Error in schedulix jobserver

30 views
Skip to first unread message

Vanessa Romero

unread,
Aug 26, 2022, 4:35:53 AM8/26/22
to schedulix
Hello to everyone,

we have a problem in a jobserver of schedulix and we can´t be able to solve it.  If we start de service of schedulix on the jobserver, it keep in status inactive

 systemctl status schedulix-salesforce.service
● schedulix-salesforce.service - Schedulix SALESFORCE
   Loaded: loaded (/etc/systemd/system/schedulix-salesforce.service; enabled; vendor preset: disabled)
   Active: inactive (dead) since Thu 2022-08-25 16:18:14 CEST; 18h ago
  Process: 1522373 ExecStop=/home/salesforce/schedulix/parada_schedulix.sh (code=exited, status=0/SUCCESS)
  Process: 1522360 ExecStart=/home/salesforce/schedulix/arranque_schedulix.sh (code=exited, status=0/SUCCESS)
 Main PID: 1522371 (code=exited, status=0/SUCCESS)


But the proccess is running.

ps -ef |grep job
salesfo+ 1522801       1  0 Aug25 ?        00:00:05 /home/schedulix/schedulix/bin/scrolllog /home/salesforce/schedulix/log/jobserver.out -e /usr/bin/java -client -XX:ParallelGCThreads=2 -cp /home/schedulix/schedulix/lib/BICsuite.jar::/usr/java/jna/dist/jna.jar: de.independit.scheduler.jobserver.JobServer /home/salesforce/schedulix/etc/jobserver.conf jobserver[salesforce@fiberco1pro]
salesfo+ 1920953 1522801  0 10:22 ?        00:00:00 /home/schedulix/schedulix/bin/scrolllog /home/salesforce/schedulix/log/jobserver.out -e /usr/bin/java -client -XX:ParallelGCThreads=2 -cp /home/schedulix/schedulix/lib/BICsuite.jar::/usr/java/jna/dist/jna.jar: de.independit.scheduler.jobserver.JobServer /home/salesforce/schedulix/etc/jobserver.conf jobserver[salesforce@fiberco1pro]
root     1920955 1920741  0 10:22 pts/0    00:00:00 grep --color=auto job



In the jobserver.out, we have all the time the same error.

FATAL   [Jobserver]     26-08-2022 10:23:29 +0200 ***ERROR*** (04301271449) Missing '=' in environment:  do
FATAL   [Jobserver]     26-08-2022 10:23:29 +0200 Program aborted


And the jobs that must be run in this server never start, all the jobs keep in state ACTIVE/RUNNABLE.

Any idea?

Thanksss!!!!

Ronald Jeninga

unread,
Aug 29, 2022, 7:45:17 AM8/29/22
to schedulix
Hi,

what's most likely happening is that at startup of a jobserver it tries to read in the environment variables and it somehow fails to understand the output of the "env" command.
This can be caused by "garbage" (entries that don't look like KEY=VALUE) that is added by your favorite distribution.
Those entries are usually intended to simplify interactive use of the shell, but hardly play a role when doing batch processing.

Maybe I can spot the culprit if you post (or send me) the output of "env".
In the jobserver startup script(s) you can the unset that entry and the jobserver will run again.

Sorry for the late reply. I was OOO last Friday.

Best regards,

Ronald
Reply all
Reply to author
Forward
0 new messages