Alana Sparx

Ranch Hand
+ Follow
since Feb 14, 2006
Merit badge: grant badges
For More
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 Alana Sparx

Hi all

Not really my area of expertise, so I'll throw all sorts of info into this.

We have deployed a couple of web apps on Websphere CE. There are two servers, which recieve requests dispatched via 2 IIS servers. When you are connected to the applications, the requests zip along. However, perfromance is sluggish at times: it zips along, and then every request hangs for up to 10 minutes. All of a sudden, everything is rosy again.

I have no idea what could be causing this, the network side of life isn't really my area of expertise. Could it be down to the AJP connections? We also have a Sitemnder security layer on the IIS server, could this be to blame.

Obviously you can't tell me what the issue here is, but does anyone have any experience of a similar issue, or any areas to look into?

Much obliged

As
15 years ago
Hello all

Bit of a heated debate going on here.

We have an enterprise level system potentially being used by a worldwide community of thousands.

There are some issue in the application code, but I would like to know what your ideas on this is



This to me is the correct way to do things, reusing the same memory address each time the loop iterates.

One of our experienced programmers says this is not an issue, and that this is perfectly acceptable:




Examples of this are littered throughout the code base (where the declaration and assignement are contained in the body of the loop). To me, this runs the risk of memory leaks and stressing the garbage collector, but I am willing to be corrected.

Any thoughts out there?

Non abbreviated thanks 30 seconds after last attempt (are you trying to stop people posting?)
15 years ago
Hi all

Statements we execute against the database are VERY SLOW INDEED.

The sql is generated by Hibernate.
The application server is Webshpere Community Edition.
The database is Sql Server 2005.
The database driver is the jTDS driver.

If we take the sql generated by hibernate and execute it directly against the database, the query flies along.
If we take the sql and execute it as a prepared statement through a jdbc connection, again it flies along.
When we use the service that ultimately calls the Hibernate classes, we are waiting in excess of 60 seconds for the query to complete.

Does anyone have any pointers that can help remedy this?

I've also posted a similar question here, on the Websphere Forum
Hi, here's hoping!

We are using Webshpere Community Edition as our application server, and Hibernate as the ORM tool. We have noticed that queries are taking an impossibly long to time to execute (over a minute to return 600 rows), and are basically blaming the Application Server for this. The Database Server is Sql Server 2005.

We have taken the sql generated by Hibernate and executed it directly against the server - the query zips along in 0.2 secs. I have created a JDBC connection and executed the statement as a prepared statement, and this also zips along. The DB driver we use is the jTDS driver.

Can anyone help explain this behaviour, or give me pointers as to possible solutions? Currently, we're thinking of shifting application server as the finger seems to be pointing in it's direction.

Many thanks
AS
16 years ago
Hi again

Here is an interim solution for anyone else who comes across this situation.

Note that the collection I'm dealing with is immutable, if yours isn't, I'm not sure what the consequences of this would be.

We augmented the Hibernate class CollectionEntry from

to


Basically, we check for the specific collection name that was causing us grief, if the collection isn't this, Hibernate carries on as usual. If the collection is our own specific one, don't bother with the flush.

loadedPersister has a mutable flag on it, wasn't sure if checking this flag and evading postFlush() is an appropriate thing to do or not, probably a question that can best be answered by the Hibernators
Hi Mark, tanks for the reply.

Well, it appears you have closed your Session.

I don't think that was the issue. I set the xdoclet 'lazy="false"', which ceased reporting the session being closed, but the AssertionError is still being reported

I have a couple of questions.

1. Why are you using xdoclet. You are using Hibernate 3.2

Only because we were previously using version 3.0.?.ga, and upgrading to 3.2.5 was in response to the error above ie no time to refactor the xdoclets (Hibernate site suggested it was an error fixed in 3.2.1 Further investigations suggest it is still a problem, as yet unresolved as reported here)

I don't think the xdoclets would affect the application in this way. The error above, for the project I'm working on, is a bit of a showstopper, so I'm in a bit of a 'brown trouser' situation at the moment.

Any other ideas?
Hi all

I've been getting the above error no matter what I try and do to solve it.

THe collection in question is

I've removed the xdoclet tag from the collection heading.
I've overridden the compateTo method to check for
I've used the latest version of hibernate (3.2.5.ga) and still I get the same exception thrown!

PolicyStatusRestriction does not contain any sets. Both PolicyStatus and PolicyStatusRestriction are immutable.

Any ideas? THe full stacktrace follows:
Hi Shailesh

Oh, it's much more serious than that. It would help if the test database I was using had a [group] table in the first place!

Sometimes my own stupidity amazes even me!

Many thanks
Hi all

I need to model a table called group which, being an SQL keyword is giving me some problems.

My Group.hbm.xml file is defined as:


When I try and query the database, the following sql is produced



which falls over with the exception message


If I run the sql against the database, it works fine!!

I've tried replacing the [] with backticks (``), the same eception is thrown.

There is also the problem that the query returns everything despite being given a QBE object with the composite keys filled in, but one thing at a time, eh?
Hi all

I'm writing a hibernate mapper for an existing database, and only need to use a handful of objects from this db.

When I run my test, I keep getting the following error:
Foreign key (FK5E0F67F881B2AA8:group [group_code]))
must have same number of columns as the
referenced primary key (group [ec_env,group_code])


The group.hbm.xml lookslike


The group model object & xdoclet tags looks like:



The composite key of group is a foreign key of two other db objects that I haven't modelled, could this be the issue?

I am missing something here in

infoCriteria.add(in(PROP_GROUP, groups));

where is the "in" method? Is that a method you have implemented, or something that is coming from a static import? And if you implemented it what does it look like.


Yes, it was a static import from the Restrictions class.

I could have implemented it myself, but it would have been awful....
Sorted it myself.

I'd tell you, but owing to your reticence, you don't deserve to know.
Hibernate is driving me crazy.

Model class (informationDisplay.java):



daoHibernate class


if I comment out the line infoCriteria.add(in(PROP_GROUP, groups));, everything works as expected. Leaving it in throws it all over the place:
1) sql error stating parameter #2 has not been set - eh?
2) the id of the Group object contained in the group set looks like it is being used to search the id of Information_Display - it should be set to the GROUP_ID of GROUP_INFORMATION_DISPLAY. I thought Hibernate took care of this.

Can anyone help. I've been battering my head against this wall all afternoon, in which time I could have JDBC'd half the bloody universe!!!
Any recommendations?

Query Analyser I have found to be a primitive piece of rubbish.

I've tried Squirrel-Sql, and although it supports nifty functions (like code complete, returning metadata and rowcounts) it has the annoying habit of endlessly writing to its heapdump logfile until I kill the process if it encounters errors / sql statements that generate large result sets. These errors can be as simple as the connection to the DB no longer existing. Latterly I tried an outer join on two table with 1 million+ rows apiece. THis should be manageable, yes?

Help!
Hi

We are possibly upgrading to Websphere 6.1 from WS 5.1.1.10, and I need some advice.

Our production and UAT / Staging environments will at some point require vertical and horizontal scaling for redundancy and failover. From my brief look , it would appear that we'll need WebSphere� Application Server Network Deployment V6.1

Our other test environments are Sys and Dev. A fully scaled Application Server environment seems here to be overkill, as at this point we're testing the business logic, so some other (less expensive!) Websphere option would seem to make sense. As there is potential for a number of projects to be on the go simultaneously, we'll need numerous instances of the Application Server each containing different instances of the middleware (which would ultimatley be merged at UAT prior to release). Would this point us to licences for IBM� WebSphere� Application Server, V6.1 as a less heavyweight version of the networked edition? Is it too simplistic to think of using the Community Edition for development? And if we use the Community Edition, we can run it inside Eclipse, avoiding the purchase of WS6.1 version of WSAD.

Having the full Network Deployment option across all environments seems like overkill to me, and would also be prohibitivly expensive. Our hardware is set out as follows:

Prod: 2 Servers, Dual 3.8GHz processor(HS20), 4Gb Ram, 40Gb Disk
UAT: Virtual Server x2, LS41 Quad AMD Opteron,
SYS: Virtual Server, HS20 Dual Processor, 2Gb Ram, 40Gb Disk
DEV: Virtual Server, HS20 Dual Processor, 2Gb Ram, 40Gb Disk

Any recommendations gratefully received!
17 years ago