Constant Back End Corruption

[CPGDeveloper]

Hi guys --

I've posted separately about this before, but the issue has seemed to have gotten worse so I thought I'd try again.

I created and administer an MS Access Application -- 20 + users -- each user has a FE on their desktop, while the BE resides on a Network Drive. Everyone has Windows 10, 32 bit version of Access 2016. Through VBA, I maintain a persistent connection to the back end.

Today, the back end crashed twice. First thing in the morning, and again around mid-afternoon.

A user will get the 'unrecognized database format' error when opening -- I tell everyone to get out of the db, open the back end, it repairs itself, and everything works again. I sometimes also create a brand new back end -- open a new access db, and import the tables into it. And this is the 3rd time this week. This used to happen once in awhile, so we started replacing the back end once a week. I started thinking we simply need to repair and replace it every night, but after today I'm thinking something else must be wrong.

I realize I have 'Auto Correct' checked in the back end (and front end), and it does not clear the cache on close -- could this be causing any issues?

Our network is notoriously slow (if you open an excel workbook, close it, and then open again, it opens read only thinking you are still in it). Could this be the issue? I do notice that I often cannot create the .accde front end version on a network drive -- it gives me a 'database already in use' error. But when I move it to my hard drive and try to create the .accde, it works just fine. This strikes me as a slow network issue? Could this also be causing back end corruption?

I also noticed earlier this week, a different access application, that does some automation and reporting tasks with Excel Spreadsheets, would not work -- it was constantly getting corrupted -- as i would step through the code, sooner or later the dreaded 'unrecognized database format error' would pop - up. I played with it for some time, constantly recreating it with copies with no luck...finally moved it to the hard drive, and it ran fine. This again seems to be a network issue?

Any thoughts would be greatly appreciated -- have a great weekend all!

 

[stumac]

There is a known corruption issue with Access running on Windows 10 where the back end is on certain versions of Windows Server - I think Windows Server 2012 is the worse culprit for it but also happens on 2016.

Unfortunately it seems to be inconsistent and unresolved. Although there is a workaround/hack for it, it isn't really suitable for many organisations (it requires a change to the server registry to disable leasing). There is quite a lot about it if you search for Access Windows 10 corruption issues. Here is a link that gives some more information

https://www.devhut.net/2018/06/13/access-bug-database-is-in-an-unrecognized-format/

As you will see the best 'fix' is to actually move your data to another DB client such as SQL Server - maybe not a bad option for an application with so many users.

 

[CPGDeveloper]

Thanks -- as a stop-gap, we've set up a new server for our back end, applying the Microsoft Suggested Registry Change -- I'll post an update about our results with that soon. Ultimately, our plan is to move our back end to Azure SQL Server.

 

[stumac]

Glad you have a solution, going down the Azure route seems sensible. Something I have been wanting to test with an Access FE, be good to hear how you get on with that!