Lars Fosdal

Would it be possible to wrap the assembly versions of these in Delphi? https://www.nayuki.io/page/fast-sha2-hashes-in-x86-assembly

If the need is simply parallel remote control like starting/stopping scripts/apps or services - you would do much better by using PowerShell.

@David Schwartz Have you guys tried https://remotedesktopmanager.com/ ? It is freaking awesome for managing a large server park! We have literally hundreds of servers and would be lost without it. It supports pretty much every remoting protocol you can think of: https://remotedesktopmanager.com/features/remote-connection-management You can have the windows tabbed, cascaded, tiled vertically or horizontally. You can lock the Windows hotkeys so that you have to use the mouse to get out of the context of the remote view. I.e. Alt-Tab happens within the focused remote connection. There is a free version for personal use, but if you have a team, you want the enterprise version.

Do all of them involve creating a modified FMX.FontGlyphs.Android.pas !? Is there a QP I can vote for to improve this somehow?

Must it be an overload? Why not FindComponentsByClass vs FindComponentsWithInterface ?

It is a joke - implying that if you surf from the space station, you get dating ads that are "near you" as spammers often use geolocation to generate fake ads for offers in your area.

Geolocation from IP address: https://www.iplocation.net/ip-address

Same as @dummzeuch We also leave range checks on for release versions. For us, API and database access times dwarf the array access times, and there is nothing quite like live data to throw curveballs at your algorithms.

@PeterPanettone I tried to fix that quote but only succeeded in making it worse. @Daniel - Something is broken when editing posts with quotes.

Integrity control.

If you want to use Hyper-V for virtualization, you need Pro. You can use VirtualBox instead. If you ever would need to join a company network, Pro is needed.

The process of making a change to the example file in the QP case would be painstaking and error-prone. I'd compress the original file, add it as a resource, and decompress in memory. No worries about encoding, quoting, and such.

Would it be feasible to memory map the utf8 file?

@bazzer747 Added a rough outline of how to use the FireDAC monitor

How to do FireDAC monitoring or logging to file. Here is a rough outline of what you need to do. The code is not ready to use as is and serves only to give you the general gist. Add FireDAC.Moni.Base, FireDAC.Moni.RemoteClient, FireDAC.Moni.FlatFile to your uses clause Somewhere appropriate, create the following - I keep them in my DatabasePool class. FTracing: Boolean // init to False FMonitorBy: TFDMonitorBy; // init to mbRemote FMonitorLink: TFDMoniClientLinkBase; // init to nil // Here is where the actual establishing of the tracing happens depending on the trace model. procedure TDBPool.SetMonitorBy(const Value: TFDMonitorBy); begin Lock; try if (FMonitorBy <> Value) and Assigned(MonitorLink) then FreeAndNil(FMonitorLink); FMonitorBy := Value; // mbNone or mbCustom disables the monitoring if MonitorBy = mbRemote then begin MonitorLink := TFDMoniRemoteClientLink.Create(nil); end else if MonitorBy = mbFlatFile then begin MonitorLink := TFDMoniFlatFileClientLink.Create(nil); TFDMoniFlatFileClientLink(MonitorLink).FileName := ParamStr(0)+'.'+FormatDateTime('yyyymmdd-hhnnss', Now)+'.FireDAC.log'; TFDMoniFlatFileClientLink(MonitorLink).FileAppend := False; end; if Assigned(MonitorLink) then MonitorLink.Tracing := Self.Tracing; finally Unlock; end; end; // Use Tracing to enable / disable the tracing function TDBPool.GetTracing: Boolean; begin Result := FTracing; end; procedure TDBPool.SetTracing(const Value: Boolean); begin FTracing := Value; if not Assigned(MonitorLink) then SetMonitorBy(FMonitorBy) else MonitorLink.Tracing := Self.Tracing; end; procedure TDBPool.SetMonitorLink( const Value: TFDMoniClientLinkBase); begin FMonitorLink := Value; end; // When the DB connection is created - it should assign the MonitorBy value procedure TDB_FD.CreateFireDACConnections; const OSAuthent = 'No'; begin if not Assigned(FConnection) then begin OnConnectionCreate; FConnection := TFDConnection.Create(nil); FConnection.DriverName := DBPool.DataBaseDriverName; // that you have selected - This example assumes an MSSQL driver - see other code block below FConnection.Params.Values[MSSQLParam.Server] := Trim(FHost); FConnection.Params.Values[MSSQLParam.Database] := Trim(FDatabaseName); FConnection.Params.Values[MSSQLParam.OSAuthent] := OSAuthent; FConnection.Params.Values[MSSQLParam.User_Name] := Trim(FUserName); FConnection.Params.Values[MSSQLParam.Password] := Trim(FPassword); FConnection.Params.MonitorBy := DBPool.MonitorBy; ... end; BTW - Selecting the right DatabaseDriverName can have a significant impact on performance. Here is how I currently pick my preferred MSSQL driver. Why are those constants are not in the interface section of FireDAC.Phys.MSSQL, @Dmitry Arefiev? class function TDBPoolMSSQL.FindBestDriver(const Link: TFDPhysMSSQLDriverLink): String; const // Constants copied from implementation section of FireDAC.Phys.MSSQL C_2018_ODBC = 'ODBC DRIVER 18 FOR SQL SERVER'; C_2017_ODBC = 'ODBC DRIVER 17 FOR SQL SERVER'; C_2016_ODBC = 'ODBC DRIVER 13 FOR SQL SERVER'; C_2012_ODBC = 'ODBC DRIVER 11 FOR SQL SERVER'; {$IFDEF POSIX} C_FreeTDS = 'FreeTDS'; {$ENDIF} {$IFDEF MSWINDOWS} C_2012_NC = 'SQL SERVER NATIVE CLIENT 11.0'; {$ENDIF} var DriverList : TStringList; WantedList : TArray<String>; begin Result := ''; // Blank = Default WantedList := {$IFDEF MSWINDOWS} {$IFDEF SQLNative} // defined if supposed to prioritize the old SQLNCLI11 [C_2012_NC, C_2017_ODBC, C_2016_ODBC, C_2012_ODBC] {$ELSE} [C_2018_ODBC, C_2017_ODBC, C_2016_ODBC, C_2012_NC, C_2012_ODBC] {$ENDIF} {$ENDIF} {$IFDEF POSIX} [C_2018_ODBC, C_2017_ODBC, C_2016_ODBC, C_2012_ODBC, C_FreeTDS] {$ENDIF}; DriverList := TStringList.Create; try Link.GetDrivers(DriverList); for var Wanted in WantedList do for var Driver in DriverList do begin // DebugOut('ODBC: "' + Driver + '"'); if CompareText(Wanted , Driver) = 0 then Exit(Wanted); end; finally DriverList.Free; end; end;

That was my first reaction as well - do you have a memory leak - i.e. allocate objects which are never disposed of? Private bytes is the one to watch. Our services go belly up when it reaches about 1.6Gb - due to something leaking. I use Process Hacker 2 over Process Explorer as I can get at more statistics for each process with that one. As suggested above, use FastMM to check for leaking objects. Check the number of handles allocated to the application. Do you do allocations through Windows APIs? Are those deallocated?

Looks hard at the provided examples of the problem, but draws a blank...

It is a good way of black-boxing old code. The drawback of using DLLs is that versioning the interface becomes important. Want to add a parameter to a function? It will break the compatibility with the old DLL, so better to add a second function. Want to change a structure? Another break - unless you planned for it and have structure versioning and size info, and always add content at the end of the structure. Win32 is jam-packed with the above.

Thanks for sharing the root cause! Such mysteries can be so hard to track down! I've seen this before on Windows - but didn't make the connection this time, if you pardon the pun. Does an Android service have a Console define? Our FireDAC helper unit uses that to handle Windows services and other console apps. unit PSD_db_FireDAC; /// <summary> Generic FireDAC wrapper. Use PSD_db_FireDAC_MSSQL and PSD_db_FireDAC_PostgreSQL units </summary> interface uses System.Classes, System.SysUtils, Winapi.Windows, System.Variants, System.AnsiStrings, System.TimeSpan, System.Diagnostics, System.TypInfo, System.Math, StringFunctions, TineInterfaces, TineClasses, PSDResourcePool, Generics.Defaults, Generics.Collections, PSD_Db_Abstract, PSDConstants, tiConsts, PSDLogEvent, PSDLogDrivers, PSDConnectivityStatus, Data.DB, FireDAC.Stan.Intf, PSDStatisticsEngine, {$ifdef Console} FireDAC.ConsoleUI.Wait, {$else} FireDAC.VCLUI.Wait, FireDAC.Comp.UI, {$endif} FireDAC.UI, FireDAC.UI.Intf, FireDAC.Stan.Option, FireDAC.Stan.Error, FireDAC.Phys.Intf, FireDAC.Stan.Def, FireDAC.Stan.Pool, FireDAC.Stan.Async, FireDAC.Phys, FireDAC.Stan.Param, FireDAC.DApt, FireDAC.comp.DataSet, FireDAC.comp.Client, FireDAC.comp.Script, FireDAC.Moni.Base, FireDAC.Moni.RemoteClient, FireDAC.Moni.FlatFile;

Defeatist? More like a realist. After getting rid of the per field micro locking and properly handling contention, the problem was greatly reduced. Even so - carefully crafted sins appear from time to time. Such is the life of dealing with legacy code.

My apologies for being unclear: One crit sec per object instance, with separate locking per object field. Return to previous description.

Can you reproduce the problem in a bare minimum app?

The deadlock on single lock instance for multiple shared instances (read; individual access locking to multiple fields of a class) is most likely to happens when multiple fields for multiple objects are accessed from multiple threads for the same objects. In our case, it was parallel processing of warehouse picks. F.x. Same article, same location, different picker, needing to update the stock saldo at the same time. It involves interaction between a relatively large number of object instances, such as object cache list retrieval and the objects themselves. I could exemplify in detail, but then I'd only get "but you are doing it wrong" - which I was - and that is why I work hard to minimize the use of implicit locking per field, and rather lock the entire object when needed for as short a time as possible, and avoid locking as far as can be deemed safe. This is not academically flawless code. It is code developed by a multitude of people over a long time.

Is there any design time settings on any of the db components that may be in violation at time of startup?

Here is another unhelpful blanket statement, based on real-world, empirical evidence. Multiple threads, multiple objects, cross-thread cross-object access and micro locking per field using the same single lock per object = high risk of deadlock. Sure, you can attribute it to bad design - but perfect designs for software that evolves over decades are really hard to do.