Special Weapons!

High processor usage on Websphere

MJLee ‎ 2,017 Views

If you run into a situation where the processor utilization is high - the process for determining the cause isn't the easiest thing to find on your typical Google search. So here it is:

For the sake of brevity I'm just going to do Linux

1) Take a javacore. This is easy on a Unix-type system. Just find your process and run "kill -3 <PID>"

2) Collect some top data

# top -bH -d 5 > top.out

3) Find the threads that are using the most processor usage in the top output

12771 user 20 0 6028m 2.3g 59m R 19.6 7.4 48:33.59 java
11880 huser 20 0 6028m 2.3g 59m R 19.6 7.4 46:41.38 java

4) Open up calculate and convert the PID to a hexadecimal value

For example 12771 -> 31E3

5) Open up the javacore and find your thread via the native thread ID.

3XMTHREADINFO      "WebContainer : 46" J9VMThread:0x00000000C4506C00, j9thread_t:0x00007FA9F66D2E10, java/lang/Thread:0x00000000091B2350, state:CW, prio=5
3XMTHREADINFO1            (native thread ID:0x31E3, native priority:0x5, native policy:UNKNOWN)
3XMTHREADINFO2            (native stack address range from:0x00007FA99EFE0000, to:0x00007FA99F021000, size:0x41000)
3XMTHREADINFO3           Java callstack:
4XESTACKTRACE                at java/awt/image/DataBufferInt.<init>(DataBufferInt.java:52(Compiled Code))
4XESTACKTRACE                at java/awt/image/Raster.createPackedRaster(Raster.java:469(Compiled Code))
4XESTACKTRACE                at java/awt/image/DirectColorModel.createCompatibleWritableRaster(DirectColorModel.java:1026(Compiled Code))
4XESTACKTRACE                at java/awt/image/BufferedImage.<init>(BufferedImage.java:323(Compiled Code))

You want to do this on different datasets so that you can confirm the pattern behavior is the same. So essentially just repeat the process

And that's it. Now that you have found the culprit, all you have to do is determine why that thread path is using up too much cpu.

Apple iPhone users: Are your texts being ignored by Android users?

CaseyToole ‎ 1,794 Views

Apple iPhone owners please read.

Do you have friends who previously had iPhones and switched to Android, who now ignore all of your texts?

Well maybe you aren't being ignored. Apple has a setting that will not send messages to Android numbers if the connections were previously established via iMessage.

To fix this:

Go to > Settings > Messages > and turn on the feature "Send as SMS" (Send as SMS when iMessage is unavailable.).

If this is working properly, your SMS texts to Android users should appear green. iMessage messages are blue.

The difference between following and being in network

MJLee Tags: connections ‎ 9,266 Views

in regards to notifications, the functionality of being in network and following has changed dramatically between Connections 3 and Connections 4. in Connections 3, if someone was in your network, you would be notified of things that person was doing in other applications. So you didn't need to follow that person or that person's blog in order to be notified of activity that the person generated.

But in Connections 4, the only way to get notified of a person's activity in another application (like Files, Forums, Activities, etc) is to follow them or the specific resource in the application. This may throw a lot of people off since it's not highly documented.

In the end, the functionality will be fixed and eventually the users will become accustomed to the new style of notifications. In the meantime, here is some info on what each of the states will do for you.

My network notifications in C4 were greatly reduced to ease the congestion in the activity stream.   It will focus on
    Status updates,
    Microblogging.
    Related comments,
    Reposting,
    Liking
    and possibly other information that could be associated with network activity.

My Notifications

Directed notifications from other users
Responses to end-user content across Connections
Any events explicitly targeted to the end-user by an IBM service

I'm Following

Despite its label in the UI ("I'm Following"), it displays more than only events about content the end-user is following. Events are surfaced as follows:

Any event surfaced under “My Notifications” (target events, such as responses to the end-users content)
Any event surfaced under “Action Required”
Any event from content being followed. Specifically in other applications

Checking out what's going on in the SystemOut.logs

MJLee Tags: websphere connections ‎ 3,492 Views

If you're in a Unix environment and hate having to check SystemOut.logs in multiple locations- Did you know you can get around that with a little scripting and taking advantage of ssh. The way it works it you use a private key to allow your script to authenticate and run a script on the node. This way you can just run a script to tail or grep for what you want to see and hit all the servers from a single machine.

I had to write this as part of a monitoring subsystem for another customer. Though this would be handy for almost anything repetitive in a multinode environment. So here is the snippet.

userHost=( user@host05 user2@host01 )

exitCode=0

# Location of the News SysOut logs.
newsSysOutDir=/local/opt/IBM/WebSphere/AppServer/profiles/AppSrv01/logs/clusterLC_server1

# Name of the monitor script
mon_script=SO_Mon.sh

# Test SSH access
for var in "${userHost[@]}"
do
echo "Directory listing for $var"
ssh -oBatchMode=yes -n $var "ls -l $newsSysOutDir"

    if [ $? != 0 ]; then
    exitCode= $?
    echo "SSH access is not configured"
       ssh-keygen -t rsa
    cat ~/.ssh/id_rsa.pub | ssh $var "chmod g-w ~/; mkdir -p .ssh; chmod 700 .ssh; cat - >> .ssh/authorized_keys2"

    echo "Configured SSH private key for $var. You will need to try again"
    fi

done
if [ $exitCode != 0 ]; then
    exit $exitCode
fi

# Upload the monitor scripts
if [ -z $isScheduledRun];then

echo "Uploading the monitor script"

for var in "${userHost[@]}"
do
scp $runDir/$mon_script $var:$newsSysOutDir/$mon_script
if [ $? != 0 ]; then
   echo "Upload of monitor script failed"
   exit 1
fi

done
fi

# Run the SystemOut.log monitor
for var in "${userHost[@]}"
do
# This is configured to work with sendEmail
ssh -oBatchMode=yes -n $var "$newsSysOutDir/$mon_script sendEmail wtfmail02.lotus.com michael_lee@us.ibm.com $newsSysOutDir "
done
echo "Kicked off the monitors"

# You can obviously run something here now to clean up the script .

Stats for Connections and Portal

MJLee Tags: connections statistics portal ‎ 1 Comment ‎ 5,912 Views

Me and a colleague from the Portal side put together a list of stats and thresholds that we thought every implementation needs to monitor of every Portal/Connections environment: This isn't a hard and fast list of course. It's a nice starting point though.:

PMI statistics:
JDBC
PoolSize ( > 80% Max - i.e. 80 on Connections servers)
UseTime (> 5 second)
WaitTime - ( > 2 second)
WaitingThreadCount - (> 3)

Connections Only - JMS (JCA)
Wait Time ( > 10 seconds)

Messaging Engine (SIB)/Queues
Queues.AvailableMessageCount (> 500)
Queues.LocalMessageWaitTime ( >600 seconds)

JVM
HeapSize ( just monitor)
UsedMemory ( >80% for a significant time -especially increases over time)
ProcessCpuUsage (> 60%)
% Free after GC, (< 30% of max heap)
% Time spent in GC (> 10%)
Uptime ( Ensure process is up)

HTTP (Servlet) Session Manager [For WPS EAR only - 004,005]
ActiveCount (Monitor to evaluate what the expected concurrent usage is)

Threadpool
ActiveCount ( > 80%)
PoolSize ( Connections only > 90%)
DeclaredThreaHungCount (> 1)

Web Applications
ActiveCount (Just monitor)
PoolSize ( Connections only > 90%)
DeclaredThreaHungCount ( > 1)

Web Applications
Web Applications.ConcurrentRequests (Monitor to evaluate what the expected concurrent usage is)
Web Applications.ServiceTime ( > 10 seconds for Portal ; > 2 seconds for Connections)

Portlet Application- All portal server
Number of concurrent portlet requests (Monitor to evaluate what the expected concurrent usage is)
Response time of portlet render (>10 seconds)
Response time of portlet action (>10 seconds)

Object Cache
LruInvalidationCount (> 100)
MissCount (Monitor to evaluate what the expected concurrent usage is)

Dynamic Caching
LruInvalidationCount (>100)
MissCount (>100)

Data Store
JDBCTransactionTime (> 5 seconds)

Security Authentication
WebAuthenticationTime ( > 5 seconds)

Security Authorization
WebAuthorizationTime (Connections only > 5 seconds)

Log File Health
Known Application errors that are critical: Development team to provide
Portal: Syndication errors from authoring to production
Hung threads
Large amount of exceptions

DB's

DB2 execution times above 1 second on average or long running transactions
Free pool for Database connections below 10% free on WAS/Portal
Monitor db2diag for ConnectionTimeOut and LockTimeout errors

TDS
LDAP response time > 3 seconds

TDS - DB
DB2 threads for Queries (execution times) above 1 second on average
Free pool for Database connections below 10% free on WAS/Portal
ConnectionTimeOut and Lock Timeout Exceptions in the logs

OS level stats

OS level - each of the above hosts should have

CPU	Memory	Disk
Total % CPU (> 75% for extended time) User %CPU (> 75% for extended time) Wait %CPU (Monitor for problems) System %CPU (> 75% for extended time) Avg. CPU RunQueue ( > 10 per CPU) CPU per processor (> 75% for extended time) CPU usage per process (> 75% for extended time)	Swap Space Used ( > 1GB ) Page In (Monitor - should be close to 0) Page Out (Monitor - shoud be close to 0)	Disk Service Time Per Disk ( > 100 ms for extended time) Avg. Disk Queue Length ( > 10) %Busy ( > 50% for extended time) Percent Free (<10% Free)

An LDAP service monitor
For example

http://cnmonitor.sourceforge.net/

IHS
Response Times
Busy Threads
Log File monitoring for "marked down"

IHS Plugin
Monitor log file for "Failed to connect to app server"

Loadbalancers
Any end user type of monitoring tool. I know of at least one customer that uses Business Process Monitor.

Lotus Mobile Connect is Traveler Aware - use it for automatic assignment of users to HA Pools

CaseyToole Tags: enterprise-mobility mobility traveler lmc traveler-ha ‎ 1 Comment ‎ 5,689 Views

Today's post is about Lotus Mobile Connect and how it can be used with Traveler to make sure that mobile clients can reach the right Traveler HA Pool or Traveler server, in a secure manner with a clientless VPN connection.

First of all, many of you don't even know what LMC is. What is Lotus Mobile Connect? In a nutshell...

VPN solution supports all access points

● Laptop and Desktop (client)

● Handheld mobile device (client)

● Browser (clientless HTTP Access)

• Proven ROI - least cost routing & data optimization

● Network optimizations reduces network data costs

• Secure end-to-end connect to corporate intranets

● Strong authentication and support for RSA Secure ID

● End-to-end encryption: DES, Triple DES, RC5, AES and FIPS 140-2 certified

Clientless: No client software installation, application and platform agnostic!

Single Sign On: LMC Gateway + application authentication

Enhanced Security

– Same strong authentication and encryption algorithms as the LMC Mobility Client

– FIPS 140-2 certified platform. Contains latest secure SSL/TLS ciphers and industry standard authentication mechanisms

– Two way certificate validation available for additional layer of trust

– Configurable to run simultaneously with full VPN sessions. Multifunction, IT configurable remote-access solution (Example: Subset of users restricted to specific apps via clientless access only. Other set of users allowed expanded enterprise access with LMC Mobility Client.)

So what does it mean to be "Traveler aware"?

LMC 6.1.4 Traveler Aware Features

Recognition of Traveler URLs,
applies data flow optimizers and HTTP header tags,
previous password recognition (prevent LMC server lockout).

LMC 6.1.5 (NOW IN BETA!!) Traveler Aware Features

With the current set up of Traveler, a user has to know their URL to their Traveler server or Traveler HA (High Availability) Pool. When customers are deploying to large numbers of people, it can be a pain to have so many profiles out there pointing to different Traveler servers or pools. In the deployment that I just did, they are planning on 4 pools (I completed 2), each pool supports about 10,000 users. However, this customer did not want to have 4 different possible Traveler profiles on their Apple iOS Mobile Devices.

Enter the new LMC 6.1.5 Traveler aware feature...

The new LMC 6.1.5 Traveler aware feature allows LMC to go into a single URL mode. LMC "learns" which server or pool of servers a user belongs to, and makes sure that the user is routed properly. So in a High Availablility scenario, you still need a load balancer in front of the pool, but if you have more than one Pool, LMC will route the users to the right pool. It knows how many users belong to each server or pool of servers, and will divvy up new users to a less loaded pool. Pretty cool, huh? Look for LMC 6.1.5 some time in Q3!

Transaction timeouts are your friend

MJLee Tags: connections ‎ 2,707 Views

It's odd how some things come in waves. Recently I have been seeing a lot of issues at once where the database server that Lotus Connections uses can get itself stuck on a call. The next thing you know, the whole system is fried and people are calling in to complain. You can avoid a lot of pain by proactively tuning your system.

First off, the JDBC connection timeouts in WAS are set to 180 seconds. This is a little conservative but ultimately what's right for your users may differ between environments. Some user populations may be good with 15 seconds. Others may prefer to have 60+ seconds.

Secondly, you will want to trim the time the database will get stuck on a transaction. For most applications, it's set to unlimited. That's a problem waiting to happen. Each DB handles their systems differently.

For Oracle, you need to implement a user profile and apply it to the user. I would recommend dropping the CPU_PER_CALL value to something less than the JDBC connection timeout. This will let the application know that it's a database issue as the timeout will general a reason code.Though, since it's measured in CPU time and not real time, it can be pretty tricky to do that.

http://www.stanford.edu/dept/itss/docs/oracle/10g/server.101/b10759/statements_6010.htm

For DB2, you need to implement the LOCKTIMEOUT value. It's part of each database's configuration. Again, you want to make sure that the timeout is less than the JDBC connection timeout. This is a bit easier to configure since it's measured in real seconds.

http://pic.dhe.ibm.com/infocenter/wasinfo/v6r0/index.jsp?topic=%2Fcom.ibm.websphere.express.doc%2Finfo%2Fexp%2Fae%2Frprf_db2parameters.html

I know an environment that has had Connections for awhile and have a couple hundred thousand users. They found they lowering the locktimeouts to 45 seconds helped a lot.

A Lotus Connections general troubleshooting steps

MJLee ‎ 2,957 Views

IBM does a great job of putting together the documentation you need to support Lotus Connections problems. Though finding it all when you need it can sometimes be a challenge. Below is a list of the documents you can use to take that first step in diagnosing your issues and even using debug to take it one step further.

Test the reproducibility and try variations

Different browsers
Different users
Different services

Different locations

Follow the general troubleshooting in the docs below. If that fails, implement the trace strings for each service and open a PMR:

IBM MustGather: Lotus Connections Profiles 3.0 - United States

IBM Collecting Data: MustGather for Lotus Connections 3.0 Files ...

IBM MustGather: Lotus Connections Communities 3.0 - United States

IBM Collecting Data: MustGather for Lotus Connections 3.0.x Search ...

IBM MustGather: Lotus Connections Activities 3.0 - United States

IBM MustGather: Lotus Connections 3.0 Forums - United States

IBM MustGather: Lotus Connections Blogs 3.0 - United States

IBM MustGather: Lotus Connections Bookmarks 3.0 - United States

IBM Collecting data for Lotus Connections Wikis 3.0 - United States

MustGather: Performance, hang, or high CPU issues with WebSphere

Setting up the Connections accounts in Lotus Notes to use SPNEGO

MJLee Tags: notes connections lotus spnego ‎ 12,641 Views

Having a SPNEGO configuration with your Connections account in Lotus Notes can make using and administering Connections widgets so much easier. But right now, setting it up to use is not a simple proposition. Unfortunately I learned the hard way that there are some issues that make using this configuration difficult/impossbible to implement until some of these issues are fixed in the future code. That is unless you know some tricks to get around those problems.

Let's start out by going over how you would think to implement the configuration in Lotus Notes. If you are going to set up SPNEGO in your Connections account for an individual user you might go to your preferences and click on the Connections entry the list. Then you would check the "Enable IBM Connections Features" checkbox. Next you would enter the URL into the "Server URL" field. Then click on the "Advanced" button and choose the "OS Credential" option. That sounds like it should be pretty simple and someday that may be all you need to do. But for now let's go over some of the things that you will need to do to get this working.

The first thing to take care of is the krb5.ini. This is the Kerberos configuration file that Lotus Notes is going to need to initialize SPNEGO in your Windows environment. It will need to be placed in the < Lotus Notes>/framework/rcp/deploy/extras directory. Below is an example of how this file needs to be configured.

[libdefaults]
default_realm = <change to something like IBM.COM/Should be the AD Domain>
default_tkt_enctypes = des-cbc-md5 rc4-hmac
default_tgs_enctypes = des-cbc-md5 rc4-hmac
[realms]
SERVER.IBM.COM = {                                                   (This line should not be changed)
kdc = logonserver.ibm.com:88                                (change to the value of your logonserver.   You can get this by typing echo %logonserver%)
admin_server= logonserver.ibm.com:88                  (change to the value of your logonserver.   You can get this by typing echo %logonserver%)
default_domain = <change to something like ibm.com>
}
[domain_realm]
.ibm.com = <change to the realm value to be used. ie. IBM.COM>
ibm.com = <change to the realm value to be used. ie. IBM.COM>

A couple gotchyas that are relatively minor are that when you set up your Connections preferences you need to specify "https://" address as the server the server URL. And you will want to use the "https://<hostname>/activities/service/authredirect.jsp" for the Authentication URL field. (NOTE: Greenhouse doesn't support SPNEGO. I'm just using this as an example)

After making these changes you may still find that SPNEGO doesn't work as you would expect. Well one reason may be that you are an administrator on a Windows 7 PC. This is a quirk of the UAC in Windows. In order to get SPNEGO to work you can right-click the Notes icon when you start and choose "Run as Administrator".

This should work to get your Activities widget working. But many times you'll want to do something like use another widget. For example the "Status Updates" widget is very popular. But if you're using SPNEGO you'll notice that the thumbnail pictures of users may not be coming down. this is a known issue that you can address by adding a line to your plugin_customization.ini file. This is located in your <Lotus Notes>/framework/rcp/ directory. This changes the behavior of the plug-in to pull the images via a feed mechanism.

com.ibm.lconn.statusupdates/download.image.enabled = true

PUSHING IT OUT

So after making all those changes you should now have a working SPNEGO enabled Connections account on your Lotus Notes client. What you usually end up wanting to do from this point forward is to push these changes out to a large group of users. The way to do that is through Policies. If you go to your Name and Address book go to your Policies folder and click on the Accounts view. Click on the Add Account button and start filling out the form. The key fields will be the "Account type" which should be Connections; the "Connections service URL"; and the "Authentication URL" on the Advanced tab. Also on the Advanced tab, you need to choose the Authentication type and set that as OS-CRED.

The most important trick to pushing this policy out is to place the krb5.ini in the Account file field. This will place that file in the rcp/deploy/extras folder. Note: You can only place one file in this field. So this will be problematic if you need to create a custom icon.

For the issues associated with the UAC causing authentication failures when using SPNEGO, there are a couple options. One you can create a VBS script that will starts the notes program with elevated privileges. Or you can modify the shortcut. Right-click on the icon and select properties. You will see a Compatibility tab and on that you can choose to check the "Run this program as an administrator" option. Both of these options have the negative side effect of creating a pop-up that alerts you to the fact the program will need to run with elevated privileges. If you want to bypass that's pop-up, you can create a scheduled task that will run with elevated privileges and then start that task via a shortcut.

Note: You need to be an administrator for these. Regular users will not need to implement these options to get SPNEGO to work.

Dim oShell,oExec
If WScript.Arguments.Named.Exists("elevated") = False Then
   CreateObject("Shell.Application").ShellExecute "wscript.exe", """" & WScript.ScriptFullName & """ /elevated", "", "runas", 1
   WScript.Quit
Else
   Set oShell = CreateObject("WScript.Shell")
    oShell.CurrentDirectory = CreateObject("Scripting.FileSystemObject").GetParentFolderName(WScript.ScriptFullName)
End If
Set oExec = oShell.Exec("notes.exe")

Start tasksch.msc
In the Left hand pane, go to the Task Scheduler Library
In the Right hand pane, click Create Task

General Tab: 1) Give the task a name; 2) check Run with highest privileges; 3) Configure for Windows 7
Actions Tab: 1) New - Start a program, 2) Start in <Notes data directory>

Conditions Tab: 1) Uncheck "Stop if the computer to battery power"; 2) Uncheck "Start the task only if the computer is on AC power

Settings Tab: Uncheck "Stop the task if it runs longer than"

The task you created will now show up in the middle pane.

Create a shortcut
Right click on the desktop and choose New -> Shortcut
For the location -> schtasks /run /tn <taskname>

click Next
Type what you want to call the shortcut and Finish

Social everywhere

MJLee Tags: connections ‎ 1 Comment ‎ 4,492 Views

It's surprising to me that people are not aware of how important social media is to business communication. I mentioned how important social business was to the Lotus brand to a bunch of email guys. Their jaws just dropped.

I recently heard a podcast where Connections was called the flagship product for Lotus. I'm not sure what my take on that statement is but it is definitely a strong indicator of sentiment.

The SWAT team is dedicated to embracing this. They have been taking on the challenge of doing their business in a social fashion which in itself is a huge undertaking. The question is invariably the same as it was 4 years ago. How do you communicate with people that aren't used to communicating that way previously?

Social media is a new paradigm now in the way that email used to be a new paradigm. Back in the early 90's when email was starting to really take hold, it was a requirement to directly communicate with people. The key word there being "direct". Using tools like Connections is a great way to indirectly communicate with a greater audience. The goal is to reach people that weren't or are not immediately available. Social is about leveraging those next tier relationships where you get value that you didn't know was there. I think what you are going to start to see is more and more social integration until we reach a tipping point and the paradigm shifts.

Right now, there are a bunch of different players in this space and I believe it's still evolving. A fairly recent Gartner report shows Connections in a close race with Jive and Microsoft. But having seen the others, I have to think that Connections has hands down more to offer. IBM has the most impressive stack of social business applications and integration. When you are talking depth and breadth of a product, I'm still trying to wrap my head around how Connections isn't blowing everyone away. Perhaps that's just because I'm an IBM'er.

Cool tooling for moving users

MJLee Tags: adminp domino ‎ 1 Comment ‎ 4,889 Views

Our team gets called in to doing a wide range of tasks to help get through some pretty painful situations. As part of that, we have a pretty healthy production of tools to make it possible to do somethings that aren't currently available. Rob Mathews created an AdminP application that makes mass moving users much much simpler. Rather than trying to track down status and management features in different places, he put it all into one place. This is especially helpful if you have to move a great deal of users.

A lot of these tools are not available through normal channels. But if this is something that you think you might need, let me or your local SWAT team member know.

Remember your IO

MJLee Tags: domino performance sametime ‎ 3,550 Views

I was talking some recent SWAT activity with our local Sametime Guru - Casey Toole. We had a couple really painful issues that came up recently. Issues like a server going down, long service outages and tons of scrambling to get things back in order. Well once the dust settled, these problems all came back to being able to conduct high quality reads and writes to the filesystem.

What makes this an especially painful process is that in most shops, the application administrator (Domino, Sametime, Connections etc.) doesn't have the power to address disk subsystem issues. And the last thing you want to do is be stuck in a situation where people are screaming about service and you are having to prove the problem is with someone else. Let's say responsibility is never readily accepted 9 times out of 10,

So before you do anything, make sure that your disk subsystem is healthy . Check your response times and your error logs regularly. And don't forget a daily dose maintenance.

Multi-Database search

MJLee Tags: lotus domino notes ‎ 8,550 Views

One of the coolest things you can do with the Notes client is plug in functionality that someone else has written. If you are like me then your mail archive has way more documents than your mail database. And going from database to database is a pain.

To help fix this, Bob Balfe wrote a plugin that can search multiple databases for you. All you need to do is drag and drop the widget from the following page:

http://blog.balfes.net/2010/10/25/new-openntf-project-multiple-database-search/

After that it's just about clicking the databases you want to search and then running a search from the toolbar. The most amazing thing about all this is how few people know that it is available to them.

Our team

MJLee Tags: support swat lotus ‎ 3,224 Views

Our team started out as an area of support somewhere between L2 and development. We are particularly suited to issues that dealt with more complexity or required a more in depth approach to resolution. Though most customers don't even know who we are. I did a couple user group presentations and found out that SWAT sightings may seem like seeing a yeti. But that can be a good thing. It can mean that our front end teams and processes are doing well enough they don't need us.

But seeing things through our eyes might be handy. Our team generally gets involved when the stakes have been raised for one reason or another. So we get to see some interesting things. And sometimes shocking things.

We have people that cover Lotus Notes/Domino, Sametime, Quickr, Connections and Portal. Hopefully, this will be a good place to learn some interesting things about the products, support and what you can/should be doing in your environment.

Stay tuned.

Re: Automatic Application of Portal Tuning Parameters

jmadl Tags: tuning portal performance ‎ 5,600 Views

In response to: Automatic Application of Portal Tuning Parameters

Many Customer are asking what is changed by using this tool. I have performed some research on the tool and based on the Tool update from 03/19/14 here is what I have found related to the configuration changes that are performed by the tool

The Bulk of the tuning changes that will be performed for the JVM and web container service are found in the TuningTask directory in the archive in the file named tuning.properties. The first 5 lines are what the user sets for the environment they are configuring as follows
# - this is set to true if wcm rendering is being used on the site
tuning.wcm=true
# - set the following to true for a subscriber only :ie this does not syndicate content to other environments
tuning.wcm.subscriberOnly=true environments
#- this is set to false of running the command against a rendering environment and true if it is an authoring environment
tuning.wcm.authoring=false - only set this to true for authoring environments

# managed pages is a Portal V8+ feature that allows WCM to manage page #definitions. If you want the "edit" button on the V8 theme, you have to set #managedPages to "true"
tuning.wcm.managedPages.inlineEdit=true
#set the bit level of the jvm environment being configured
tuning.os.bit=64

#The other properties in this file sets the JVM Heap initial amd Max size as well as
# other generic jvm arguments depending on the OS being used per the tuning
# guide recommendations the values here are the values that will be configured in
# the Process definition for the Server being configured

# The properties from here to the end of file are recommendations from IBM on the
#initial tuning recommendations
tuning.64.linux.maxHeap=3584
tuning.64.zlinux.maxHeap=3584
tuning.64.zos.maxHeap=3584
tuning.64.windows.maxHeap=3584
tuning.64.aix.maxHeap=3584
tuning.64.unix.maxHeap=3584
tuning.64.os400.maxHeap=3584

tuning.32.linux.maxHeap=2048
tuning.32.windows.maxHeap=1408
tuning.32.aix.maxHeap=1792
## The next three OSes are not known to be used in 32 bit mode but are included
# here for completeness
tuning.32.unix.maxHeap=1792
tuning.32.zlinux.maxHeap=1792
tuning.32.zos.maxHeap=1792
tuning.32.os400.maxHeap=1792

# this adds - Xmn<value> "nersey initial and max heap size setting to the
# generic jvm arguments for the Process definition
# for the jvm being configured depending on whether its a 32 bit or 64 bit jvm
tuning.64.linux.nursery=1024M
tuning.64.zlinux.nursery=1024M
tuning.64.zos.nursery=1024M
tuning.64.windows.nursery=1024M
tuning.64.aix.nursery=1024M
tuning.64.unix.nursery=1024M
tuning.64.os400.nursery=1024M

tuning.32.linux.nursery=256M
tuning.32.unix.nursery=256M
tuning.32.aix.nursery=512M
tuning.32.zlinux.nursery=512M
tuning.32.zos.nursery=512M
tuning.32.windows.nursery=512M
tuning.32.os400.nursery=512M

## All these settings would be in the Generic JVM ArgumentsM
# sets -XX:MaxDirectMemorySize=256000000
tuning.all.DirectMemorySize=256000000
# sets --Xjit:iprofilerMemoryConsumptionLimit=67108864
tuning.all.MemoryConsumptionLimit=67108864
# sets -Dcom.ibm.websphere.alarmthreadmonitor.threshold.millis=
tuning.all.AlarmMonitor=40000
# sets -Xscmx150m note it must also be destroyed for this to take effect
tuning.all.SharedClassCache=150m
# sets -Xcodecache20m if running on power 7 hardware
tuning.power7.CodeCache=20m

#for Portal servers running with Oracle Java the following generic jvm arguments will be used
tuning.unix.Server=true
tuning.unix.MaxPermSize=-XX:MaxPermSize=768m
tuning.unix.UseConcMarkSweepGC=-XX+UseConcMarkSweepGC
tuning.unix.SurvivorRatio=-XX:SurvivorRatio=6
tuning.unix.UseParNewGC=-XX:+UseParNewGC
tuning.unix.ParallelGCThreads=-XX:ParallelGCThreads=5
tuning.unix.PrintGCDetails=-XX:+PrintGCDetails
tuning.unix.PrintGCTimeStamps=-XX:+PrintGCTimeStamps

#
## These sections set the min and max WebContainer thread pools to 50
#WebSphere Integrated Solutions Console
#Servers > Server Types > WebSphere application servers > WebSphere_Portal # > Additional Properties: Thread Pools > Web Container > Thread Pool
#minimum size = 50
#maximum size = 50
#inactivity timeout = 60000 this is in ms
tuning.all.webcontainer.min=50
tuning.all.webcontainer.max=50
tuning.all.webcontainer.timeout=60000

### This section update the datasources for JCR and Release. The name of the
#datasource is obtained from wkplc_dbdomain.properties
#WebSphere Integrated Solutions Console: Resources > JDBC Providers >
#provider name > Data Sources > data source name > Connection pool
#properties
#JCR datasource connection pool min/max set to 10/110
# Release datasource connection pool min/max set to 10/55
tuning.all.datasource.jcr.size.max=110
tuning.all.datasource.jcr.size.min=10
tuning.all.datasource.release.size.max=55
tuning.all.datasource.release.size.min=10

## Security > Global security > Custom properties > New
#name: com.ibm.CSI.disablePropagation.CallerList
#value: true
tuning.all.disablePropagationCallerList.name=com.ibm.CSI.disablePropagation.CallerList
tuning.all.disablePropagationCallerList.value=true

## Security > Global security > Custom properties > New
#name: com.ibm.ws.security.ltpa.useCRT
#value: true
#Note the tuning Guid rcommends this be removed
#the code appears to be adding this property
tuning.all.ltpaUseCRT.name=com.ibm.ws.security.ltpa.useCRT
tuning.all.ltpaUseCRT.value=true

## Disable the Internalization Service on the server
#In the WebSphere Integrated Solutions Console
#Servers > Server Types > WebSphere application servers > WebSphere_Portal
# > Container Services: Internationalization service
# Uncheck “Enable service at server startup”.
tuning.all.enableInternationalization=false

## Set the WAS session timeout
#Servers> Server Types> WebSphere application servers>
# WebSphere_Portal > Container Settings: Web Container Settings > Session
# Management > Session Timeout > Set Timeout
tuning.all.webcontainer.sessionManagement.sessionTimeout=30

## Avoid Refreshing Static Content which allows the modular theme to avoid
#reloads on login.
## These resources include the ra: collection URLs that are part of the theme. The
#same URL can safely be used for authenticated and unauthenticated users.
## The default is "lazy". "Persisting" provides better performance. "Persisting"
#means the checkbox "Use available authentication data when an unprotected
#URI is accessed" was checked.
# WebSphere Integrated Solutions Console: Security > Global security
#2. Expand Web and SIP security
#3. Click on General Settings
#4. Check 'Use available authentication data when an unprotected URI is
#accessed'
tuning.all.security.avoidRefreshStaticContent=persisting

## Set the WAS Enable Servlet Cache
#
#WebSphere Integrated Solutions Console Servers > Server Types >
# WebSphere application servers > WebSphere_Portal > Web Container
#Settings > Web container
#Check Enable servlet caching
tuning.all.webcontainer.enablecache.value=true
#WebSphere Integrated Solutions Console Servers > Server Types >
# WebSphere application servers > WebSphere_Portal > Portlet Container
#Settings > Portlet container
#Check Enable portlet fragment caching
## Set the Portet Container to enable portlet fragment caching
tuning.all.portletcontainer.enablecache.value=true

Resource Environment Provider Changes
WebSphere Integrated Solutions Console > Resources >Resource Environment
> Resource Environment Providers <ResourceEnvironmentProviderName>
The following are performed for Portal servers version 7 or version 8+
WP CacheManagerService

# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.ac.AccessControlUserContextCache.size=6000
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.ac.ChildResourcesCache.lifetime=28800
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.ac.CommonRolesCache.size=33000
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.ac.ExternalOIDCache.lifetime=-1
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.ac.OwnedResourcesCache.enabled=FALSE
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.ac.PermissionCollectionCache.lifetime=-1
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.ac.ProtectedResourceCache.lifetime=14400
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.datastore.services.Identification.SerializedOidString.cache.size=5000

########
# This cache uses more memory per object than other caches.
# The value should be set based on the number of concurrent, logged in users. Larger values may require a larger JVM heap size.
# A value of 6,000 was used to support 6,000 virtual users in the benchmark measurements.
# Increasing to 7,000 in other measurements increased throughput but required a heap size of 4,608 MB.#
#########

# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.model.factory.UserSpecificModelCache.size=6000

# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.pe.portletentity.lifetime=28800
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.pe.portletentity.size=45000
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.policy.services.PolicyCacheManager.lifetime=43200
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.puma.CommonPrincipalCache.size=30000
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.puma.DN_OID_Cache.size=30000
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.puma.OID_DN_Cache.size=5000
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.resolver.data.cache.DataSourceCache.size=8000
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.resolver.data.cache.FirstLevelDataSourceCache.size=2003
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.services.vpmapping.HostnameToVirtualPortalIDCache.lifetime=-1
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.spa.parser.locale.FirstLevelLocalizationParserCache.size=1009
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.spa.parser.skin.FirstLevelSkinParserCache.size=1009
# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.spa.parser.theme.FirstLevelThemeParserCache.size=2003
# Description = Portal/WCM initial tuning
cacheinstance.com.lotus.cs.services.UserEnvironment.size=4500
# Description = Portal/WCM initial tuning
cacheinstance.DigestCache.cache.size=45000

# Description = Portal/WCM initial tuning
cacheinstance.com.ibm.wps.resolver.resource.AbstractRequestDispatcherFactory.size=200

WP CommonComponentConfigService
# Description = Portal/WCM initial tuning
cc.multipart.enabled=false
# Description = Portal/WCM initial tuning
cc.multipart.correlatehosts=false

WP ConfigService
# Description = Portal/WCM initial tuning
cache.dynamic.content.spot=false
# Description = Portal/WCM initial tuning
resourceaggregation.cache.markup=true
# Description = Portal/WCM initial tuning
resourceaggregation.enableRuntimePortletCapabilitiesFilter=false

WP CPConfigService
# Description = Portal/WCM initial tuning
com.ibm.wps.cp.tagging.isTaggingEnabled=false
# Description = Portal/WCM initial tuning
com.ibm.wps.cp.rating.isRatingEnabled=false

WP NavigatorService
# Description = Portal/WCM initial tuning
public.expires=3600
# Description = Portal/WCM initial tuning
public.reload=3600
# Description = Portal/WCM initial tuning
remote.cache.expiration=28800

WP RegistryService
# Description = Portal/WCM initial tuning
default.interval=28800
# Description = Portal/WCM initial tuning
bucket.transformationapplication.interval=28800
# Description = Portal/WCM initial tuning
bucket.transformation.interval=28800

For Websphere Portal 8+ and WCM Rendering Only
### Do these only if in a WCM rendering environment only
# Description = Portal/WCM initial tuning
global.toolbar.enabled=false
# Description = Portal/WCM initial tuning
default.toolbar.enabled=false

WCM WCMConfigService
# Description = Portal/WCM initial tuning
deployment.subscriberOnly=true
# Description = Portal/WCM initial tuning
connect.moduleconfig.ajpe.contentcache.defaultcontentcache=SECURED
# Description = Portal/WCM initial tuning
connect.moduleconfig.ajpe.contentcache.contentcacheexpires=REL 2H
# Description = Portal/WCM initial tuning
user.cache.enable=true

For Authoring Only
WP AccessControlDataManagementService
# Description = Portal/WCM initial tuning for Authoring environments
accessControlDataManagement.acucIgnoreResourceTypes=null
# Description = Portal/WCM initial tuning Authoring environments
accessControlDataManagement.loadRolesParentBased=true</ResourceEnvironmentProviderName></value>

Feed for Blog Entries Feed for Blog Comments

▼ Similar Blogs

▼ Similar Ideation Blogs

▼ Archive

▼ Blog Authors

Special Weapons!