How often do people reboot their serves

Terry,
Due to memory leaks, typically in the lase process, I tell clients to bounce LSF once a week, and the server about once a month to once every two months.

Now with that said, if you can identify a patch release and build number that's extremely stable, I have seen Windows systems run for over a year, without so much as a hiccup, but this is the exception and not the rule.

A general rule is the more frequently you restart LSF and clean up processes related to it, the less frequently a full OS restart is needed

Hi Kwane, thanks for the info, I kind of figured it was leading down that path unfortunately.
When you say LSF, you are referring to the lawson env, or the node agent/app1, app2 etc or both?
Thanks
Terry

LSF refers to the entire stack, as it's best practice to restart WebSphere when restarting LSF Core processes.

We reboot LSF LMK and DB every Monday early am. It takes approx 30 mins mainly due to websphere coming down and back up and the need to have them reboot in order. This was originally suggest by our Infor consultant on v9 . Now on V10 we have had multiple instances where issues arise during a week of processing that get cleared up with a clean reboot.

AlanK,
There should NEVER be a need to reboot the DB server. In fact, you're hurting performance doing restarts of the DB regularly

We reboot our LSF and Landmark servers every two weeks.

JimY,
I'm surprised you're getting that much mileage. With the quantity of patches that have been released as of late for the LSF stack, the quality of those patches has dropped a fair bit.

I'm having a number of my clients needing to restart LSF (and in some cases LMRK) once a week, as the system starts to fail in odd ways about 10-12 days after the prior restart. More strangely, this is happening on flavors of *NIX (which has historically been stable) as well as Windows.

As I posted above, I'm telling people restart weekly, until the patch releases begin to slow down.

Hi Terry, We're on AIX and rarely reboot the servers.  The only times are when we're doing DR or failover testing.  We've been on AIX for several years now.  When we moved to v10, it seems like things became even more stable.  I've been incredibly impressed!  Landmark purrs along without issues.  So that's the situation regarding *server* restarts.  Actual Lawson/Landmark environment restarts (not the server) happen pretty much only when we do patching, especially environment, since that requires app restarts.  I would encourage everyone to move to AIX (ha ha!!).  Let me know if you have questions, Lenny

Sometimes we do have problems with it slowing down between the restarts. Usually due to a run away java process in WebSphere. We decided that 2 weeks was enough and could live with the occasional restarts in between. We were on Unix prior to upgrading to V10 and also rarely rebooted the server. I wish we had stuck with Unix, but went with Windows due to cost. We are currently in the process of upgrading from Windows 2008 to 2012 along with our applications. Not sure if that will make a difference in speed though. So far in testing we haven't seen much of a difference.

@Kwane McNeal
Yes my mistake - DB is not included in the weekly reboot.