Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Data corruption - any ideas? (Not VO)
44 views
Skip to first unread message
Martin
May 25, 2012, 4:13:31 AM5/25/12
to
Not what you think
Our database schema is in a DBF, and is sent up as part of an update,
after an update a database was missing indexes and kept crashing.
So here is a small section of it from a screen dump.
This was sent up
70 │ 0005 │ DEL_INVSUP │ L │ 1 │ 0 │
70 │ 0006 │ DEL_SHTSUP │ L │ 1 │ 0 │
70 │ 0007 │ DEL_TOTENT │ N │ 8 │ 0 │
70 │ 0008 │ DEL_WEIGHT │ N │ 10 │ 2 │
70 │ 0009 │ DEL_AREA │ N │ 8 │ 2 │
70 │ 0010 │ DEL_USER │ C │ 15 │ 0 │
70 │ 0011 │ DEL_TYPE │ C │ 1 │ 0 │
70 │ 0012 │ DEL_INVBAT │ C │ 10 │ 0 │
70 │ 0013 │ DEL_STOCK │ L │ 1 │ 0 │
70 │ 0014 │ DEL_UPD │ L │ 1 │ 0 │
70 │ I001 │ │ I │ 0 │ 0 │
70 │ I002 │ │ I │ 0 │ 0 │
70 │ I003 │ │ I │ 0 │ 0 │
70 │ I004 │ │ I │ 0 │ 0 │
70 │ I005 │ │ I │ 0 │ 0 │
This is what arrived on the network drive
70 │ 0005 │ DEL_INVSUP │ L │ 1 │ 0
70 │ 0006 │ DEL_SHTSUP │ L │ 1 │ 0
│ │ │ │ 0 │ 0
│ │ │ │ 0 │ 0
70 │ 0009 │ DEL_AREA │ N │ 8 │ 2
│ │ │ │ 0 │ 0
70 │ 0011 │ DEL_TYPE │ C │ 1 │ 0
70 │ 0012 │ DEL_INVBAT │ C │ 10 │ 0
70 │ 0013 │ DEL_STOCK │ L │ 1 │ 0
70 │ 0014 │ DEL_UPD │ L │ 1 │ 0
│ │ │ │ 0 │ 0
│ │ │ │ 0 │ 0
│ │ │ │ 0 │ 0
67 │ 0004 │ PI_ITEM │ C │ 2 │ 0
70 │ I005 │ │ I │ 0 │ 0
67 │ 0005 │ PI_VAL │ N │ 10 │ 2
Does anyone have any ideas why this would happen?
Windows 7 32 bit (which is broken anyway - no NETBIOS) and Windows 2008
server I think. Net Support was used.
What ever it is this is nasty.
Thanks
Martin
Our database schema is in a DBF, and is sent up as part of an update,
after an update a database was missing indexes and kept crashing.
So here is a small section of it from a screen dump.
This was sent up
70 │ 0005 │ DEL_INVSUP │ L │ 1 │ 0 │
70 │ 0006 │ DEL_SHTSUP │ L │ 1 │ 0 │
70 │ 0007 │ DEL_TOTENT │ N │ 8 │ 0 │
70 │ 0008 │ DEL_WEIGHT │ N │ 10 │ 2 │
70 │ 0009 │ DEL_AREA │ N │ 8 │ 2 │
70 │ 0010 │ DEL_USER │ C │ 15 │ 0 │
70 │ 0011 │ DEL_TYPE │ C │ 1 │ 0 │
70 │ 0012 │ DEL_INVBAT │ C │ 10 │ 0 │
70 │ 0013 │ DEL_STOCK │ L │ 1 │ 0 │
70 │ 0014 │ DEL_UPD │ L │ 1 │ 0 │
70 │ I001 │ │ I │ 0 │ 0 │
70 │ I002 │ │ I │ 0 │ 0 │
70 │ I003 │ │ I │ 0 │ 0 │
70 │ I004 │ │ I │ 0 │ 0 │
70 │ I005 │ │ I │ 0 │ 0 │
This is what arrived on the network drive
70 │ 0005 │ DEL_INVSUP │ L │ 1 │ 0
70 │ 0006 │ DEL_SHTSUP │ L │ 1 │ 0
│ │ │ │ 0 │ 0
│ │ │ │ 0 │ 0
70 │ 0009 │ DEL_AREA │ N │ 8 │ 2
│ │ │ │ 0 │ 0
70 │ 0011 │ DEL_TYPE │ C │ 1 │ 0
70 │ 0012 │ DEL_INVBAT │ C │ 10 │ 0
70 │ 0013 │ DEL_STOCK │ L │ 1 │ 0
70 │ 0014 │ DEL_UPD │ L │ 1 │ 0
│ │ │ │ 0 │ 0
│ │ │ │ 0 │ 0
│ │ │ │ 0 │ 0
67 │ 0004 │ PI_ITEM │ C │ 2 │ 0
70 │ I005 │ │ I │ 0 │ 0
67 │ 0005 │ PI_VAL │ N │ 10 │ 2
Does anyone have any ideas why this would happen?
Windows 7 32 bit (which is broken anyway - no NETBIOS) and Windows 2008
server I think. Net Support was used.
What ever it is this is nasty.
Thanks
Martin
dlzc
May 25, 2012, 9:59:31 AM5/25/12
to
Dear Martin:
On Friday, May 25, 2012 1:13:31 AM UTC-7, Martin wrote:
> Not what you think
>
> Our database schema is in a DBF, and is sent up
> as part of an update, after an update a database
> was missing indexes and kept crashing.
Can your describe the update process? This computer does this, the network architecture is this, these files are used...
> So here is a small section of it from a screen
> dump.
...
Is the update file previously existing, or a new file?
> Windows 7 32 bit (which is broken anyway - no
> NETBIOS) and Windows 2008 server I think. Net
> Support was used.
>
> What ever it is this is nasty.
Its Windoze... what else could it be but nasty? It either works, or it is like tracing one piece of spaghetti from end to end, without using your fork...
The network drive, it is mapped how? How does the workstation connect to the drive? How does the workstation connect to the network (is it domain / user name / password)?
David A. Smith
On Friday, May 25, 2012 1:13:31 AM UTC-7, Martin wrote:
> Not what you think
>
> Our database schema is in a DBF, and is sent up
> as part of an update, after an update a database
> was missing indexes and kept crashing.
> So here is a small section of it from a screen
> dump.
> Does anyone have any ideas why this would happen?
Is either computer on a wireless network connection?
Is the update file previously existing, or a new file?
> Windows 7 32 bit (which is broken anyway - no
> NETBIOS) and Windows 2008 server I think. Net
> Support was used.
>
> What ever it is this is nasty.
The network drive, it is mapped how? How does the workstation connect to the drive? How does the workstation connect to the network (is it domain / user name / password)?
David A. Smith
D.J.W. van Kooten
May 25, 2012, 3:28:03 PM5/25/12
to
On Fri, 25 May 2012 09:13:31 +0100, Martin <sp...@spam.spam> wrote:
Hello Martin,
>Our database schema is in a DBF, and is sent up as part of an update,
>after an update a database was missing indexes and kept crashing.
Sent how?
Dick
Hello Martin,
>Our database schema is in a DBF, and is sent up as part of an update,
>after an update a database was missing indexes and kept crashing.
Dick
Martin
May 28, 2012, 6:51:58 AM5/28/12
to
The update went wrong because the control file was corrupted on the way up.
Connections up there - windows networking.
We use net support manager, but are at a loss as to where the data went
Connections up there - windows networking.
We use net support manager, but are at a loss as to where the data went
Martin
May 28, 2012, 6:52:36 AM5/28/12
to
Using Net Support Manager - our usual remote control client
D.J.W. van Kooten
May 29, 2012, 3:53:53 AM5/29/12
to
On Mon, 28 May 2012 11:52:36 +0100, Martin <sp...@spam.spam> wrote:
Hello Martin,
>>
>>> Our database schema is in a DBF, and is sent up as part of an update,
>>> after an update a database was missing indexes and kept crashing.
>>
It is still not clear what you sent. If you sent a dbf+cdx, you may
Hello Martin,
>>
>>> Our database schema is in a DBF, and is sent up as part of an update,
>>> after an update a database was missing indexes and kept crashing.
>>
want to try to zip it, make sure on the receiving part that you can
open the dbf exclusively and if so, unzip the content and reopen the
dbf. That way you control the whole process.
Dick
dlzc
May 29, 2012, 11:23:55 AM5/29/12
to
Dear Martin
On Monday, May 28, 2012 3:51:58 AM UTC-7, Martin wrote:
> > Can your describe the update process? This
> > computer does this, the network architecture
> > is this, these files are used...
...
> We use net support manager, but are at a loss
> as to where the data went
I find almost no discussion of NetSupport on Micro$haft's websites. So if it (NetSupport) is a recent purchase, I suggest you use its error logs to identify the source of corruption, and/or contact the supplier directly.
If you are uploading a file that is in use on either end, then you might get this behavior. Recommend you look at another method of creating this update / transaction file, such as described by Mr. van Kooten.
On Monday, May 28, 2012 3:51:58 AM UTC-7, Martin wrote:
> > Can your describe the update process? This
> > computer does this, the network architecture
> > is this, these files are used...
> > Is either computer on a wireless network
> > connection?
> > connection?
...
> > The network drive, it is mapped how? How does
> > the workstation connect to the drive? How
> > does the workstation connect to the network (is
> > it domain / user name / password)?
...
> > The network drive, it is mapped how? How does
> > the workstation connect to the drive? How
> > does the workstation connect to the network (is
> > it domain / user name / password)?
> The update went wrong because the control file
> was corrupted on the way up.
>
> Connections up there - windows networking.
Thank you, but this does not answer any of the questions asked.
> was corrupted on the way up.
>
> Connections up there - windows networking.
> We use net support manager, but are at a loss
> as to where the data went
If you are uploading a file that is in use on either end, then you might get this behavior. Recommend you look at another method of creating this update / transaction file, such as described by Mr. van Kooten.
0 new messages