Lawson is just slow

Yes, we have been hearing this over the last couple days and it is intermittent to make it worse. Difficult issue to resolve. We have checked db server for large queries, checked network traffic several times, no errors in any Lawson system logs. We have restarted Lawson several times, rebooted the AIX server. Run fiddler several times to try and capture this happening. No updates, patches, changes of any kind to the system.

We are seeing Portal hang then continue as usual. Anyone dealt with this issue and come up with a good solution or something we can try and chase?

dEPM-674_Series_Part 8_Right x Quantity.mp4

Find more posts tagged with

Infor Lawson Technology Group - Discussion

Comments

bob-heitzman

When this happened to us we traced it back to a recent product installed that was doing auditing of changes on various applications - Lawson being one of them. It caused slowness on Lawson but not other systems it was monitoring for changes. You didn't mention if you had a chance to review all known changes made on your system and network prior to the issue. If you haven't had a chance to do that fully the issue may have started with a change somewhere.

0909090854053232.pdf

jreese

When we have experienced this we find that one of the clustered app servers has hung or stopped and have to restart.

todd-brown

Yes we have seen the hung WAS app server before. This is not that unfortunately. We have bounced WAS several times. Anyone added another WAS app server to the cluster to fix an issue like this?

unknown

A while back I ran into an issue that ultimately came down to the NIC Card settings on the DB Server. Adjust those, packet size increase, and performance was much better. Not sure this is helpful for you - but worth noting. I have documentation on the fix.

jreese

We have 3 app servers and we will get calls from certain users complaining about slowness while others say they are fine and that usually tells us we have an issue with one of the app servers.

unknown

I am interested in reading your documentation if you are willing to share.
We have been seeing the same sporadic slowdown.

test.csv.zip

todd-brown

Me too. todd.brown@hhsys.org

todd-brown

Seems to be more of an issue with pulling data rather that loading screens.

unknown

Sorry I should have included my email address lbrant@rhsnet.org.

unknown

OK. I'll send it out when I find it! I can locate my RCA Document since changing jobs, and dropbox is acting up!

gwhiteford

If this issue is more with pulling data, do you have a separate web port for Lawson Addins/Queries?

unknown

I would like to read your documentation on what you did to fix also. My email address is judith.washburn@roswellpark.org. Thanks..

0712131256598360.doc

gwhiteford

Please send me documentation as well: Barbara_caporella@mednax.com

tbyczkowski

Please send to me as well, amy.cordell@tmcmed.org. We have the same issue as Todd and it has been like looking for a needle in the haystack.

todd-brown

Oh its like chasing a ghost. You have the same issue going on now?

tbyczkowski

Yes, I have had our server team and dba looking into why. Portal will freeze up for a few minutes and be non responsive then in about 5 or 10 mins be fine again. It happens intermittently at different times and not every day (at least, not reported every day). One day I did capture an error message on Portal "The Portal cannot load because the Profile service returned an error: INCIDENT_ID#1435342221515-004 An error occurred outside of IOS while accessing Lawson Database : File GEN/USER_OpenDBFile2 error is Unable to communicate with database driver process (99).Additional information is 1182." I found KB article 1383212 which said this was an issue related to latm.exe causing conhost.exe to crash applicable to Windows 2008 R2 platform only but I didn't find any errors related to this in our Event Viewer logs. Are you running Windows 2008 R2?

todd-brown

AIX. Our issue is similar however our delays are not quite that long. Maybe 60 sec, then continues on normal

todd-brown

Oh, we are on seeing any error messages at all.

teena-chopra

Please send me documentation as well, krgoodgion@kennedy-center.org. We too experience intermittent unidentified slowness. Thank you!

unknown

We are on AIX please copy me at kathy.r.williams@bannerhealth.com

unknown

Recently, our System hit a limit. Over time the number of users have increased and we felt inermittenly and then specifically when Payroll ran a large job.

Searching the logs we realize we needed to increase max memory for java. This is a setting within WebSphere.

Servers -> Server Types -> Websphere Application Servers.

Click on server1 (or your LSF server's name).

under Server Infrastructure (subgroup on right side), expand Java and Process Management.
Click Process Definition.

On the right, Click Java Virtual Machine.

We increased the Maximum Heap Size from 1024MB to 2048 MB.

We need to add new ap servers specific to Add-ins and LSO. But, this helped immediately.

unknown

Please add me to your document distribution list also. Thank you

tom.dewey@dal.frb.org

todd-brown

Our JVM max heap sizes are at 1024. Maybe we should up thoses. What ramifications are there with increasing that, and do you have to recycle WAS?

todd-brown

Did anyone get that db server NIC setting documentation?

unknown

Not me, I did request it but haven't seen it.

unknown

When we mae the change, we took the opportunity to restart everything. I believe you can just restart the app server. As always, do this in test first. We can't recreate the load that production has, but it will indicate if there is an immediate problem. I used topas (on AIX) to monitor the processes and their cpu and memory usage.

Before, java would reach 90%+ cpu and not recover. Now, bounces between 2% - 21%. Spikes happen, but it quickly drops.

unknown

DerekL: - Can you tell us exactly what version of Java you are using? The systems that I currently administer are Windows, not AIX. So all I can add from prior experience, when I was working with AIX systems, is that I have noticed that certain versions of Java work better than others. Some versions of Java developed a memory leak, filled up CPU, and brought the system to its knees. Others worked great.

lbecker

Tools can help. Try running the IBM Java Heap Analyzer. Requires X-Windows on Unix, at least for me.

www.ibm.com/.../0811_gunasekaran.html

The download for the tool is near the bottom.

Long ago, we had something similar. What we found (highly suspected, never truly found) was a badly written Excel Add-in would cause Java to use quickly 100% of its memory, then halt as it tried to collect garbage, only to reclaim 2%, and then immediately try to collect garbage again. By "badly written", I mean over 100 conditions in the selection criteria. Nobody would admit to this.

An environment patch fixed the problem of bad Excel Add-ins.

bhillz

We have a similar situation with the exception that we have sufficient CPU, memory and IO (including database) capacity. Some portal (principally RQC) users will report the system is sporadically slow. We have been able to, in most cases, trace it back to a database contention between an batch job (RQ500, WH130, etc.) and the user activity. In a few instances there appear to be no relation to the batch activity.

We would like to determine if this is user vs user contention.

Has anyone determined a method to track portal activity per user per screen/form, including start/stop times. We would need this to be across multiple JVMs.

unknown

WebSphere 7.0.0.29

java version "1.6.0"
Java(TM) SE Runtime Environment (build pap6460sr8-20100409_01(SR8))
IBM J9 VM (build 2.4, JRE 1.6.0 IBM J9 2.4 AIX ppc64-64 jvmap6460sr8-20100401_55940 (JIT enabled, AOT enabled)
J9VM - 20100401_055940
JIT - r9_20100401_15339
GC - 20100308_AA)
JCL - 20100408_01

Copyright © 2025 Infor. All rights reserved.