Posted on 02-09-2018 04:39 AM
I have a question for anyone else running their JSS on a windows server.
Do you have it set to periodically reboot? Do you have any issues when it has 'been up too long'?
I have two instances of our JSS, one to serve inside the network and one that serves outside.
We get reports from time to time that people can't use this service or that... something funky. We reboot the server and all is OK.
I am thinking that periodic reboots might be good but was wondering what everyone else is doing.
Do you experience this same issue? Did you find it was something in your config that causes it to 'sour'?
Posted on 02-09-2018 04:50 AM
Yes, I'm running my JSS on Windows 2012 R2. I'm not experiencing this issue. With that said, our servers are patched and rebooted on a monthly schedule. I have a Master with two web app servers.
Posted on 02-09-2018 06:04 AM
We have three JSS running in our environment. One internal and two DMZ. The internal one which is the primary one usually stops working once in a while. I've been tweaking the Tomcat settings but unable to find the right match for it. In our case, tomcat just crashes and service is stopped.
To mitigate this issue, I have a scheduled batch file running around 5 AM to restart Tomcat service. This seems to mitigate the issue.
Our servers are patched and rebooted on a monthly basis but the issue still happens.
Posted on 02-09-2018 06:27 AM
I ran a JSS on 2012 R2 for two years, never had to reboot it because of an issue. It was rebooted monthly after patching, and taken down for JSS upgrades, but otherwise ran great. MySQL was on a separate RedHat box.
Posted on 02-09-2018 06:43 AM
No issues here. Ours probably gets rebooted about every 4-6 weeks, depending on what patches are needed for Windows itself. No complaints.
Posted on 02-09-2018 06:58 AM
We are in the process of moving from 2008R2 to 2016. We have been on windows since we started 4 years ago. I can't really say that any of the issues we have had were directly related to the server OS. Although we do generally patch and reboot monthly. Most issues revolved around Tomcat getting overwhelmed and clients not being able to communicate with the webapps. We have implemented two memcached servers and our problems seem to have been resolved.
Posted on 02-09-2018 07:02 AM
I was running a single 2012 R2 server in AWS at my previous job, with infrequent restarts/updates. I did have some issues with reliability/accessibility, and it was related to memory/storage. Make sure that Tomcat has enough memory and that the HD is not running out of space.
Yes it's a simple thing, but it was once extremely compounded by the fact that some large package upload/sync once got stuck in the DB as it did not complete. This meant that every backup of the DB went from being MBs to GBs and ate up the entire HD, unknown to me until the JSS was unavailable. Had to remove all the bloated backups and fix the DB to get rid of that incomplete package.
So... worth checking the simple things to make sure restarts aren't just flushing RAM and caches and making things run smooth for a little while.
Posted on 02-09-2018 09:08 AM
Reboots shouldn’t be necessary but they help us. They get performed during server patching typically
Posted on 02-09-2018 11:54 AM
Those of you running memcached servers, are you running the memcached service on the same server as the JSS or do you have it on the separate server?
Posted on 02-10-2018 01:02 AM
Also running 2012R2 for our JSS's and DPs - they get rebooted once a month for patching and that's pretty much it. Don't really have issues with them, all good!
Posted on 02-12-2018 05:02 AM
memcached on two separate boxes.