I don’t think this is strictly a Filemanager question but it is part of the question.
I have a C6 application that is being hosted on Azure for some users and self hosted on local PCs for other users.
The latest update has some file changes and it is triggering a Filemanager process to convert some TPS files, which I suspect haven’t been updated in some time. This update process is working fine for all the self hosted users but about half of the Azure hosted users are getting a problem with the updated files.
I’m not 100% sure what the problem is, it could be that…
The file is being updated, but the temporary file created is not being renamed to the appropriate file name, so when the app runs the customer.tps file, for example, is empty.
The file has a corruption of some sort? The techos who handle this report that restoring the original file and running TPSFix before calling the Filemanager process allows Filemanager to complete successfully.
I haven’t dug too deep into the process at this stage, I am unable to reproduce it, even with the original data that has problems on Azure, but I’m not running it on Azure.
Does anyone have any insight into what might be happening? e.g. a resources issue with Azure?
Nothing to do with Azure, but I’ve just found some undocumented features of windows 11, namely power & timer events and possibly more, not detecting them, occur 1hr 50mins after the machine is put in S0 Sleep mode (sleep with network).
These shouldnt occur, but the 1hr 50mins more than covers an unproductive officeworker with a 1hr lunch break, before finally putting the computer into true S0 sleep.
Knowing Azure & Amazon can mirror a machine and its components like a disk & ram to another machine (in another data centre), then freeze the first machine and switch network to the other machine (in the other datacentre), it sounds to me like you have a disk version control issue at the Azure level.
Thanks Richard for the suggestion, but the process is running for other files and it only seems to fail on two of the files, so I’m not sure how disk version control would explain my situation?
By “hosted on Azure” you mean on a VPS running Windows Server and accessed via remote desktop?
At some point I did have my C6.3 app hosted the same way i.e. on a Windows Server Standard, just with another VPS hosting company and I was experiencing a lot of strange issues in that environment while the same app running on desktop was behaving perfectly fine. I upgraded to C11 and all problems were gone.
As for migrating virtual machines from one physical machine to the other, I was doing that quite a lot on VMWare using VMotion and users working on these did to even notice it was happening. There was a millisecond delay while doing the actual switch.