Veritas-bu

[Veritas-bu] Strangness on user backups - filelist sent is not filelist backed -up for bpbackup

2003-03-03 18:10:26
Subject: [Veritas-bu] Strangness on user backups - filelist sent is not filelist backed -up for bpbackup
From: Mark.Donaldson AT experianems DOT com (Donaldson, Mark)
Date: Mon, 3 Mar 2003 16:10:26 -0700
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C2E1DA.1368F180
Content-Type: text/plain;
        charset="iso-8859-1"

I have a script that allows users to make their own backups.  It recorded
this today:

## Filelist Begin...
/stage2/vue/temp/xref.tpid
## ...Filelist End

Backup started Mon Mar 03 16:19:37 2003

16:19:38 Initiating backup
16:21:06 INF - Processing /stage2/vue/temp/xref.rid
16:21:06 /
16:21:06 /stage2/
16:21:06 /stage2/vue/
16:21:06 /stage2/vue/temp/
16:21:10 INF - Beginning backup on server citidev of client saturn.
16:44:12 /stage2/vue/temp/xref.rid
16:44:12 INF - Client completed sending data for backup

16:44:23 INF - Backup by root on client saturn using policy User, sched u36:
th
e requested operation was successfully completed.


The filelist used for the "-f" inputfile on the bpbackup command is echoed
between the dual hashes above.  Before I back it up, I just cat it out into
this logfile so I can see what the users are doing.  

The problem, as you may notice, is that the filelist contained this file
called "xref.tpid" and the backup grabbed "xref.rid".  The latter is a valid
file but not the same. 

I see two possibilties, and there may be more...  Once the "-f" file is read
by the bpbackup command, it must persist until the command completes.  The
file data may not be cached at read time.  This is one backup of a large set
& perhaps the filename, while based on PID, may have duplicated.

The other idea is some sort of inode confusion - if one file replaced the
other, reusing the inode number, before the bpbackup command completed.

Yet another idea has to do with gremlins or a misbalance of bodily humours
but I'll leave that for the apothacary to diagnose.

Ideas?

-M


------_=_NextPart_001_01C2E1DA.1368F180
Content-Type: text/html;
        charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2653.12">
<TITLE>Strangness on user backups - filelist sent is not filelist =
backed-up for bpbackup</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>I have a script that allows users to make their own =
backups.&nbsp; It recorded this today:</FONT>
</P>

<P><FONT SIZE=3D2>## Filelist Begin...</FONT>
<BR><FONT SIZE=3D2>/stage2/vue/temp/xref.tpid</FONT>
<BR><FONT SIZE=3D2>## ...Filelist End</FONT>
</P>

<P><FONT SIZE=3D2>Backup started Mon Mar 03 16:19:37 2003</FONT>
</P>

<P><FONT SIZE=3D2>16:19:38 Initiating backup</FONT>
<BR><FONT SIZE=3D2>16:21:06 INF - Processing =
/stage2/vue/temp/xref.rid</FONT>
<BR><FONT SIZE=3D2>16:21:06 /</FONT>
<BR><FONT SIZE=3D2>16:21:06 /stage2/</FONT>
<BR><FONT SIZE=3D2>16:21:06 /stage2/vue/</FONT>
<BR><FONT SIZE=3D2>16:21:06 /stage2/vue/temp/</FONT>
<BR><FONT SIZE=3D2>16:21:10 INF - Beginning backup on server citidev of =
client saturn.</FONT>
<BR><FONT SIZE=3D2>16:44:12 /stage2/vue/temp/xref.rid</FONT>
<BR><FONT SIZE=3D2>16:44:12 INF - Client completed sending data for =
backup</FONT>
</P>

<P><FONT SIZE=3D2>16:44:23 INF - Backup by root on client saturn using =
policy User, sched u36: th</FONT>
<BR><FONT SIZE=3D2>e requested operation was successfully =
completed.</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>The filelist used for the &quot;-f&quot; inputfile on =
the bpbackup command is echoed between the dual hashes above.&nbsp; =
Before I back it up, I just cat it out into this logfile so I can see =
what the users are doing.&nbsp; </FONT></P>

<P><FONT SIZE=3D2>The problem, as you may notice, is that the filelist =
contained this file called &quot;xref.tpid&quot; and the backup grabbed =
&quot;xref.rid&quot;.&nbsp; The latter is a valid file but not the =
same. </FONT></P>

<P><FONT SIZE=3D2>I see two possibilties, and there may be =
more...&nbsp; Once the &quot;-f&quot; file is read by the bpbackup =
command, it must persist until the command completes.&nbsp; The file =
data may not be cached at read time.&nbsp; This is one backup of a =
large set &amp; perhaps the filename, while based on PID, may have =
duplicated.</FONT></P>

<P><FONT SIZE=3D2>The other idea is some sort of inode confusion - if =
one file replaced the other, reusing the inode number, before the =
bpbackup command completed.</FONT></P>

<P><FONT SIZE=3D2>Yet another idea has to do with gremlins or a =
misbalance of bodily humours but I'll leave that for the apothacary to =
diagnose.</FONT></P>

<P><FONT SIZE=3D2>Ideas?</FONT>
</P>

<P><FONT SIZE=3D2>-M</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C2E1DA.1368F180--

<Prev in Thread] Current Thread [Next in Thread>
  • [Veritas-bu] Strangness on user backups - filelist sent is not filelist backed -up for bpbackup, Donaldson, Mark <=