size=1 face="Arial">&nbsp; &nbsp; &nbsp; &nbsp; </font>
<br><font size=1 face="sans-serif">&nbsp; &nbsp; &nbsp; &nbsp; To: &nbsp; 
&nbsp; &nbsp; &nbsp;&quot;Nash, Ebon&quot; &lt;Ebon_Nash AT compuware DOT 
com&gt;</font>
<br><font size=1 face="sans-serif">&nbsp; &nbsp; &nbsp; &nbsp; cc: &nbsp; 
&nbsp; &nbsp; &nbsp;&quot;'Weber, Philip'&quot; &lt;Philip.Weber AT egg DOT 
com&gt;, veritas-bu AT mailman.eng.auburn DOT edu</font>
<br><font size=1 face="sans-serif">&nbsp; &nbsp; &nbsp; &nbsp; Subject: &nbsp; 
&nbsp; &nbsp; &nbsp;Re: [Veritas-bu] NT OTM Questions</font></table>
<br>
<br>
<br><font size=2 face="Courier New">I don't recall seeing this option in 3.4, 
but on a 4.5 master, you can <br>
disable OTM at the client level. It's just a checkbox, and it rid me of <br>
those pesky 11 system call failure errors.<br>
<br>
Here's couple of links to some good info on NBU and OTM:<br>
<br>
http://www.backupcentral.com/foms/netbackup-serve/cache/151.html<br>
<br>
--Steve<br>
<br>
<br>
Nash, Ebon wrote:<br>
&gt; I would recommend either increasing the size of the system drive, which 
is<br>
&gt; most likely your C: drive. &nbsp;If that is not possible, I believe there 
is a<br>
&gt; way to tell NetBackup to write OTM files to a different location 
(anything<br>
&gt; other than c:). &nbsp;I believe their web site has information on how to 
do this.<br>
&gt; One other thing you can try is disabling OTM to see if that fixes the<br>
&gt; problem. &nbsp;That should be documented on Veritas' web site as well.<br>
&gt; <br>
&gt; <br>
&gt; Ebon Nash<br>
&gt; Compuware Corp.<br>
&gt; <br>
&gt; -----Original Message-----<br>
&gt; From: Weber, Philip [mailto:Philip.Weber AT egg DOT com]<br>
&gt; Sent: Thursday, February 20, 2003 6:17 AM<br>
&gt; To: veritas-bu AT mailman.eng.auburn DOT edu<br>
&gt; Subject: [Veritas-bu] NT OTM Questions<br>
&gt; <br>
&gt; <br>
&gt; Dear all,<br>
&gt; <br>
&gt; We are experiencing system problems with a large NT4 fileserver (500 Gb<br>
&gt; storage, 2 * P3 Xeon 550, 512 Gb RAM), the system has been found hanging 
the<br>
&gt; last three mornings &amp; had to be cold restarted. &nbsp;We are running 
Netbackup<br>
&gt; 3.4 patch level 3 on the Solaris servers &amp; patch level 1 on the NT 
clients.<br>
&gt; <br>
&gt; Fingers are being pointed at OTM (because yesterday when the problem<br>
&gt; occurred the winnt system drive was found to have filled up, &amp; an old 
OTM<br>
&gt; cache file was found on it).<br>
&gt; <br>
&gt; I have seen on this list that other people have had issues with OTM. 
&nbsp;I know<br>
&gt; it is a pain to configure and doesn't seem to work very well, can it 
cause<br>
&gt; system problems?<br>
&gt; <br>
&gt; I have also seen that people have turned OTM off. &nbsp;What are the 
effects of<br>
&gt; this - I would assume that all open files would therefore not be backed 
up<br>
&gt; and we would get lots of 1 return codes.<br>
&gt; <br>
&gt; When OTM is working correctly I does it create cache files &amp; delete 
after<br>
&gt; use? &nbsp;Can it create multiple? &nbsp;Why would it leave them lying 
around?<br>
&gt; <br>
&gt; Any help appreciated ... thanks, Phil<br>
&gt; <br>
&gt; Phil Weber<br>
&gt; Egg Distributed Hosts - UNIX Systems Engineer<br>
&gt; Phone: 01384 26 4136<br>
&gt; Mobile: <br>
&gt; <br>
&gt; <br>
&gt; -----Original Message-----<br>
&gt; From: Liddle, Stuart [mailto:liddles AT amgen DOT com]<br>
&gt; Sent: 19 February 2003 22:48<br>
&gt; To: veritas-bu AT mailman.eng.auburn DOT edu<br>
&gt; Subject: [Veritas-bu] Catalog backup question (NetBackup Datacenter 3.4<br>
&gt; UNIX)<br>
&gt; <br>
&gt; <br>
&gt; I need a way to find out how much data is actually being written to our<br>
&gt; catalog tape. &nbsp;I used the following:<br>
&gt; <br>
&gt; bpbackupdb -v<br>
&gt; <br>
&gt; and looked at the /usr/openv/netbackup/logs/admin/log.&lt;date&gt; file<br>
&gt; <br>
&gt; It did not have the total number of bytes written to the tape anywhere 
in<br>
&gt; the output. &nbsp;I'd like to know how close I'm getting to filling my 
catalog<br>
&gt; tape.<br>
&gt; <br>
&gt; Is there an easy way to determine this information, or do I have to 
just<br>
&gt; guess that since the data in the <br>
&gt; <br>
&gt; /usr/openv/netbackup/db <br>
&gt; <br>
&gt; directory is mostly text that it will get at least a 2:1 compression<br>
&gt; (probably closer to 4:1) and use that as a guide for how much will fit on 
my<br>
&gt; DLT7000 tapes? &nbsp;Since the hardware compression will get about 70GB 
of</font>
<br><font size=2 face="Courier New">&gt; compressed data on a tape, what's the 
realistic amount to expect that will<br>
&gt; fit on the tape? &nbsp;Would I have to worry if my db directory is about 
50GB?<br>
&gt; 100GB?<br>
&gt; <br>
&gt; thanks,<br>
&gt; <br>
&gt; --stuart<br>
&gt; _________________________<br>
&gt; Stuart W. Liddle<br>
&gt; Amgen Corp.<br>
&gt; liddles AT amgen DOT com<br>
&gt; <br>
&gt; <br>
&gt; _______________________________________________<br>
&gt; Veritas-bu maillist &nbsp;- &nbsp;Veritas-bu AT mailman.eng.auburn DOT 
edu<br>
&gt; http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu<br>
&gt; <br>
&gt; <br>
&gt; This private and confidential e-mail has been sent to you by Egg.<br>
&gt; The Egg group of companies includes Egg Banking plc<br>
&gt; (registered no. 2999842), Egg Financial Products Ltd (registered<br>
&gt; no. 3319027) and Egg Investments Ltd (registered no. 3403963) which<br>
&gt; carries out investment business on behalf of Egg and is regulated<br>
&gt; by the Financial Services Authority. &nbsp;<br>
&gt; Registered in England and Wales. Registered offices: 1 Waterhouse 
Square,<br>
&gt; 138-142 Holborn, London EC1N 2NA.<br>
&gt; If you are not the intended recipient of this e-mail and have<br>
&gt; received it in error, please notify the sender by replying with<br>
&gt; 'received in error' as the subject and then delete it from your<br>
&gt; mailbox.<br>
&gt; <br>
&gt; _______________________________________________<br>
&gt; Veritas-bu maillist &nbsp;- &nbsp;Veritas-bu AT mailman.eng.auburn DOT 
edu<br>
&gt; http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu<br>
&gt; <br>
&gt; <br>
&gt; <br>
&gt; The contents of this e-mail are intended for the named addressee only. 
It<br>
&gt; contains information that may be confidential. Unless you are the named<br>
&gt; addressee or an authorized designee, you may not copy or use it, or 
disclose<br>
&gt; it to anyone else. If you received it in error please notify us 
immediately<br>
&gt; and then destroy it. <br>
&gt; <br>
&gt; _______________________________________________<br>
&gt; Veritas-bu maillist &nbsp;- &nbsp;Veritas-bu AT mailman.eng.auburn DOT 
edu<br>
&gt; http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu<br>
&gt; <br>
<br>
<br>
-- <br>
===================================<br>
<br>
Steven L. Sesar<br>
Ops. Sys. Programmer/Analyst, Sr.<br>
Application Operations R10A<br>
The MITRE Corporation<br>
202 Burlington Road - R101<br>
Bedford, MA 01730<br>
tel: (781) 271-7702<br>
fax: (781) 271-2600<br>
email: ssesar AT mitre DOT org<br>
mobile: (617) 893-9635<br>
<br>
===================================<br>
<br>
<br>
_______________________________________________<br>
Veritas-bu maillist &nbsp;- &nbsp;Veritas-bu AT mailman.eng.auburn DOT edu<br>
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu<br>
</font>
<br>
<br>
--=_alternative 005DF94B86256CD3_=--

<Prev in Thread] Current Thread [Next in Thread>