I have a FME Server (2018.1.01) set up with two local engines and one remote engine. This remote engine periodically shuts down and requires restarting. I haven't found a pattern of anything causing it to shut down. Is this a known issue? I also cannot manage the number of engines for this host, it isn't listed as a host on the engines page in the web interface.
Page 1 / 1
Here's the FME Server troubleshooting guide for engines.
Looking at it, the shutting down might be related to scheduled rebooting after a certain number of jobs/features processed, but in that case it should come back automatically.
Here's the FME Server troubleshooting guide for engines.
Looking at it, the shutting down might be related to scheduled rebooting after a certain number of jobs/features processed, but in that case it should come back automatically.
This engine seems to be down more than it is up now. It isn't used all that often so I don't know immediately when it is down, just when a user goes to run a process on it. Is there any way for it to notify me when it goes down?
The engine seems to go down after a job fails on it. As long as jobs are successful it is ok.
Hi @swach
How did you add the engine to your FME Server install, if it's not showing in the web ui?
Is the other engine the same build as your FME Server?
Have you looked in the engineprocessmonitor and the engine log for that engine? You can find log files under Resources in the web ui.
Hi @swach
How did you add the engine to your FME Server install, if it's not showing in the web ui?
Is the other engine the same build as your FME Server?
Have you looked in the engineprocessmonitor and the engine log for that engine? You can find log files under Resources in the web ui.
I have attached the files showing the web ui and the log. I wish I know more about how to decipher the logs.
EWPG-APP-1110_fmeprocessmonitorengine_2019-03-19_102745.txtEngines_Licensing.jpg
I have attached the files showing the web ui and the log. I wish I know more about how to decipher the logs.
EWPG-APP-1110_fmeprocessmonitorengine_2019-03-19_102745.txtEngines_Licensing.jpg
Thanks for attaching those @swach
Is your remote engine at 142.16.16.83?
I think you need to do step 2 here, which is editing the hba.conf file for postgres so that it can communicate with your extra engine: https://docs.safe.com/fme/2018.0/html/FME_Server_Documentation/Content/AdminGuide/Post-Config-Scripts.htm
Here's the FME Server troubleshooting guide for engines.
Looking at it, the shutting down might be related to scheduled rebooting after a certain number of jobs/features processed, but in that case it should come back automatically.
I have noticed the same thing on my test machine. We're running 2018.1.1 I have a sneaky suspicion it's related to invalid parameters (possibly a decrypt on a password that's not a valid hash) on a schedule or the python caller crashing that's crashing the engine. I've since stopped all my schedules on my test server except one schedule and the engine never crashes. I'm slowly going to enable my schedules and see which one crashes the engine. Like you, the logs were not useful to understand why the engine was not restarting.
What is this patch they are talking about in this post? https://knowledge.safe.com/questions/50783/fme-20171-engines-shuts-downstops-unexpectedly.html
I have noticed the same thing on my test machine. We're running 2018.1.1 I have a sneaky suspicion it's related to invalid parameters (possibly a decrypt on a password that's not a valid hash) on a schedule or the python caller crashing that's crashing the engine. I've since stopped all my schedules on my test server except one schedule and the engine never crashes. I'm slowly going to enable my schedules and see which one crashes the engine. Like you, the logs were not useful to understand why the engine was not restarting.
What is this patch they are talking about in this post? https://knowledge.safe.com/questions/50783/fme-20171-engines-shuts-downstops-unexpectedly.html
Hi @justincornell
The patch referenced in that article was only related to FME Server 2017.1, so what you're experiencing will likely be different. Based on the steps you've taken so far, I think you're on the right lines to determining which job(s) are causing your engine to shut down. Once you find that, it would be good to look at the engine process monitor log which should contain details about why the engine is shutting down.
Thanks for attaching those @swach
Is your remote engine at 142.16.16.83?
I think you need to do step 2 here, which is editing the hba.conf file for postgres so that it can communicate with your extra engine: https://docs.safe.com/fme/2018.0/html/FME_Server_Documentation/Content/AdminGuide/Post-Config-Scripts.htm
Great! That seemed to fix it. One additional question though, the .bat file indicated in step 3 of that process doesn't exist on my server. What does it do?
In this situation the engine had been installed and configured but was not able to communicate with the FME Server database, as shown by a warning in the fmeprocessmonitorengine log file:
FATAL: no pg_hba.conf entry for host
Editing the hba.conf file to allow connections from the remote engine machine is necessary to allow the engine to communicate with the FME Server database so that it can receive and process jobs.
Step 2 in this documentation outlines how to do this.
Great! That seemed to fix it. One additional question though, the .bat file indicated in step 3 of that process doesn't exist on my server. What does it do?
It initialises FME by putting the right things in the database. More recently they've been running automatically on install so there shouldn't be a need to manually run them any more.