Sybase NNTP forums - End Of Life (EOL)

The NNTP forums from Sybase - forums.sybase.com - are now closed.

All new questions should be directed to the appropriate forum at the SAP Community Network (SCN).

Individual products have links to the respective forums on SCN, or you can go to SCN and search for your product in the search box (upper right corner) to find your specific developer center.

Drive crash, recovering DB from dump w/out some info

5 posts in Windows NT Last posting was on 2000-07-03 13:00:43.0Z
Frank Burleigh Posted on 2000-06-26 21:18:45.0Z
Message-ID: <3957C8B5.D44C2EA0@indiana.edu>
Date: Mon, 26 Jun 2000 16:18:45 -0500
From: Frank Burleigh <burleigh@indiana.edu>
X-Mailer: Mozilla 4.73 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
Subject: Drive crash, recovering DB from dump w/out some info
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Newsgroups: sybase.public.sqlserver.nt
Lines: 39
NNTP-Posting-Host: holdout.law.indiana.edu 129.79.131.77
Path: forums-1-dub!forums-master.sybase.com!forums.sybase.com
Xref: forums-1-dub sybase.public.sqlserver.nt:1969
Article PK: 1089416

Hello. We had a hard drive crash over the weekend on our ase 11.9.2
install. I had recent dumps of the substantive db and of master. I
don't have printed output from master showing the layout of the
substantive db but it was pretty simple, and my memory of it is
"close." This db was on two fragments over two devices:

dev01 held the db and was 600mb
dev02 held the log and was 200mb (I had at one point moved the log here)

On a fresh drive with ase 12 I created the two devices (800 and 200
respective, I assume a device bigger than the db fragment is ok). Then
I used

create database name
on dev01=600
log on dev02=200
for load

Then I used sybcentral to load the db. The load seemed to go fine and
dbcc claims the db is ok. Of course user names are in bad shape and
I've seen at least one case where the db user name was not the same as
the server's idea of login names. Since there are only 20 or so of
these I'm happy to delete and recreate those.

My questions:

1. Is it safe to take dbcc's "word" that the db is ok?

2. What could one do with the dump of the old master to check the
assumptions I've made above with respect to the substantive db, and what
steps would I use to get access to the system tables in the master dump?

I am particularly concerned about allowing this db back into production,
just to have to dump it and reload into the new system once replacement
hardware arrives...
--
Frank Burleigh
Indiana University School of Law
burleigh@indiana.edu


Frank Burleigh Posted on 2000-07-02 00:44:10.0Z
Message-ID: <395E905A.28EE7E4A@indiana.edu>
Date: Sat, 01 Jul 2000 19:44:10 -0500
From: Frank Burleigh <burleigh@indiana.edu>
X-Mailer: Mozilla 4.73 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
Subject: Re: Drive crash, recovering DB from dump w/out some info
References: <3957C8B5.D44C2EA0@indiana.edu>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Newsgroups: sybase.public.sqlserver.nt
Lines: 47
NNTP-Posting-Host: holdout.law.indiana.edu 129.79.131.77
Path: forums-1-dub!forums-master.sybase.com!forums.sybase.com
Xref: forums-1-dub sybase.public.sqlserver.nt:1950
Article PK: 1089366

I have installed 11.9.2 on a machine, created a 1000mb 'device' and
loaded the dump there. No errors. I wonder why loading to 12 has
troubles?

Frank Burleigh wrote:
>
> Hello. We had a hard drive crash over the weekend on our ase 11.9.2
> install. I had recent dumps of the substantive db and of master. I
> don't have printed output from master showing the layout of the
> substantive db but it was pretty simple, and my memory of it is
> "close." This db was on two fragments over two devices:
>
> dev01 held the db and was 600mb
> dev02 held the log and was 200mb (I had at one point moved the log here)
>
> On a fresh drive with ase 12 I created the two devices (800 and 200
> respective, I assume a device bigger than the db fragment is ok). Then
> I used
>
> create database name
> on dev01=600
> log on dev02=200
> for load
>
> Then I used sybcentral to load the db. The load seemed to go fine and
> dbcc claims the db is ok. Of course user names are in bad shape and
> I've seen at least one case where the db user name was not the same as
> the server's idea of login names. Since there are only 20 or so of
> these I'm happy to delete and recreate those.
>
> My questions:
>
> 1. Is it safe to take dbcc's "word" that the db is ok?
>
> 2. What could one do with the dump of the old master to check the
> assumptions I've made above with respect to the substantive db, and what
> steps would I use to get access to the system tables in the master dump?
>
> I am particularly concerned about allowing this db back into production,
> just to have to dump it and reload into the new system once replacement
> hardware arrives...
> --
> Frank Burleigh
> Indiana University School of Law
> burleigh@indiana.edu

--
Frank Burleigh
Indiana University School of Law
burleigh@indiana.edu


Jason Froebe Posted on 2000-06-28 16:06:22.0Z
Message-ID: <395A227E.F3898F50@sybase.com>
Date: Wed, 28 Jun 2000 11:06:22 -0500
From: Jason Froebe <jfroebe@sybase.com>
Organization: Sybase, Inc.
X-Mailer: Mozilla 4.73 [en] (X11; U; Linux 2.2.15 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: Frank Burleigh <burleigh@indiana.edu>
Subject: Re: Drive crash, recovering DB from dump w/out some info
References: <3957C8B5.D44C2EA0@indiana.edu>
Content-Type: multipart/mixed; boundary="------------C8071B4E3D79DECF1D8047AC"
Newsgroups: sybase.public.sqlserver.nt
Lines: 74
NNTP-Posting-Host: jfroebe-desktop.sybase.com 157.133.46.52
Path: forums-1-dub!forums-master.sybase.com!forums.sybase.com
Xref: forums-1-dub sybase.public.sqlserver.nt:1962
Article PK: 1089439


Frank Burleigh wrote:

> Hello. We had a hard drive crash over the weekend on our ase 11.9.2
> install. I had recent dumps of the substantive db and of master. I
> don't have printed output from master showing the layout of the
> substantive db but it was pretty simple, and my memory of it is
> "close." This db was on two fragments over two devices:
>
> dev01 held the db and was 600mb
> dev02 held the log and was 200mb (I had at one point moved the log here)
>
> On a fresh drive with ase 12 I created the two devices (800 and 200
> respective, I assume a device bigger than the db fragment is ok). Then
> I used
>
> create database name
> on dev01=600
> log on dev02=200
> for load
>
> Then I used sybcentral to load the db. The load seemed to go fine and
> dbcc claims the db is ok. Of course user names are in bad shape and
> I've seen at least one case where the db user name was not the same as
> the server's idea of login names. Since there are only 20 or so of
> these I'm happy to delete and recreate those.
>
> My questions:
>
> 1. Is it safe to take dbcc's "word" that the db is ok?
>
> 2. What could one do with the dump of the old master to check the
> assumptions I've made above with respect to the substantive db, and what
> steps would I use to get access to the system tables in the master dump?
>
> I am particularly concerned about allowing this db back into production,
> just to have to dump it and reload into the new system once replacement
> hardware arrives...
> --
> Frank Burleigh
> Indiana University School of Law
> burleigh@indiana.edu

Hi Frank,

If "dbcc checkdb", "dbcc checkalloc", and "dbcc checkstorage" are clean
then it is unlikely there is any problem with the structure of the database
or the tables. You may want to run a few queries on your user tables and
verify the data looks okay to be on the safe side.

In order to look at the old master, you will need to create a new server
and load it in. It won't load into a user database.

j


Download VCard jfroebe.vcf


Frank Burleigh Posted on 2000-07-01 01:36:55.0Z
Message-ID: <395D4B37.5567A2B3@indiana.edu>
Date: Fri, 30 Jun 2000 20:36:55 -0500
From: Frank Burleigh <burleigh@indiana.edu>
X-Mailer: Mozilla 4.73 [en] (Windows NT 5.0; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Jason Froebe <jfroebe@sybase.com>
Subject: Re: Drive crash, recovering DB from dump w/out some info
References: <3957C8B5.D44C2EA0@indiana.edu> <395A227E.F3898F50@sybase.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Newsgroups: sybase.public.sqlserver.nt
Lines: 90
NNTP-Posting-Host: holdout.law.indiana.edu 129.79.131.77
Path: forums-1-dub!forums-master.sybase.com!forums.sybase.com
Xref: forums-1-dub sybase.public.sqlserver.nt:1952
Article PK: 1089391

Thanks for your suggestions. For a time I was lulled into believing the
db was in good shape, but then we began to experience problems in a web
application. For a few days I thought this might be an IIS or OLE DB
driver problem but now that I look at dbcc run a different way I see
this

Data size mismatch occurred while checking a TEXT value.

The message goes on to say (in the places where this happens) that the
expected size is x while the found size is y. I don't see a
pattern--tables that haven't been touched have this problem as do tables
that have been touched. I've also good to recheck the dbccs I did after
the load, but I'm pretty sure they were clean. Of course I relied on
sybase central to perform those... ;-)

I've talked to sybase tech support, which so far doesn't have anything
specific to offer. They feel that as long as I created the device and
db to be large enough to accomodate the dump, the db should have loaded
fine.

They speculate that the db was always in a bad way, but that wouldn't
explain why the web app only began to act badly *after* the drive died.
They also question the backup. To test that I'm going to reload 11.9.2
on a server and try the load again, and also from an earlier tape
backup.

I continue to appreciate any advice. Thanks.

Jason Froebe wrote:
>
> Frank Burleigh wrote:
>
> > Hello. We had a hard drive crash over the weekend on our ase 11.9.2
> > install. I had recent dumps of the substantive db and of master. I
> > don't have printed output from master showing the layout of the
> > substantive db but it was pretty simple, and my memory of it is
> > "close." This db was on two fragments over two devices:
> >
> > dev01 held the db and was 600mb
> > dev02 held the log and was 200mb (I had at one point moved the log here)
> >
> > On a fresh drive with ase 12 I created the two devices (800 and 200
> > respective, I assume a device bigger than the db fragment is ok). Then
> > I used
> >
> > create database name
> > on dev01=600
> > log on dev02=200
> > for load
> >
> > Then I used sybcentral to load the db. The load seemed to go fine and
> > dbcc claims the db is ok. Of course user names are in bad shape and
> > I've seen at least one case where the db user name was not the same as
> > the server's idea of login names. Since there are only 20 or so of
> > these I'm happy to delete and recreate those.
> >
> > My questions:
> >
> > 1. Is it safe to take dbcc's "word" that the db is ok?
> >
> > 2. What could one do with the dump of the old master to check the
> > assumptions I've made above with respect to the substantive db, and what
> > steps would I use to get access to the system tables in the master dump?
> >
> > I am particularly concerned about allowing this db back into production,
> > just to have to dump it and reload into the new system once replacement
> > hardware arrives...
> > --
> > Frank Burleigh
> > Indiana University School of Law
> > burleigh@indiana.edu
>
> Hi Frank,
>
> If "dbcc checkdb", "dbcc checkalloc", and "dbcc checkstorage" are clean
> then it is unlikely there is any problem with the structure of the database
> or the tables. You may want to run a few queries on your user tables and
> verify the data looks okay to be on the safe side.
>
> In order to look at the old master, you will need to create a new server
> and load it in. It won't load into a user database.
>
> j

--
Frank Burleigh
Indiana University School of Law
burleigh@indiana.edu


Jason Froebe Posted on 2000-07-03 13:00:43.0Z
From: "Jason Froebe" <jfroebe@sybase.com>
References: <3957C8B5.D44C2EA0@indiana.edu> <395A227E.F3898F50@sybase.com> <395D4B37.5567A2B3@indiana.edu>
Subject: Re: Drive crash, recovering DB from dump w/out some info
Date: Mon, 3 Jul 2000 08:00:43 -0500
Lines: 106
Organization: Sybase, Inc.
X-Newsreader: Microsoft Outlook Express 5.00.2919.6700
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700
Message-ID: <rYV24AP5$GA.298@forums.sybase.com>
Newsgroups: sybase.public.sqlserver.nt
NNTP-Posting-Host: 158.159.8.25
Path: forums-1-dub!forums-master.sybase.com!forums.sybase.com
Xref: forums-1-dub sybase.public.sqlserver.nt:1949
Article PK: 1089364

Hi Frank,

Can you email the case number to me? (not to the newsgroup)
j

"Frank Burleigh" <burleigh@indiana.edu> wrote in message
news:395D4B37.5567A2B3@indiana.edu...
> Thanks for your suggestions. For a time I was lulled into believing the
> db was in good shape, but then we began to experience problems in a web
> application. For a few days I thought this might be an IIS or OLE DB
> driver problem but now that I look at dbcc run a different way I see
> this
>
> Data size mismatch occurred while checking a TEXT value.
>
> The message goes on to say (in the places where this happens) that the
> expected size is x while the found size is y. I don't see a
> pattern--tables that haven't been touched have this problem as do tables
> that have been touched. I've also good to recheck the dbccs I did after
> the load, but I'm pretty sure they were clean. Of course I relied on
> sybase central to perform those... ;-)
>
> I've talked to sybase tech support, which so far doesn't have anything
> specific to offer. They feel that as long as I created the device and
> db to be large enough to accomodate the dump, the db should have loaded
> fine.
>
> They speculate that the db was always in a bad way, but that wouldn't
> explain why the web app only began to act badly *after* the drive died.
> They also question the backup. To test that I'm going to reload 11.9.2
> on a server and try the load again, and also from an earlier tape
> backup.
>
> I continue to appreciate any advice. Thanks.
>
> Jason Froebe wrote:
> >
> > Frank Burleigh wrote:
> >
> > > Hello. We had a hard drive crash over the weekend on our ase 11.9.2
> > > install. I had recent dumps of the substantive db and of master. I
> > > don't have printed output from master showing the layout of the
> > > substantive db but it was pretty simple, and my memory of it is
> > > "close." This db was on two fragments over two devices:
> > >
> > > dev01 held the db and was 600mb
> > > dev02 held the log and was 200mb (I had at one point moved the log
here)
> > >
> > > On a fresh drive with ase 12 I created the two devices (800 and 200
> > > respective, I assume a device bigger than the db fragment is ok).
Then
> > > I used
> > >
> > > create database name
> > > on dev01=600
> > > log on dev02=200
> > > for load
> > >
> > > Then I used sybcentral to load the db. The load seemed to go fine and
> > > dbcc claims the db is ok. Of course user names are in bad shape and
> > > I've seen at least one case where the db user name was not the same as
> > > the server's idea of login names. Since there are only 20 or so of
> > > these I'm happy to delete and recreate those.
> > >
> > > My questions:
> > >
> > > 1. Is it safe to take dbcc's "word" that the db is ok?
> > >
> > > 2. What could one do with the dump of the old master to check the
> > > assumptions I've made above with respect to the substantive db, and
what
> > > steps would I use to get access to the system tables in the master
dump?
> > >
> > > I am particularly concerned about allowing this db back into
production,
> > > just to have to dump it and reload into the new system once
replacement
> > > hardware arrives...
> > > --
> > > Frank Burleigh
> > > Indiana University School of Law
> > > burleigh@indiana.edu
> >
> > Hi Frank,
> >
> > If "dbcc checkdb", "dbcc checkalloc", and "dbcc checkstorage" are clean
> > then it is unlikely there is any problem with the structure of the
database
> > or the tables. You may want to run a few queries on your user tables
and
> > verify the data looks okay to be on the safe side.
> >
> > In order to look at the old master, you will need to create a new server
> > and load it in. It won't load into a user database.
> >
> > j
>
> --
> Frank Burleigh
> Indiana University School of Law
> burleigh@indiana.edu
>