One of our deployment scripts uses Python which Bamboo is able to run no problem. The issue comes when the Build Log doesn't seem to record the output in the same logical order specified in the Python script.
For example when the Python script is run on the command line you might expect to see the following output:
sending incremental file list graphics/ images/ files/ support/ SERVICE_NAME: apache TYPE : 10 WIN32_OWN_PROCESS STATE : 1 STOPPED WIN32_EXIT_CODE : 0 (0x0) SERVICE_EXIT_CODE : 0 (0x0) CHECKPOINT : 0x0 WAIT_HINT : 0x0
Instead the Bamboo Build log displays the following log:
sending incremental file list graphics/ imaSERVICE_NAME: apache TYPE : 10 WIN32_OWN_PROCESS STATE : 1 STOPPED WIN32_EXIT_CODE : 0 (0x0) SERVICE_EXIT_CODE : 0 (0x0) CHECKPOINT : 0x0 WAIT_HINT : 0x0 ges/ files/ support/
Does this imply that the commands are not working in the same logical order compared to the command prompt or is it just the way Java reads the output stream?
Either way is it possible to improve the output of the log as knowing the sequence of events is an important tool in the debugging stage.
Your script probably sends some of the output on stdout and the rest on stderr stream. These streams are separate, one is buffered and the other one is not. The fact that you are seeing them in certain order when you run the script on command line cannot be depended upon.
You can get the same result on command line and in Bamboo if you make the script output everything on stdout - ideally directly in the script, but if you can't, redirect the output.
Thanks for the tip which does seem possible in Python. At the moment I use one of the following commands:
# Basic call subprocess.call(apache_cmd, stderr=subprocess.STDOUT) # Advanced call to record output p = subprocess.Popen(apache_cmd, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) stdout, stderr = p.communicate()
The basic call on line 2 forces all STDOUT data to be handled by the parent’s file handler and all STDERR output should be captured by the same file handler as STDOUT. The same is happening on line 4 except STDOUT now goes to a new PIPE which Python can communicate with. Further explanation is given in the Python documentation:
stdin, stdout and stderr specify the executed program’s standard input, standard output and standard error file handles, respectively. Valid values are <tt>PIPE</tt>, an existing file descriptor (a positive integer), an existing file object, and <tt>None</tt>. <tt>PIPE</tt> indicates that a new pipe to the child should be created. With the default settings of <tt>None</tt>, no redirection will occur; the child’s file handles will be inherited from the parent. Additionally, stderr can be <tt>STDOUT</tt>, which indicates that the stderr data from the child process should be captured into the same file handle as for stdout.
Unfortunately these settings don't seem to help so unless you have any other ideas I could take this question to a more Python oriented forum.
The change has been applied in following way:
Beginning to execute external process for build 'Test Web Deploymet'
... running command line:
... in: E:\bamboo-home\xml-data\build-dir\TESTDEPLOY\web\etc\trunk\scripts
Still no luck
Looks like simply outputting to a text file outside of Bamboo causes inconstancies. After looking into it further the issue was indeed due to buffering, after forcing stdin, stdout and stderr to be totally unbuffered using the –u flag: http://docs.python.org/using/cmdline.html#cmdoption-u the issue resolved itself.
Welcome to your weekly Jira Ops Early access program update, where we’re sharing news and updates on Jira Ops' progress as we work toward our 1.0 release. If you ever want to drop us feedback or idea...
Connect with like-minded Atlassian users at free events near you!Find a group
Connect with like-minded Atlassian users at free events near you!
Unfortunately there are no AUG chapters near you at the moment.Start an AUG
You're one step closer to meeting fellow Atlassian users at your local meet up. Learn more about AUGs