Win a copy of Securing DevOps this week in the Security forum!

sander hautvast

Ranch Hand
+ Follow
since Oct 18, 2002
Cows and Likes
Cows
Total received
0
In last 30 days
0
Total given
0
Likes
Total received
0
Received in last 30 days
0
Total given
0
Given in last 30 days
0
Forums and Threads
Scavenger Hunt
expand Ranch Hand Scavenger Hunt
expand Greenhorn Scavenger Hunt

Recent posts by sander hautvast

not that hard
get a mailsession from the container (see admin) (or, not as good: create your own)
create a javax.mail.Message with the session. Populate it and send.
see javax.mail package for details
11 years ago
>>Start java com.netcast.common.NC_AdminReportHandler %1
not exactly sure but try removing 'Start'. This command forks another shell, why bother?
12 years ago
google's so great:
http://publib.boulder.ibm.com/infocenter/wasinfo/v5r1//index.jsp?topic=/com.ibm.websphere.base.doc/info/aes/ae/trun_app_hotupgrade.html

haven't read it yet, as we never do this
used to do it in WAS4 all the time: copy the new jars and (ripple)start

WAS6 has interesting new features called rollout update: try googling on that, if you need
12 years ago
and the files are about 2Mb in size
12 years ago
our machines are pretty fast, and I don't know about yours...
but it takes just a second or less
12 years ago
you can also use AST (Application Server Toolkit) (ships with websphere)
it's Eclipse, with plugins for easier editing of ibm descriptors like extensions and bindings.
12 years ago
hi all,

we have a product called jintegra: it's a bridge between DCOM and java. The java part runs as a jar in a webapp on websphere 5.1.1.4 on AIX 5.3 (J2RE 1.4.2 IBM AIX build ca1420-20040626)
Works fine most of the time, but occasionally the listener (netstat -an) indicates that the listener is gone...(telnet won't work either)
Established connections continue to work properly.

A restart of the application server resolves the the situation.
Analysis of threaddumps suggests that the thread running serverSocketInstance.accept has not stopped.

Posted this in the Socket saloon too, but could this be a websphere thing?? J2ee spec disallows openening your own sockets, but websphere doesn't log anything and it happened only twice in 4 months.
12 years ago
I use kill -3 on a running production environment (AIX) without any problem. gives a java core txt file that you can analyse (i think the tool is on developerworks).
(but make sure you do it in test before trying it in production)

I have seen hung threads in a situation where websphere 4 acted as java-client. solution was upgrading websphere 4 with the latest security fix
12 years ago
hi all,

we have a product called jintegra: it's a bridge between DCOM and java. The java part runs as a jar in a webapp on websphere 5.1.0.4 on AIX 5.3 (so it's J2RE 1.4.2 IBM AIX build ca1420-20040626)
Works fine most of the time, but occasionally the listener (netstat -an) indicates that the listener is gone...(and telnet won't work either)
Established connections continue to work properly.
A restart of the application server resolves the the situation.
Anyone seen behavior like that in a java app???
Analysing thread dumps suggests that the thread running ServerSocketInstance.accept has not stopped. Could this be an OS-jdk thing?
There's seldom a good reason to use a Hashtable anyway.

why is that?
13 years ago
but I mentioned this method in my question...

I think I got the whole picture now:
The static inner is not accessible through the instance (as opposed to a static member). it's only accessible through the class (hence static).

So my question becomes irrelevant.
13 years ago
you say "admin server": websphere 4 ??
in a clustered websphere 4 environment, the jndi request is handled by the adminserver that's in the provider_url. WLM passes the request to any of the appservers available (the adminserver repository is a single DB schema for both adminservers and it contains info on the appservers in the cluster), gets a home object reference (stub) from it and sends this back to the client. (i guess this is similar for other j2ee containers.)
You could also specify the second adminserver in the url. works the same. both need to be running to make use of the cluster
i'd say your pc does something evil...I have never seen sth like that on Windows and any java app, including Eclipse

hmmm. My eclipse process takes 60 megs footprint or so. try a newer eclipse version??
13 years ago
in my humble opinion i'd say that you don't have a memory problem. Because that would give an OutOfMemoryError. Try evaluating the heap programmatically by using Runtime.freeMemory() once every interval
13 years ago
you can use the File object, it has an exists() method.
with that you can check for the existence of a file. enough for an answer?
13 years ago