Veritas-bu

[Veritas-bu] error 96 but have scratch

2003-05-13 00:02:28
Subject: [Veritas-bu] error 96 but have scratch
From: Tim.McMurphy AT telus DOT com (Tim McMurphy)
Date: Mon, 12 May 2003 22:02:28 -0600
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_01C31904.7857CF10
Content-Type: text/plain; 
 charset=iso-8859-1
Content-Transfer-Encoding: 7bit

I had the same problem and had 2 cases opened with Veritas on it. If you
specify "any available" storage unit change it to force the class to use one
not "any available" (shades of the evil "auto negotiate"). That was the
workaround I came up with while waiting for Veritas. Here is what they said
to do. By the way it showed up as status 96 in the job monitor but was
really a status 174 in the debug logs.
 
  

---   From Veritas (case # 181-316-761)


Try this. Create a file on the problem media server:
echo 0 >/usr/openv/volmgr/VMDB_QUERY_BUFFER_SIZE

See if that resolves it.

 

---   From me (case # 181-316-761)


What is this one supposed to do that would fix it?

 

---   From Veritas (case # 181-316-761)

 

An enhancement was made to vmd that caused it to buffer some of the
workload. This did not always work. 

Here are the errors in your bptm log that show this :

10:07:44 [1170] <16> string_to_record: cannot convert unrecognized string:
me Manager
10:07:45 [1170] <16> count_media: unable to query Media Manager volume
daemon on host odin, unexpected data received
10:07:45 [1170] <2> bptm: EXITING with status 174 <----------

 

The fix is already coded into the next patch release.



 

 -----Original Message-----
From: Dave Markham [mailto:dave.markham AT fjserv DOT net]
Sent: Friday, May 09, 2003 4:33 AM
To: veritas-bu AT mailman.eng.auburn DOT edu
Subject: [Veritas-bu] error 96 but have scratch



I am seeing a weird activity
Im running netbackup 3.2 on solaris 8 with an L1000 unit attached
 
I have various classes and schedules set up which all seem to work ok but im
am seeing more and more error 96 no media to assign which is fine if there
is none, but when I look I see plenty of scratch in the jukebox which is
odd.
It seems that tapes from scratch are not being taken and changed to the pool
required but at times they might be.
 
I have no frozen tapes and no down drives although bperror -U -media shows
reports :-
 
1052446020 1 132 8 tpf-pbs1 19280 0 0 tpf-pas3 bptm media id DBQ856 is in a
DOWN drive, misplaced, or unmountable; attempting retry with different media
id
1052446164 1 132 8 tpf-pbs1 19280 0 0 tpf-pas3 bptm media id DBQ838 is in a
DOWN drive, misplaced, or unmountable; attempting retry with different media
id
1052446312 1 132 8 tpf-pbs1 19280 0 0 tpf-pas3 bptm media id DBQ883 is in a
DOWN drive, misplaced, or unmountable; attempting retry with different media
id

Anyone seen anything like this? or have any idea why scratch may not be used
?
 
I have an auto-sratch script which runs and places all deassigned media into
scratch pool each day too so there should be scratch.
 
Dave
 

------_=_NextPart_001_01C31904.7857CF10
Content-Type: text/html; 
 charset=iso-8859-1
Content-Transfer-Encoding: 7bit

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 6.00.2800.1170" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><SPAN class=714100404-13052003><FONT face=Arial color=#0000ff size=2>I had 
the same problem and had 2 cases opened with Veritas on it. If you specify "any 
available" storage unit change it to force the class to use one not "any 
available" (shades of the evil "auto negotiate"). That was the workaround I 
came 
up with while waiting for Veritas. Here is what they said to do. By the way it 
showed up as status 96 in the job monitor but was really a status 174&nbsp;in 
the debug logs.</FONT></SPAN></DIV>
<DIV><SPAN class=714100404-13052003><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=714100404-13052003>&nbsp;
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%"><SPAN 
class=714100404-13052003>---&nbsp;&nbsp; From Veritas (</SPAN>case # 
181-316-761<SPAN class=714100404-13052003>)</SPAN><BR></P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">Try this. Create a 
file on the problem media server:<BR>echo 0 
&gt;/usr/openv/volmgr/VMDB_QUERY_BUFFER_SIZE</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">See if that 
resolves it.</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">&nbsp;</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%"><SPAN 
class=714100404-13052003>---&nbsp;&nbsp; From me (</SPAN>case # 
181-316-761<SPAN 
class=714100404-13052003>)</SPAN><BR></P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">What is this 
one<SPAN class=714100404-13052003> </SPAN>supposed to do that would fix it?</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">&nbsp;</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%"><SPAN 
class=714100404-13052003>---&nbsp;&nbsp; From Veritas (</SPAN>case # 
181-316-761<SPAN class=714100404-13052003>)</SPAN></P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%"><SPAN 
class=714100404-13052003></SPAN>&nbsp;</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">An enhancement was 
made to vmd that caused it to buffer some of the workload. This did not always 
work. </P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">Here are the 
errors 
in your bptm log that show this :</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">10:07:44 [1170] 
&lt;16&gt; string_to_record: cannot convert unrecognized string: me 
Manager<BR>10:07:45 [1170] &lt;16&gt; count_media: unable to query Media 
Manager 
volume daemon on host odin, unexpected data received<BR>10:07:45 [1170] 
&lt;2&gt; bptm: EXITING with status 174 &lt;----------</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%"><FONT face=Arial 
color=#0000ff size=2></FONT>&nbsp;</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%">The fix is already 
coded into the next patch release.<BR><BR></SPAN><FONT face=Tahoma><FONT 
size=2><SPAN class=714100404-13052003></SPAN></FONT></FONT></P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%"><FONT 
face=Tahoma><FONT size=2><SPAN 
class=714100404-13052003></SPAN></FONT></FONT>&nbsp;</P>
<P style="MARGIN: 0px; WORD-SPACING: 0px; LINE-HEIGHT: 100%"><FONT 
face=Tahoma><FONT size=2><SPAN 
class=714100404-13052003>&nbsp;</SPAN>-----Original 
Message-----<BR><B>From:</B> 
Dave Markham [mailto:dave.markham AT fjserv DOT net]<BR><B>Sent:</B> Friday, 
May 09, 
2003 4:33 AM<BR><B>To:</B> veritas-bu AT mailman.eng.auburn DOT 
edu<BR><B>Subject:</B> 
[Veritas-bu] error 96 but have scratch<BR><BR></P></DIV></FONT></FONT>
<DIV><FONT face=Arial size=2>I am seeing a weird activity</FONT></DIV>
<DIV><FONT face=Arial size=2>Im running netbackup 3.2 on solaris 8 with an 
L1000 
unit attached</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>I have various classes and schedules set up which 
all seem to work ok but im am seeing more and more error 96 no media to assign 
which is fine if there is none, but when I look I see plenty of scratch in the 
jukebox which is odd.</FONT></DIV>
<DIV><FONT face=Arial size=2>It seems that tapes from scratch are not being 
taken and changed to the pool required but at times they might be.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>I have no frozen tapes and no down drives although 
bperror -U -media shows reports :-</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>1052446020 1 132 8 tpf-pbs1 19280 0 0 tpf-pas3 
bptm 
media id DBQ856 is in a DOWN drive, misplaced, or unmountable; attempting retry 
with different media id<BR>1052446164 1 132 8 tpf-pbs1 19280 0 0 tpf-pas3 bptm 
media id DBQ838 is in a DOWN drive, misplaced, or unmountable; attempting retry 
with different media id<BR>1052446312 1 132 8 tpf-pbs1 19280 0 0 tpf-pas3 bptm 
media id DBQ883 is in a DOWN drive, misplaced, or unmountable; attempting retry 
with different media id<BR></FONT></DIV>
<DIV><FONT face=Arial size=2>Anyone seen anything like this? or have any idea 
why scratch may not be used ?</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>I have an auto-sratch script which runs and places 
all deassigned media into scratch pool each day too so there should be 
scratch.</FONT></DIV>
<DIV><FONT face=Arial size=2></FONT>&nbsp;</DIV>
<DIV><FONT face=Arial size=2>Dave</FONT></DIV>
<DIV><FONT face=Arial size=2>&nbsp;</DIV></FONT></BODY></HTML>

------_=_NextPart_001_01C31904.7857CF10--


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