PK99815: Supercede the interim fix for PK94209

PK99815; 7.0.0.5: Supercede the interim fix for PK94209

APAR status

• Closed as program error.

Error description

• PK94209 corrects the following problem:
  Improve application deployment performance by expanding
  nestedarchives to temporary storage when an access threshold
  is reached.
  
  However, there was an error in the packaging of the interim
  fix for PK94209 which would cause the installation of
  subsequent interim fixes to fail. PK99815 supercedes PK94209
  and repackages the interim fix to correct this problem.
  
  One of the problem symptiom is that when user is trying to
  deploy a application from console when selecting the app after
  clicking on browse buttin it just hangs there and following hung
  thread message with stack will get printed into SystemOut.log
  file :
  
  [10/31/09 10:57:08:903 EDT] 0000000c ThreadMonitor W
  WSVR0605W: Thread
  "WebContainer : 2" (0000002e) has been active for 652568
  milliseconds and may be hung.  There is/are 1 thread(s) in
  total in the server that may be hung.
   at java.util.zip.Inflater.inflateBytes(Native Method)
   at java.util.zip.Inflater.inflate(Inflater.java:223)
   at
  java.util.zip.InflaterInputStream.read(InflaterInputStream.java:
  135)
   at
  java.io.BufferedInputStream.fill(BufferedInputStream.java:218)
   at
  java.io.BufferedInputStream.read1(BufferedInputStream.java:258)
   at
  java.io.BufferedInputStream.read(BufferedInputStream.java:317)
   at java.io.FilterInputStream.read(FilterInputStream.java:116)
   at
  java.io.PushbackInputStream.read(PushbackInputStream.java:169)
   at
  java.util.zip.InflaterInputStream.fill(InflaterInputStream.java:
  221)
   at
  java.util.zip.InflaterInputStream.read(InflaterInputStream.java:
  141)
   at java.util.zip.ZipInputStream.read(ZipInputStream.java:146)
   at
  java.util.zip.ZipInputStream.closeEntry(ZipInputStream.java:92)
   at
  java.util.zip.ZipInputStream.getNextEntry(ZipInputStream.java:70
  )
   at
  org.eclipse.jst.j2ee.commonarchivecore.internal.strategy.NestedA
  rchiveLo
  adStrategyImpl.getZipInputStreamSkippedTo(NestedArchiveLoadStrat
  egyImpl.  java:249)
   at
  org.eclipse.jst.j2ee.commonarchivecore.internal.strategy.NestedA
  rchiveLo
  adStrategyImpl.getInputStreamSequentially(NestedArchiveLoadStrat
  egyImpl.  java:207) at
  org.eclipse.jst.j2ee.commonarchivecore.internal.strategy.NestedA
  rchiveLo
  adStrategyImpl.getInputStream(NestedArchiveLoadStrategyImpl.java
  :203) at
  org.eclipse.jst.j2ee.commonarchivecore.internal.impl.ContainerIm
  pl.primG etInputStream(ContainerImpl.java:431)
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED:  All users of IBM WebSphere Application      *
  *                  Server version 7.0                          *
  ****************************************************************
  * PROBLEM DESCRIPTION: Slow application deployment resulting   *
  *                      from zip file decompression overhead    *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  This update provides a performance enhancement by reducing the
  overhead of accessing entries in nested archives.  Overhead is
  reduced by expanding the nested archive to a temporary
  location, and by using that temporary copy of the nested
  archive for further accesses.  For medium to large sized
  applications, without this update, the overall cost of
  accessing entries within nested archives can grow to be in
  excess of an hour.
  
  Frequent access is made to entries within nested archives
  during the deployment of new enterprise application.  The
  performance improvements provided by this update apply mostly
  to application deployment operations.
  
  Getting input streams for entries from within nested archives is
  expensive due to archive decompression overhead.  A request
  for an input on an entry in a nested archive requires a scan and
  decompression from the beginning of that nested archive.
  

Problem conclusion

• This update provides a performance enhancement by enabling the
  extraction of temporary copies of nested archive files when
  repeated access is made to entries within the nested archives.
  
  When obtaining an input stream for a entry from within a nested
  archive, for example, when loading the web module deployment
  descriptor, "web.xml" from a web module archive (WAR file)
  that is stored within enterprise application archive (EAR
  file), as long as the nested web module archive is left inside
  the enterprise application archive, accessing web module
  deployment descriptor requires first that the nested web module
  archive be accessed, then, second, that the entries of the
  nested web module archive be scanned and decompressed until the
  target web module deployment descriptor is reached.
  
  When the number of entries which are accessed is small, or when
  the overall web module archive is small, the cost of the
  extra scan and decompression is relatively low, and is not a
  problem.
  
  When the number of entries which are accessed is more than a
  few, or when the web module archive has more than a few tens of
  kilobytes (KB) of data, the cost of the access can grow to be
  large.
  
  When the number of entries which are accessed is large, or
  when the web module archive is very large, the cost of the
  access can grow to be quite excessive.  (Application
  deployment times in excess of an hour have been seen.)
  
  This update causes the creation of new temporary files.  To
  disable creation of the new temporary files, use the following
  custom JVM property setting:
  
  com.ibm.config.eclipse.wtp.dynamic.expansion=false
  
  **************************************************************
  All the previous APAR text describes the problem corrected by
  PK94209. However, the interim fix for PK94209 had a packaging
  problem which would cause the installation of subsequent
  interim fixes to fail. This APAR, PK99815, supercedes PK94209
  and repackages that interim fix only. There is no additional
  code change made with this APAR.
  **************************************************************
  
  *** Note that there is no problem with PK94209 as shipped in
  any fix pack.
  The fix for PK94209 is currently targeted for inclusion in
  fix pack 7.0.0.7.  Please refer to the Recommended
  Updates page for delivery information:
  http://www.ibm.com/support/docview.wss?rs=180&uid=swg27004980
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  WEBSPHERE APP S

• Fixed component ID

  5724J0800

Applicable component levels

• R700 PSY

     UP