Intermittent Security Violation errors on screens

We have been struggling with a couple of our payroll users (restricted in security by process level) getting security violation messages when they try to update their Payroll screens (PR39, PR36, PR14). These are screens they use all the time but are intermittently getting security violation messages that at this time only seem to disappear for a period of a day or two by restarting the LSF server. We are also seeing connection failure errors in our ladb.log not just for this user but other users that do not report any issues.

Has anyone else had this issue and what did you do to resolve it? Thank you in advance!

Find more posts tagged with

Comments

Legacy Contributor

We have been having same exact issues, but no resolution yet. Appears something is causing a timeout somewhere, and have been focusing on the app server to DB server tracing. Have even had security violations pop up when Lawson job is running files out to user Print folders, violations reflected in job log, which makes no sense in general.

deanna-plant

Just curious...are you using VMWare? And is LSF & database on separate servers?

dEPM-653 - Introduction to dEPM Office integration for Excel.mp4

lmerrill

We had the same issues for more than a year. We opened several tickets with support and they couldn't figure out what it is. They told us it is our network issue, and our network guys, say, there is no network issue. It was more happening on our test environment. So, finally, we moved out test environment to our production Veam(internal cloud) a couple of months back and knock on wood, so far, it is stable. It doesn't happen all the time. Anytime, there is a Veam issue, test goes down with these symptoms and we have to restart all the servers, and it would only last for a few hours and again, we get these messages. And suddenly, one day, it stays stable and continues to be stable for a few weeks. We used to see 'socket closed' error in ladb.log.

0911060553270289.pdf

Legacy Contributor

I have had similar issues - not exactly the same, because we don't run Lawson PR module - but here is my advice: If it's network-related, do not, I repeat, DO NOT take the network guys' word that it isn't their fault. If t hey want to prove that it's not their fault, they HAVE TO prove it by installing and monitoring a continuous trace on network traffic. I've seen "quick" 5 - to - 15 minute network pauses. They happen a lot. Often this is when network relay equipment (switches, routers, etc) are rebooted, or patches are applied. The network picks up afterward without a hitch. But in the meantime, my applications (not just Lawson!) encounter timeouts at various levels of the network stack, and, boom ! Users are disconnected. Guess what! The application, and thus _I_, get blamed. I have been fighting this battle for years. I can tell you right now, the people responsible for network maintenance do not like being called out ... but, they usually have no way of continuously monitoring network flow. Or, it's "too expensive", or, they "don't have the time". When I encounter this kind of excuse, I ask, Well, is there anywhere in the network where timeout values can be increased? Often, there are. Another thing -- I haven't pursued this in Lawson in a while, but found this in other applications -- there are application timeout values that can be increased. Adjusting these has helped a lot.

clara-tan

We found the issue with "connection failure errors in our ladb.log" to be related to the number of database processes allowed by the database. we increased that number and the problem for connection went away - its more than connecting for account but each query to retrie data would be a connection. the database releases connections after they are done but if there's just too many that error can show up. Not sure for security since it's intermittent, i woudl have assume it was a patch that added a new file/field in the background that the security admin tool needs to now grant access to new underlying tables even though accessing the same user screen. a security type report would seem a good route to capture more

FordUS_ship_req.pdf

lmerrill

Good advices Milo and JBoncha(sorry, don't know your actual name). I will definitely keep these in mind, if this ever happens again.

Legacy Contributor

JBONCHA - that is also a very relevant point.

Legacy Contributor

For over one week we have been getting the following error on our test Smart Office environment every time any user attempts to open any screen: "Error retrieving form definition for token: AP90.1 Please check the log file for details."

So far, Infor Support's advice is to upgrade Smart Office.

Legacy Contributor

[Your question deserves its own thread.] I am not arguing with Infor Support, but, I am suggesting that the "Error retrieving" be addressed as well. Have you tried re-compiling AP90?

Smart Office logging (if I remember correctly) requires an individual user to turn on logging on their canvas. Then, after the error occurs, you are supposed to export or download the log file. Check the log file & see what it says. That will help to narrow down the issue.

plancor

jboncha, when you say "number of database processes", do you mean connections? If you are using SQL, where is the location of the number you increased?

deanna-plant

I am interested in this as well. We have out SQL database "Max number of connection = Unlimited" already and getting the errors.

Copyright © 2025 Infor. All rights reserved.