[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
Re: [mat-dev] Refreshing MAT's web pages
|
I submitted some proposed edits for the
IBM section:- Used the most recent version of MAT+DTFJ
to update the list of heap dump types
- Removed the ancient information about
1.4.2
- Added discussion about system dumps
vs PHDs
- Removed the reference to using MAT in
IBM Support Assistant as we're moving away from that
System Dumps and Heap Dumps from IBM Virtual MachinesMemory Analyzer may read memory-related information from
IBM system dumps and from Portable Heap Dump (PHD) files with the IBM
DTFJ feature installed. Once installed, then
File > Open Heap Dump should give the following options
for the file types:
- All known formats
- HPROF binary heap dumps
- IBM 1.4.2 SDFF
- IBM Javadumps
- IBM SDK for Java (J9) system dumps
- IBM SDK for Java Portable Heap Dumps
For
a comparison of dump types, see Debugging
from dumps. System dumps are simply operating
system core dumps; therefore, they are a superset of portable heap dumps.
System dumps are far superior than PHDs, particularly for more accurate
GC roots, thread-based analysis, and unlike PHDs, system dumps contain
memory contents like HPROFs. Older versions of IBM Java (e.g. < 5.0SR12,
< 6.0SR9) require running jextract on the operating system core dump
which produced a zip file that contained the core dump, XML or SDFF file,
and shared libraries. The IBM DTFJ feature still supports reading these
jextracted zips; however, newer versions of IBM Java do not require jextract
for use in MAT since DTFJ is able to directly read each supported operating
system's core dump format. Simply ensure that the operating system core
dump file ends with the .dmp suffix for visibility in the MAT Open
Heap Dump selection. It is also common to zip core dumps because they are
so large and compress very well. If a core dump is compressed with .zip,
the IBM DTFJ feature in MAT is able to decompress the ZIP file and read
the core from inside (just like a jextracted zip). The only significant
downsides to system dumps over PHDs is that they are much larger, they
usually take longer to produce, they may be useless if they are manually
taken in the middle of an exclusive event that manipulates the underlying
Java heap such as a garbage collection, and they sometimes require operating
system configuration (Linux,
AIX)
to ensure non-truncation. In recent versions of IBM Java (> 6.0.1), by default,
when an OutOfMemoryError is thrown, IBM Java producesa system dump, PHD, javacore, and Snap file on the first occurrence for
that process (although often the core dump is suppressed by the default
0 core ulimit on operating systems such as Linux). For the next three occurrences,
it produces only a PHD, javacore, and Snap. If you only plan to use system
dumps, and you've configured your operating system correctly as per the
links above (particularly core and file ulimits), then you may disable
PHD generation with -Xdump:heap:none. For versions of IBM Java older than
6.0.1, you may switch from PHDs to system dumps using -Xdump:system:events=systhrow,filter=java/lang/OutOfMemoryError,request=exclusive+prepwalk
-Xdump:heap:none
In addition to an OutOfMemoryError, system dumps may be
produced using operating system tools (e.g. gcore in gdb for Linux, gencore
for AIX, Task Manager for Windows, SVCDUMP for z/OS, etc.), using the IBM
Java APIs, using the various options of -Xdump,
using Java
Surgery, and more.
Versions of IBM Java older than IBM JDK 1.4.2 SR12, 5.0
SR8a and 6.0 SR2 are known to produce inaccurate GC root information.
--
Kevin Grigorenko
IBM WebSphere Foundation SWAT Team
kevin.grigorenko@xxxxxxxxxx
Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/kevgrig/
From:
"Tsvetkov, Krum"
<krum.tsvetkov@xxxxxxx>
To:
Memory Analyzer Dev
list <mat-dev@xxxxxxxxxxx>
Date:
11/08/2016 05:24 AM
Subject:
[mat-dev] Refreshing
MAT's web pages
Sent by:
mat-dev-bounces@xxxxxxxxxxx
Hi, A few days ago I started cleaning up the
MAT web site and wiki pages. There were a number of broken links, outdated
information, very old styling theme, etc …I am not capable of creating a really
good and modern looking page, therefore I focused on cleaning up broken
stuff and trying to put some better structure. Here is a summary of what changed so far:- Updated
the home page https://www.eclipse.org/mat/ to use similar styling as most of the other projects – the “solstice”
theme. Again, it is not the super-good-looking page, but at least we
are not two themes behind most of the other eclipse projects- Used
the navigation bar on the top, which comes with the new theme, to
structure groups of links related to developers/contributors, documentation,
support. The info/links was available already before, but spread
unstructured across our home page - Removed
the links to previous presentations from the home page (the newest
one was 2010), created a wiki page with the still working links, and linked
it under documentation as additional learning material. They were a few
links which didn’t work any longer. For some I managed to find the
replacement, the others I deleted. If you happen to know some new tutorial
or presentation which we could link – let me know. A couple of weeks ago
I was at EclipseCon Europe and was pleasantly surprised to be in two sessions:
https://www.eclipsecon.org/europe2016/session/optimising-eclipse-plug-insand https://www.eclipsecon.org/europe2016/session/performance-testing-budget. - I removed
the links to the webinars. The links to adobedev.adobe.acrobat.com
are not responding any longer. I spend some time looking for new links
to the old info, but found none. If anyone has an idea where to find them
(e.g. the adobedev.adobe.acrobat.com/p76554151/ one) let me know.- On
the wiki (linked as Getting Started) I tried to keep the Getting
started part minimal – install and a links to two (in my opinion) good
first-reading materials. Please have a look and let me know if you
still find some broken links / wrong data! Suggestions for improvement
are also welcome! Some further changes I will try to do (and
I need some help):- On
the wiki we still have a lengthy “Getting a Heap Dump” description
https://wiki.eclipse.org/MemoryAnalyzer#Getting_a_Heap_Dump. This is pretty much the same as the one in the documentation within MAT,
which is also available online: http://help.eclipse.org/neon/topic/org.eclipse.mat.ui.help/tasks/acquiringheapdump.html. In addition, both descriptions look quite outdated – they are mostly
talking about java 1.4 to 1.6, there is java 8 beta mentioned on one place,
there are some descriptions specific to MAT 1.2 and 1.3, etc… While most
of the information is still correct, it leaves the impression that the
info is out-of-dateo I
will try to update the description – add some details about java 8, remove
the paragraphs about MAT 1.2, etc…o I
would appreciate if one of the IBM colleagues can check if the descriptions
for using IBM dumps are still fine and update if necessary (also add new
java versions to the list if needed)o The
open question is where to fix it – at the end I see no use in having the
info maintained twice. I would like to have an up-to-date documentation
within the tool, and have it also online. The issue is, that with our release
tempo, the next online version of the docu will be June 2017, and I’d
like to have the update sooner than that. I am thinking about fixing both
documentation and Wiki and then removing the Wiki once we have an online
version of the documentation which is up-to-date. If we manage to do improvements
to the doc, we could probably o Talking
about java versions and heap dumps – we’ll have to watch how java 9 evolves
and see if there are some changes relevant for MAT- Another
piece of documentation which needs to be refreshed is the contributor
guide. I’d like to review it again and come up with an up-to-date
description. This would need more time, so let’s see if I manage to get
to it. Feedback is welcome! Krum _______________________________________________
mat-dev mailing list
mat-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe
from this list, visit
https://dev.eclipse.org/mailman/listinfo/mat-dev