Automic Workload Automation

  • 1.  Unable to execute the job, ending with error code 210

    Posted Mar 16, 2016 09:26 AM
    While trying to execute a simple job on the server through uc4, it is ending with return code 210 and not report is showing up. Sometimes the same job is going through fine, is it connection issue, agent connectivity from uc4 to server. Why some times the job is going through fine and why sometimes same job with the error code 210. Load balancing is implemented on the target server.  This is production env and very critical for us to schedule the job. Job is going through fine while triggering from the backend. 


  • 2.  Unable to execute the job, ending with error code 210

    Posted Mar 17, 2016 05:28 AM

    Hi John,

    does the job end wit ENDED_NOT_OK and errorlevel 210? If so, Automic executes your program correctly on the target machine, but the program itself fails.

    To rule out Automic as error source, you could manually login on the target machine with the same credentials used in the job and execute the program/command on the command line. If it also fails, you can be sure, that the failure lies inside the program.

    I'd also suggest to look up in the documentation of the program you are trying to execute what error level 210 means.

    If you need further help, please provide more information about the Job you try to execute. (OS, Command, ..)



  • 3.  Unable to execute the job, ending with error code 210

    Posted Jul 17, 2017 02:27 PM

    Guenter Schulmeister

     - Strangely enough - this same thing is happening to us starting today.  We have some Linux servers and suddenly today - we can run jobs on them as UserA without any issue, but when we go to run jobs on that same server with UserB - they fail with a rc=210.  Even an simple "ls -al" or "pwd" command".  They are able to login to the server with UserB and run commands, so not sure what is going on.  Any ideas anyone?


  • 4.  RE: Unable to execute the job, ending with error code 210

    Posted Nov 01, 2021 04:47 AM
      |   view attached
    Unfortunately we have also the same issue as described above. We figured out
    - ENDED_NOT_OK
    - Return code = 210
    - Job is not executed because simply comands are not executed.

    We have set the trace modus ON and d like to upload the trace file, so maybe you can find the error.
    In my opinion, the error can be seen in the file 89727199.REP.txt Line 125 after calling the function

    execle(/usr/bin/bash, -bash, "-c", /opt/automic/v11/agent/unix/bin/./../temp/JAHOJCOD.TXT, (char *)0, env_child(TERM=xterm, SHELL=/usr/bin/bash, USER=oracle, MAIL=/var/spool/mail/oracle, PATH=/usr/local/bin:/usr/bin:/usr/local/sbin:/usr/sbin, PWD=/home/oracle, LANG=en_US.UTF-8, HOME=/home/oracle, LOGNAME=oracle,

    Thanks a lot
    Luis


    Attachment(s)

    txt
    89727199.REP.txt   10 KB 1 version


  • 5.  Unable to execute the job, ending with error code 210

    Posted Jul 18, 2017 06:17 AM
    (edit): This is mostly in response to Laura, who reports getting the 210 return code for one user but not the other. This may not neccessarily apply to the original poster:

    Instead of "ls" or "pwd", I like doing something like "touch /tmp/testfile" from UC4 instead. If no such file is generated, you know your job didn't run. It would then appear your agent has a problem switching to user "B" but not user "A".

    Check the agent logs. Nothing there? If not, try to "su -" to the user from root. Any errors with that? Compare /etc/passwd and /etc/group for user A and B. Any significant differences? If still nothing, check if your agent binary is still owned by root, as is Automic recommendation (unless your company knowingly operates it under some other uid on purpose), and whether the setuid bit is still set on the agent binary (ls -la should have "rws" for the first triplet, not rwx or somesuch).

    Still nothing? You might then want to strace the agent. Find its PID (ps -ef | grep ucxjlx6), then run "strace -f -p <pid_of_agent>".   (edit: do that as root).

    Execute your test job. If it is indeed a problem with the agent not able to switch to the user, you should probably see a syscall of one of it's child processes in the output at the exact moment you run your test job, which ends with the 210 return code. Google or the man page for the syscall (if any) for possible pointers what's wrong with that account.

    Best,
    Carsten


  • 6.  Unable to execute the job, ending with error code 210

    Posted Jul 20, 2017 05:35 AM
    LauraAlbrecht608310 I'd try the method Carsten_Schmitz_7883 described to find out if there is a problem with the agent or the payload.