Page 1 of 1
Exit codes from spawned processes
#1
Posted 06 October 2008 - 02:33 PM
Hi yall,
My client is running Maestro 8.2 (yes, it's true), and there's a bit of disagreement in regards to how Maestro handles exit codes from processes.
If a job kicks off a process, which kicks off a process, which kicks off a process (think great-grandchildren here), does Maestro itself pick up the exit codes from all these spawns? Or do the exit codes get passed back up the chain and then to Maestro?
Thanks for your help!!!
Jen
My client is running Maestro 8.2 (yes, it's true), and there's a bit of disagreement in regards to how Maestro handles exit codes from processes.
If a job kicks off a process, which kicks off a process, which kicks off a process (think great-grandchildren here), does Maestro itself pick up the exit codes from all these spawns? Or do the exit codes get passed back up the chain and then to Maestro?
Thanks for your help!!!
Jen
#5 Guest_Timmy_*
Posted 20 October 2008 - 04:58 PM
Jen, on Oct 20 2008, 05:08 PM, said:
Timmy,
I appreciate it, but it's not your responsibility to do that. I was just hoping someone knew. I searched through all the pertinent Redbooks, and couldn't find anything.
Jen
I appreciate it, but it's not your responsibility to do that. I was just hoping someone knew. I searched through all the pertinent Redbooks, and couldn't find anything.
I would be the worst admin in the world if i didnt try and help out my users
Its no probs we'll do our best
Timmy
#6
Posted 09 November 2008 - 12:50 AM
You are probably having such a difficult time because ultimately the answer is "it depends".
If each application in the stream returns it's return code back to the parent application and the parent application in turn returns the return code of the child as its own, then the answer is "yes".
Unfortunately this is extremely unlikely and you are most likely stuck with the return code of the original application that is spawned as this application is going to have activities that it performs after the child application has completed (e.g. logically the child application was spawned in order to return some data to the parent or complete some sub-process on the parent's behalf) and as a result the return code of the parent application will be reflective of the results of the cumulative activities of the parent where the success of the child sub-process may or may not be relevant, but most certainly will be interpreted and/or handled.
If you are launching applications that effect a pipeline (e.g. Windows cmd.exe or powershell.exe) then you have a much better chance that the return codes will be passed back up the pipe.
Ultimately, I would recommend that you design your processing based on the answer to this question being "no" as that would be more practical.
If each application in the stream returns it's return code back to the parent application and the parent application in turn returns the return code of the child as its own, then the answer is "yes".
Unfortunately this is extremely unlikely and you are most likely stuck with the return code of the original application that is spawned as this application is going to have activities that it performs after the child application has completed (e.g. logically the child application was spawned in order to return some data to the parent or complete some sub-process on the parent's behalf) and as a result the return code of the parent application will be reflective of the results of the cumulative activities of the parent where the success of the child sub-process may or may not be relevant, but most certainly will be interpreted and/or handled.
If you are launching applications that effect a pipeline (e.g. Windows cmd.exe or powershell.exe) then you have a much better chance that the return codes will be passed back up the pipe.
Ultimately, I would recommend that you design your processing based on the answer to this question being "no" as that would be more practical.
#7
Posted 11 November 2008 - 09:38 AM
Jeff, on Nov 9 2008, 01:50 AM, said:
You are probably having such a difficult time because ultimately the answer is "it depends".
If each application in the stream returns it's return code back to the parent application and the parent application in turn returns the return code of the child as its own, then the answer is "yes".
Unfortunately this is extremely unlikely and you are most likely stuck with the return code of the original application that is spawned as this application is going to have activities that it performs after the child application has completed (e.g. logically the child application was spawned in order to return some data to the parent or complete some sub-process on the parent's behalf) and as a result the return code of the parent application will be reflective of the results of the cumulative activities of the parent where the success of the child sub-process may or may not be relevant, but most certainly will be interpreted and/or handled.
If you are launching applications that effect a pipeline (e.g. Windows cmd.exe or powershell.exe) then you have a much better chance that the return codes will be passed back up the pipe.
Ultimately, I would recommend that you design your processing based on the answer to this question being "no" as that would be more practical.
If each application in the stream returns it's return code back to the parent application and the parent application in turn returns the return code of the child as its own, then the answer is "yes".
Unfortunately this is extremely unlikely and you are most likely stuck with the return code of the original application that is spawned as this application is going to have activities that it performs after the child application has completed (e.g. logically the child application was spawned in order to return some data to the parent or complete some sub-process on the parent's behalf) and as a result the return code of the parent application will be reflective of the results of the cumulative activities of the parent where the success of the child sub-process may or may not be relevant, but most certainly will be interpreted and/or handled.
If you are launching applications that effect a pipeline (e.g. Windows cmd.exe or powershell.exe) then you have a much better chance that the return codes will be passed back up the pipe.
Ultimately, I would recommend that you design your processing based on the answer to this question being "no" as that would be more practical.
In other words, no.
Page 1 of 1



Start a new topic
Add Reply
MultiQuote

Sign In
Register
Help