As this requires admin (still) and keeps giving slight trouble (like in failing
testcases with missing/other versions of rxapi) I much rather have Moritz’s
patches applied and tested for Linux and macOSX; this would bring the USB
version closer and would simplify the infrastructure. Because I build on every
machine and start rxapi by hand I am personally not bothered by this too much;
but it would be good to get rid of it.
Does the patch also work for Windows? I forgot.
A not working or firewalled (in the case of my first attempt on the Rockhopper
Linux on Z OS) rxapi yields deplorable performance (actually, after the IP
timeout all code runs quickly, but it is hard to spot the difference).
So my suggestion would be to apply the patch and keep on testing with that. The
old version would still be in svn when we need it.
best regards,
René.
> On 23 May 2017, at 16:14, CVBruce <[email protected]> wrote:
>
> I think I originally wrote this plist file. In the old make process the
> ‘make install’ would tailor the file for the current install location (i
> think).
>
> There was some discussion around this file and the rxapi process on MacOS.
> The problem stems from the MacOS architecture for daemons doesn’t like
> daemons that fork and exec when starting. The launch daemon looses
> connection to them. With the KeepAlive attribute the launch daemon thinks
> that the process rxapi has stopped/failed, so every few seconds it tries to
> restart the rxapi process. You can see this in the system log. I had
> re-written the rxapi code with MacOS conditionals, and these changes were
> rejected, because the whole process was scheduled to be re-written.
>
> As a work around, one user suggested that since the rxapi process rarely ever
> fails, that the KeepAlive attribute is unnecessary, and can be set to false.
>
> Bruce
>> On May 23, 2017, at 5:37 AM, Rony G. Flatscher <[email protected]
>> <mailto:[email protected]>> wrote:
>>
>> Enclosed you will find a MacOSX plist-file named
>> "org.rexxla.oorexx.rxapid.plist" which I use for the BSF4ooRexx MacOSX
>> installation.
>>
>> It turns out (by mistake) that ooRexx will run quite slowly, if rxapi is not
>> running as a daemon, hence mentioning this here and enclosing a working
>> plist file which has the following content:
>> F:\debug_oorexx\from_macosx>type org.rexxla.oorexx.rxapid.plist
>> <?xml version="1.0" encoding="UTF-8"?>
>> <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN"
>> "http://www.apple.com/DTDs/PropertyList-1.0.dtd";
>> <http://www.apple.com/DTDs/PropertyList-1.0.dtd>>
>> <plist version="1.0">
>> <dict>
>> <key>Disabled</key>
>> <false/>
>> <key>KeepAlive</key>
>> <true/>
>> <key>Label</key>
>> <string>org.rexxla.oorexx.rxapid</string>
>> <key>ProgramArguments</key>
>> <array>
>> <string>/usr/local/bin/rxapi</string>
>> </array>
>> <key>RunAtLoad</key>
>> <true/>
>> </dict>
>> </plist>
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Oorexx-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/oorexx-devel
