Marc, In my opinion, a user is not capable of being the cause of database corruption, no matter how ill-behaved he or she is, short of unplugging a cable in the middle of an update.
A bad user might be able to cause a data integrity problem, but not broken pointers or indexes. Bill On Thu, Jul 15, 2010 at 6:36 PM, MDRD <[email protected]> wrote: > Sorry, true I should have said Off Topic > I think it is a click happy user on a slow workstation > > Marc > > > *From:* Paul InterlockInfo <[email protected]> > *Sent:* Thursday, July 15, 2010 5:16 PM > *To:* RBASE-L Mailing List <[email protected]> > *Subject:* [RBASE-L] - RE: Off Topic Data Corruption > > > Network so change the title/Subject to Off-Topic > > > > > > You have a network error and need to look hard and what time. Have them > b/u hourly if needed and compare till you find it. I would write more but > we have a HUGE storm here right now and I am double b/u as we speak sorry! > > > > Sincerely, > > Paul D > > www.SystemNets.com > > > > > > *From:* [email protected] [mailto:[email protected]] *On Behalf Of *MDRD > *Sent:* Thursday, July 15, 2010 6:05 PM > *To:* RBASE-L Mailing List > *Subject:* [RBASE-L] - RE: Data Corruption > > > > Jim and Paul > > > > They do an Unload ALL at least month, they did one a week ago when they had > corruption in 1 table, > > The deleted the last 20 rows then did an Unload All which seemed to fix the > problem for a few says > > but now have the corruption again. > > > > Marc > > > > > > *From:* Paul InterlockInfo <[email protected]> > > *Sent:* Thursday, July 15, 2010 4:33 PM > > *To:* RBASE-L Mailing List <[email protected]> > > *Subject:* [RBASE-L] - RE: Data Corruption > > > > The answer to your question “yes” that is why Rscope or do a Razzak > routine Unload / load with error messages on etc… It is all at his ftp > site. It will show you the errors. > > > > > > > > Sincerely, > > Paul D > > www.SystemNets.com > > > > > > > > > > You also might try using a slightly different data pack (larger/more info > in every packet) protocol “NetBios” it has more information in the data > pack and a different arrangement for the data/bit check authorization (front > end check w/rear verification). Back in the old days Novell (hint on the > name “NetBois”) was very stable but when Windows and the Internet came along > “TCP/IP” won out. Control Panel> Network Connections > “This connection > uses the following items:” > Install > Client > NetBois DONE! > > > > Now you can run both at the same time and Windows seems to handle it well. > The IT person will be familiar with the protocol. There are also ways to > scan and watch for dropped packets and do a before and after comparison. > Run it as I recall with Wins in the DNS area. > > > > Perhaps some others will correct any errors I have made. But it is what I > recall. > > > > referances: http://support.microsoft.com/kb/314067 > > > > > > > > > > > > > > *From:* [email protected] [mailto:[email protected]] *On Behalf Of *MDRD > *Sent:* Thursday, July 15, 2010 4:37 PM > *To:* RBASE-L Mailing List > *Subject:* [RBASE-L] - RE: Data Corruption > > > > Can Autochck pass but there still be some corrupt data inside the Varchar > column? > > > > Thanks > > Marc > > > > > > *From:* Jim Belisle <[email protected]> > > *Sent:* Thursday, July 15, 2010 3:25 PM > > *To:* RBASE-L Mailing List <[email protected]> > > *Subject:* [RBASE-L] - RE: Data Corruption > > > > Marc, > > > > We just had a long discussion on the list about the bugaboo “Access > Violation”. > > From what the list told me, it is a very hard thing to track down. > > That could be anything from hardware to wires to communication between > server and the PC. > > > > I know here, that problem was only on two stations, mine and one other > station. > > Mine was limited to a specific transaction so I was not as worried about > that. > > The one PC however, I believe was more to do with the PC itself. > > That PC has had all sorts of problems, from the email locking up RBASE to > viruses. > > > > If the data is being corrupted, that may cause the violation as well form > what I’ve been told. > > > > Jim > > > ------------------------------ > > *From:* [email protected] [mailto:[email protected]] *On Behalf Of *MDRD > *Sent:* Thursday, July 15, 2010 3:18 PM > *To:* RBASE-L Mailing List > *Subject:* [RBASE-L] - Data Corruption > > > > Hi > > > > Well that office with the slow Delete problem had data corruption for the > second time in a about 10 days. > > I wonder if they are clicking like crazy or pressing ESC because of the > delay? I am sending them a new > > Main menu that runs the Delete code when they close the day, which is run > on the server and that should > > solve that issue for now at least. > > > > My guess is a network issue. They have a good network tech and he is not > blaming me or RBase but he > > was talking about the network sending packets of info and things over my > head. > > > > I told him that if there is a hiccup or brown power during a Save Row or > something that might corrupt the data. > > He said he thought that if a packet was not received that the packet gets > resent and that should not > > cause a problem as far as he knew. > > > > I am not sure if this is the difference between C/S and file server? or is > there something else I can tell him to look for? > > He is going to change the cables and do a few other hardware things in the > morning. No AV software is running. > > > > Also they got an Access Violation the other day if that means anything. > > > > Thanks > > Marc > > > > > > >
