Solved Automated search for Database Corruption Assistance

According to your comment, the user's desktop is on OneDrive and thus the FE file is also on OneDrive.
This is not the issue and is resolved - although it wouldn't have been if you had not commented - much appreciated. See Reply #192. The FE was on OneDrive on Monday night and Tuesday morning, but the user isn't actively working in the database yet, and the file duplication happened apparently around the 18th.
PLEASE tell me that the BE is also not subject to OneDrive periodic backups. Because if you have a crash during a long transaction and try to restore, your backup copy will be unusable.
I don't believe the BE is subject to OneDrive, but I'm not certain. The BE is on \\networkname\networkdirectory. There is no OneDrive in the directory tree and there was in the new users Desktop folder location. OneDrive is new to us at the user level, so I'm not certain how it and where it is being utilized.
Your solution MUST be to have the FE on a purely local (i.e. outside of the range of OneDrive's reach) folder on the user's desktop system, or one of Pat Hartman's CITRIX/RDP solutions with a private folder on the RDP server.
Let's look at the second part of this ... Half of our team (8 users total) operates with the FE on the users desktop on the local machine (C:\Users\username\desktop). The newest team member is using a SHORTCUT on the OneDrive Desktop to open the FE in the stated folder.

The other half of to team (4 users) accesses the database through Citrix. Per policy, they can't install the FE on their local PC. For them, the FE is installed on a network (\\networkname\username\) folder that only the individual user has access to. Could this be part of the issue? One of our db admins on a different database tends to blame it, and other databases in the company the don't rely on Citrix don't seem to have the issues we typically run into.

I don't know how much support I might get trying to have the FE moved to a private folder on the RDP Server. I'm not familiar with Pat's solution. Generally when there has been an issue, the Citrix team has told me that Access is not supported on Citrix.

Question 2: You mentioned VERY early in this thread that the issues I mentioned could happen if a record from one table was pasted into a different table, and I said that seems VERY likely what was happening (at that time, but somewhat still). How could this happen? The users can't directly access the tables, the layout/datasheet/Excel view is disabled, the navigation pane, ribbon, and navigation bar and F11 are disabled. Is there something obvious I overlooked like Ctrl-C in one form and Ctrl-V in another form? I know it seems silly, but it's a serious question.

In addition: table definitions do not simply change at will, sometimes PK, sometimes not, values duplicate themselves or disappear. This is nonsense.
This can only be understood as corruption in the tables, and a third party is involved.
I don't want to get into an argument or semantics of whether or not it is considered corruption. I have tables in the backup file that were labelled as PK and are not labelled as PK in the current file. I have tables in the backup file where two records have identical values in the PK field, which is not supposed to be possible with Access, as far as I understand it (which isn't that far).

It may well be that a third party is involved, but I'm not sure how to find that third party and prevent it.

Are we talking about the backend being an Access file and all the tables being viewed being in this backend? A file is a stupid creature and cannot defend itself against external measures. If a user is currently carrying out write operations and another user or an admin service is simultaneously carrying out copying or compression measures on this file, this is potentially problematic and can lead to corruption.
Yes, backend is an Access file. Tables in question are linked into the FE from the BE. As far as I can tell, the third sentence shouldn't be happening. Most users don't know where the BE file is located to try to run backup or compression on it. I run a nightly unattended routine that does this, but there is a routine that kicks all users out first and the routine will fail if anyone is in the file (and I can verify it failed), but that hasn't happened during the time the duplication has occurrred.
 
or one of Pat Hartman's CITRIX/RDP solutions with a private folder on the RDP server.
I'm not familiar with this. Is there more to it than ensuring the FE is installed on the Citrix server and each users runs their own copy of it?

Could the front end being on one network folder and FE being on a different network folder (neither part of the Citrix Server that is called both of them) cause the issues I'm seeing?

Finally, we run multiple servers. Would the one user running his FE on Server 1 and another user running his FE on Server 2 (from folders on the Server) cause any issues? (I'm assuming not - conceptually this would be no different from me running the FE from my local computer and another user running the front end from their local computer.)
 
1. You said
The newest team member is using a SHORTCUT on the OneDrive Desktop to open the FE in the stated folder.
The shortcut can be anywhere since it isn't active once the launch of the targeted file/app occurs. Nothing else happens to the shortcut. If I remember this correctly, the shortcut is treated as though it was a one-line command line equivalent to "RUN X with parameter Y". So running via a shortcut is SIMILAR to running from a batch script... a very SHORT batch script. (Could be wrong - but I think that is correct.) The shortcut's properties and content are not altered, so even if OneDrive was involved, it would have no updates to make to the cloud copy.

The crucial aspect is where the targeted app resides. In theory, since the implied launch occurs on the user's PC, the launched code - the FE in this case - will have to have a clear, unobstructed path (in the network sense) to the PC and its memory. If you can't put a shortcut on the desktop but CAN have a user's files on that machine in a local folder, don't make the shortcut a desktop shortcut, but instead make it just an icon in the user's home folder. (Technically, a desktop shortcut is just another file in the implied folder that is the desktop.) Open Windows Explorer (or whatever they are calling it these days) to the user's working folder and launch from there.

2. You asked
For them, the FE is installed on a network (\\networkname\username\) folder that only the individual user has access to. Could this be part of the issue?
As long as each user's FE resides in a folder where a different user won't try to open it, you should be OK.

The thing you want to avoid is that two users will open the same physical file. (Not talking about users opening COPIIES of the same file.) When two users open the same file simultaneously, you have locking issues AND you have the potential for the file's contents to get scrambled. That is why we tell you to copy the FE and only share the BE file.

3. You asked about using CTRL/C and CTRL/V as copy/paste to move bulk data from one place to another.

If you have any tabular forms (similar to datasheet view) where a group of controls could be copied at once, I suppose it is possible to get all bollixed up with a bulk copy/paste. People used to Excel-like behavior have some bad habits to break and that is one of them. But for that to happen, you have to have a grid-like form in two places where a copy-from and a paste-to can occur. I would think you would know whether you had such a thing.

Let me digress for just a moment. In Excel, you have the explicit ability to select multiple cells and do a CTRL/C copy, then move to another sheet or segment thereof and do a CTRL/V paste. When you do that, though, Excel asks you whether you wanted to jam everything into a single cell or do a paste maintaining separation of cells or do some other type of paste. It also matters whether you select enough cells in the desitnation area to hold the cells you selected from the copy area. In any case, you get options.

When you do that grid-oriented copy/paste with Access, you don't get options because Access treats grids differently. In Excel, each cell is a unique entity unto itself. EVERY CELL can be something else. In Access, the cells in a tabular grid are constrained by record layout definitions and explicit or implied data types. You can build an Excel spreadsheet with a column that you are using as a scientific number. You can then copy/paste a number from that column into another column that is text. (Or vice versa.) And Excel WILL NOT CARE. In Access, try that and you might get a type misatch. Therefore, I would expect bizarre behavior in trying to do a bulk copy/paste in any grid-like display in Access.

Can this happen? Yes. Does it happen for you? Only if your users can get to grid-like displays.

This is getting long. I'll start a new one.
 
Is there more to it than ensuring the FE is installed on the Citrix server and each users runs their own copy of it?

The MSAccess.EXE app that is part of Office must be installed on that machine. But if multiple users will run on the same machine, there can be a licensing issue since the garden-variety copy of Office is single-user. @Pat Hartman has some experience with CITRIX sites and multi-user licenses so perhaps she can offer insight.

Could the front end being on one network folder and FE being on a different network folder (neither part of the Citrix Server that is called both of them) cause the issues I'm seeing?

I am not sure I understand the configuration you just described. Are you saying that the FE runs in machine A's memory but the actual FE file is on machine B's disk?

That should run but should have absolutely abysmal performance and your network guys would probably come after you one night with torches and pitchforks. For efficiency, an app must run where the physical file embodying that app is located. Otherwise, you swamp the network with program traffic (distinct from data traffic). You would have to load forms, queries, and modules with event code to even launch the app before it touches data. Then, each form action (opening or closing) ALSO triggers data movement from the BE file. If the FE is running locally, ONLY the BE data is transferred, which is normal and what most of the world of Access users see. But if neither the FE nor the BE are local to a machine, then EVERY ASPECT of the operation triggers multiple data bursts from multiple places (FE/BE).

Would the one user running his FE on Server 1 and another user running his FE on Server 2 (from folders on the Server) cause any issues? (I'm assuming not - conceptually this would be no different from me running the FE from my local computer and another user running the front end from their local computer.)

Here, you are correct. If the CITRIX/RDP user is running the FE from a private folder on the remote server, it is the same as a user running from a private folder on his desktop machine. Physically different location but logically identical isolation. (See above regarding legalities, though.)
 
I'm good, but I'll reply to clarify.

The MSAccess.EXE app that is part of Office must be installed on that machine. But if multiple users will run on the same machine, there can be a licensing issue since the garden-variety copy of Office is single-user.
I don't think that presents an issue. All of the Citrix users have the ability to launch MS Access (along with MS Word, MS Excel, and various other apps.) The issue is getting them a unique folder that they alone have write access to. I know that CAN be done, but I don't have control to authorize it.

I am not sure I understand the configuration you just described. Are you saying that the FE runs in machine A's memory but the actual FE file is on machine B's disk?
Correct - technically 4 machines involved:

User opens their laptop and launches Citrix desktop which I am going to say is running on virtual server A - but there are several Virtual Servers and they load share and they have MS Access installed. From the Citrix server, users goes to \\network_B\<username>\Frontend.accde, which only they can access and they double click and it opens MSAccess on Virtual Server A and links remotely to \\network_C\<backend_folder>\backend.accdb.

Oddly, I tried opening the fe from \\network_B on my local computer and it was abysmally slow, but it eventually worked. Under Citrix it is faster then opening the front end on my local computer directly on the network.

The way to do this is very simple. In the script that the Citrix folks create, all they need to do is to copy the master copy of the FE from the server to the user's personal Citrix directory.
The issue is there is no user's personal Citrix directory - as far as I know. Is the setup I described above problematic.

Reply #204 went WAY over my head ... I don't think non-Citrix users could access the backend if it was on the Citrix servers and they change them around fairly often. I can't put the backend on Network B b/c it is limited to the individual user. I could create restricted folders on Network_C and have the FE and BE on the same machine, but I don't know if that gains anything.
 
We handle things a bit differently. I've seen implementations where the FE is copied from the master directory to a local directory and then opened. What we do if have a version number in a table in the backend. If that number is greater than the number in the local table in the front end, than the new master version downloaded and opened.

There is ALWAYS a personal directory.
Okay - I don't know what you mean by it, perhaps. I rarely use Citrix, but I have Access to it. In Citrix, I am instructed to save my work in \\Network B\<username>\ - which is only accessible to me (and the network admins) and which I can also access from my local computer, so it is not on the Citrix server. There is also a C:\Users\<username>\Desktop folder on the Citrix server, but it is empty and I can't write to it.
If the FE is not COPIED from the master directory to the users personal directory, then they are sharing the FE and that is a problem.
They are not sharing the FE. They are all running their copy of the FE from their Network_B folder which only they can Access. Now, what concerns me is that I don't know if they are sharing the main Access program (not sure if that is possible) and also I'm not sure how the server memory is affected since the FE appears to be on a different computer than the Access program on Citrix.
Explain to them, in very small words, that you would never allow userA and userB to both open the same physical copy of Word.exe from a server folder. Access is no different.
Not following this. UserA and UserB open the same version of Word.exe from \\Server\ProgramFiles\Word - not good. UserA opens Word.exe from \\Server\UserA\ProgramFiles\Word and UserB opens Word.exe from \\Server\UserB\ProgramFiles\Word - seems to be more what is happening, but I'm not sure if that would create issues or not.
 
I explained how to tell. Look at the date/time of the master copy in the master folder. If it is static, they are not sharing that file.
If I went with your implementation.
There isn't an expert here who will tell you that sharing the FE is not dangerous. You absolutely do not want the same physical copy of an Access FE opened in memory on multiple computers at the same time. PERIOD.
And that is not happening. The front end won't open unless it is in a directory that only that user has Access to.

The central question at this point is whether under Citrix there is an issue with the FE being on one server folder and the BE being on a different server folder - neither of which are part of the Citrix Server.

There is no reason to not replace the FE every time the database opens if that method is easier to implement, especially with Citrix. Do what you want.
I'm convinced either method works. The advantage to your method is the user ALWAYS has the latest version of the front-end and it is simpler to implement. The disadvantage (especially for the VPN users) is that the database has to download and open everytime you open it, even if it hasn't changed at all.

It would be trivial to implement your method under Citrix, except I don't have write Access to ANY folders on the Citrix servers. I would either need the Citrix admins to set it up, or I could write a file to download the latest version to the network share, but the database does that already when I do an update - and I'm still not clear if that (having the FE separate from the Citrix server) causes an Issue.
 
There is also a C:\Users\<username>\Desktop folder on the Citrix server, but it is empty and I can't write to it.

This folder that you just described is a zero, a place-holder, someone's mistaken idea of giving users privacy. I'm thinking that an apt comparison is a "screen door on a submarine." It does NOTHING for you. It has no practical reason for existence if you can't write to it. But this is the kind of very short-sighted foolishness that an inexperienced admin will do to you. I was a system admin for the U.S. Navy for 28 1/2 years. I had to educate some of the security managers about Office and its true behavior. So if this configuration was what your IT guys did for you, they did so from lack of knowledge about Windows File & Printer Sharing protocol, known as SMB or Server Message Block, and its requirements. They also didn't know or forgot that the true purpose of an admin is to NOT get in the way of a user with a legit need. We admins SHOULD have the attitude of enabling the necessary and implementing the possible. Unless you are a malicious hacker, there should be an environment where YOU as a user don't even realize that you actually ARE sharing a machine. I learned that lesson my first year on the job when a program manager and I had a long discussion about security. We came away with a compromise because I still had security mandates that I had to follow and he still had goals for his people to meet.

If that C:\Users... folder is your default folder, you start life hamstrung. Yes, there are environmental variables that one can define to point your operations to a different folder. Yes, they will work. But not with complete simplicity.

I am instructed to save my work in \\Network B\<username>\ - which is only accessible to me (and the network admins) and which I can also access from my local computer, so it is not on the Citrix server.

I'm going to presume that B represents some form of network-attached storage (NAS). If that is where the FE file is located, then I'm not surprised by your comment

Oddly, I tried opening the fe from \\network_B on my local computer and it was abysmally slow, but it eventually worked.

Look in my earlier comments from a couple of posts ago. If the FE is not local to the PC, then you not only have data traffic but PROGRAMMING traffic going across your network. Each time you open a form, you have to reload the form. I hope it loads the class module with it. Every form, every query, every report, and every module that you reference has to be loaded from the FE location to the memory of the active PC. Forms and reports close when done, and when they close, that means that they must be reloaded if you open the form again.
 
We are getting away from my central question - i.e.:

I'm still not clear if that (having the FE separate from the Citrix server) causes an Issue.

This folder that you just described is a zero, a place-holder, someone's mistaken idea of giving users privacy. I'm thinking that an apt comparison is a "screen door on a submarine." It does NOTHING for you. It has no practical reason for existence if you can't write to it.
I'm not tracking here. Locally, we are using C:\Users\<username>\Desktop. I can get to that. I have other folders named C:\Users\<otherusername>\ that I can't open.

What is think it does is since the database won't open unless it is located there, it prevent two users opening the database from the same folder (the master file location on the network - for example).

Are you saying it is a BAD idea to have the front end in this folder.

From what I have seen and been told - there aren't ANY folders on the Citrix (Virtual) Servers that end users have write permissions to. Not C:\Users\<username>\Desktop, but not any other folder either.

I'm going to presume that B represents some form of network-attached storage (NAS). If that is where the FE file is located, then I'm not surprised by your comment
Sounds correct, but I'm not sure. I know we have to request it. It is typically mapped to a local drive. I can access my directory on it and subdirectories. If I am on someone else's computer, I can access their folders and subdirectories on it, but not my own.
If the FE is not local to the PC, then you not only have data traffic but PROGRAMMING traffic going across your network.
Don't know the details, but under Citrix the virtual server opens the FE on the external network and the backend on the different network extremely quickly. So I don't know what to say here ...
 
The hard bit that even a lot of system admins don't really get, is that logins to a RDP server, or a Citrix set up, have to be directed to a unique folder for each user.

Although in theory a single access front end database is multi user, in practice there can be difficulties and we advise against it..

One reason is that it's easy and convenient for a developer to store information about a process in a local table (ie inside the front end). If more than one user is using the database, they might interfere with that local information, which will cause problems.
 
I'm feeling a bit better, but still confused.

Essentially, @The_Doc_Man is saying what I have setup is horrible for network traffic and @Pat Hartman is saying everything is normal.

No two users are opening the same copy of the FE. It is not being shared.

Please read #204 again.
#204 wasn't clear to me that first time nor the second time I read it ... from 204:
Assume you have a LAN with 3 servers. A file server, a database server, and a Citrix server.
As I understand this, in our case, the BE is on the file server, the FE is on the NAS storage the @The_Doc_Man mentioned, there is NO database server, and the Access program is running on the Citrix Server, which is virtual.
Typically if the BE is Jet/ACE, both the BE and the master FE will be stored on the file server. If the BE is SQL Server, the master copy of the FE is stored on the file server and the BE is hosted on the database server. When the BE is Jet/ACE, you can give your Citrix users a boost by using the Citrix server to host the BE database.
Actually, the BE and the MASTER FE are on the file server. Each users FE is on the NAS - which I assume is also a file server.

Unless I'm missing something, you are saying I can host the BE database on the Citrix server, but above you just said I can't, shouldn't and shouldn't need to write anything to the Citrix server, so ...
If your users are using Citrix, there is absolutely no reason to be using a VPN. If you are using the VPN anyway, you access Citrix from within the VPN and the FE is NEVER downloaded to your laptop, EVERYTHING happens on the Citrix server.
I wasn't very clear here and I'm not sure how it works. The non-Citrix users need the VPN to access the BE files on the file server. I'm not sure if the VPN is required to access the Citrix Server or not. I think it is, but I agree with the rest of the statement. Nothing happens on the laptop storage - not sure about memory.
 
No two users are opening the same copy of the FE. It is not being shared.

Whatever else is going on, this part is exactly right.

My comments about having the FE directly available to the end user's PC have to do with efficiency and performance, but if you have to load things remotely, it will work. It will lose races to snails on a salted track, old turtles, and cold molasses... but it would work.

The central question at this point is whether under Citrix there is an issue with the FE being on one server folder and the BE being on a different server folder - neither of which are part of the Citrix Server.

No issue on that part of the configuration. A split FE/BE is a case of a file being in one place and a different file in another place. As long as the hosting systems in question can see each other and exchange data, no problem with connectivity. The efficiency of a particular connection is an issue. The earlier business about OneDrive is potentially an issue, though you've been attacking that. And your corruption events are another issue.

Locally, we are using C:\Users\<username>\Desktop. I can get to that. I have other folders named C:\Users\<otherusername>\ that I can't open.

That might simply be Windows "isolation" security kicking in. Old rules existed back in the days when computers were gradually infiltrating government offices on a more frequent and invasive basis. Something referred to as "The Orange Book" defined certain standards such that if a computer complied with a certain set of rules, it would have a particular security level. The U.S. Navy mandated that any computer used in any Navy office would be not less than "C2" level. Under C2 security, certain rules are relevant. BTW, the Orange Book has been superseded by a more comprehensive (and therefore less comprehensible) book so don't be surprised if you can't easily find it.

1. Fine-grained discretionary access control (DAC) - i.e. you can control specific actions for specific files and different actions for other files in the same folder. Detailed network permissions are part of DAC. The idea that permissions apply according to the current username is part of DAC as well. Object permissions and device permissions are another part of that.
2. Session login procedures are needed to establish credentials/abilities/file access rights
3. Object auditing (forensic but not necessarily fault-recovery-level audits). Recovery audits are an applications thing, most often.
4. Objects cannot be re-used without first being "sanitized" - in most common practice, erase memory before granting to a user; erase a disk block before allocating it to a user's file.
5. Isolation of resources - in practice, task A cannot interact with task B unless there is a predefined interface between them.

In case anyone was wondering, Windows NT and subsequent versions either fully complied out of the box or only needed minor option tweaks to be C2 compliant.
 
@Pat Hartman

Pat. I often work inside the front end. For instance if I am preparing a quotation, I work on temporary tables in the front end. If the user cancels, there's no issues. If the quote is finalised, then the temporary tables get appended to the back end. So in this case having multiple users sharing the front end would cause an issue.

In another app, I have a dashboard type report using a temporary table in the front end, that gets filled by running multiple queries. Again having a shared front end database with multiple concurrent users might interfere with that.
 
When did the NAS come into play? Is the NAS drive defined as a standard LAN server or is it Web enabled? If it is the former, it is identical to using any other LAN file server. If it is web enabled, it is an accident waiting to happen because of everything Doc has said about protocols.
See Reply #210.

I'm not sure if it is web-enabled or not. I tried to access it from the address bar in Google Chrome and was not able to do so. However, I was able to access the folder that the BACKEND of the database is located in through Google Chrome. Is that an issue?

Remaining questions:
  • If that is an issue - I'm not sure where IT could host the backend. Is it possible for one folder on the server to not be web-enabled, but the folders above it to be web-enabled?
  • Under Citrix and/or for corruption issues - is there an advantage to having the FE and BE on the same server? The current setup is working pretty well (maybe?) and I can't put the backend on the NAS b/c there isn't a folder there that all users have access to, but I could probably setup or have IT set up individual user folders on the same server as the backend is on. The database is very fast under Citrix, so I'm thinking the answer is no, but ...
  • Still need to figure out what is going on with records being messed up if not related to the above.
Thank you again!!!
 
@Pat Hartman I frequently use a NAS drive at home to host BE files on, and I can get to it and the files via a web based app that runs on it, no issues. It's a Synology unit.

I wouldn't try and link to the BE files from Access via the web portal route, and I think that is the key difference you are getting at?

I use it replicate customers set ups, and shared folders to check operational speed, as when everything is on my local SSD the access times are negligeable, so it's handy as a performance check.
 
Let me explain the setup in as much detail as I can. I don't know if we are using NAS Drives. @The_Doc_Man mentioned them and that seems correct.

We have two classes of users. One group uses Citrix. The other group uses local resources and uses VPN (Pulse Secure/Ivanti) when working from home.

The BE file is on <Network A>\folder structure\BE.accdb. Using VPN, I can get to this folder from Google Chrome. I assume that means it is web enabled and not acceptable. You never explained, can this folder be made non-web-enabled, or do I have to find a completely different server location? The BE has been in this location for twenty-some years and there are other database backends in the same server, but we've had intermittent data issues for that long also.

Also - if it makes a difference, nobody uses a web-based interface to get to the BE files. Not sure if that matters or not ...

The non-Citrix users have the database FE on their local machines in C:\Users\<username>\Desktop\. The database checks if it is in this location on startup. And the non-Citrix users are using Access M365 64-bit.

The Citrix users have the database FE in <Network B>\Data\<username>\ I don't know if this is NAS or not. I have a folder on this server and I know that I can't get into other users folders and they can't get into mine. This network does NOT appear to be web-enabled - i.e. I can't get to my folder in Google Chrome.

I'm not sure how Citrix is set up. I go to a Citrix Portal website and click a Desktop shortcut and it opens a virtual server. There are multiple virtual servers and the virtual servers are running Windows Enterprise 2016 and Access 2016 64-bit.

If, you can access the NAS from your home computer using Google Chrome, then YES, it is web enabled and you can't use it to host your Access BE.
I'll need to look into getting this changed. Again - can I disable the BE Folder from being web-enabled, or do I need to find a totally different server?
I explained this to you and why it is advantageous in my example way back.
You did and I misunderstood twice. At first I thought you meant I should load the FE on the same virtual server that Citrix is running Access on. Then I thought you meant Citrix loads both the BE and the FE into memory so it really doesn't matter.

Either way, Citrix operates VERY quickly. Now if I could speed up the VPN operation ...?
 
I don't have a network set up where I can test this.
Is there a way I can test it? It doesn't SEEM to be working as you describe. For example - we have multiple users in the database. I haven't heard of anyone saying their changes weren't saved b/c someone else updated a record in the same table. Also, someone can open a record and save it and someone else viewing the same record will see the changes. We do have issues where two users change the same record WITHOUT saving it and then Access asks if they want to overwrite the other changes or discard - which I think is normal.

Basically, we don't have issues with changes being dropped. We have issues with records being duplicated or disappearing.

I would not involve a VPN when I have Citrix as an option.
Thanks - Generally I have found Citrix faster then in-office which is much faster than VPN. The downside of Citrix is you have to wait for the Virtual Server to load, which takes almost as long as loading the database.
 
OK, we have to clear the air a bit.

"Web-enabled" means that a web server has been given information about where to find the files associated with a web site. Anyone coming to that computer using HTTP / HTTPS protocols would probably find a socket waiting to serve their connection. But the host system's disk is just a disk. It has folders, file owners, permission codes, and files just like any other disk.

A Network Attached Storage (NAS) disk is just a disk, too. However, it rides a network connection rather then an ATA or SCSI or FIBER channel and it also has a microprocessor acting as its controller, with a very basic device driver, maybe just barely above the smarts of the BOOT ROM disk driver that is used when you reboot a system. The ONLY thing an NAS disk's processor can do is disk I/O and network I/O. Nothing else... for the most part.

If it is a "smart" NAS disk (high-end device) then it MAY have been set up with some sort of exclusivity with regard to inbound protocols. That is actually no different from a locally attached disk whose host system may have been set up regarding exclusive accessibility on that disk. Cloud drives often have this exclusivity and that is why putting a back-end file on a cloud drive often has protocol compatibility issues. But note that a cloud drive can be an NAS disk or a system-hosted disk. So we have to watch out about mixing labels.

An NAS drive can be set up to look like it is accessible by other protocols than HTTP/HTTPS. If it is accessible to SMB protocol, which is the Windows File and Printer Sharing protocol, then a BE can comfortably reside there. I've used that kind of drive myself many times with the Navy. The key is what protocols its controller recognizes from the network to which it is attached.
 

Users who are viewing this thread

Back
Top Bottom