Well -
It has been three consecutive days without a problem! It looks as
though this bit of code did the trick.
Before this code, every morning javamonitor would be stuck because of
one bad instance.
So far no problems and I am now sleeping much better, let alone not
being obsessed with c
On 27-Feb-07, at 2:55 PM, James Cicenia wrote:
... I don't even track users. ...
On 5-Mar-07, at 11:16 AM, James Cicenia wrote:
I even set the session timeout to be like six minutes.
...
Funny thing is this is only a read only app, though a very busy one
24 hours a day.
Do you really need
My best guess is that an exception is being thrown during session
check-in/check-out or sleep. Here's some code that I use in
Application.java to prevent this. I believe Wonder also has some of
these fixes.
Application.java:
public WOResponse handleActionRequestError(WORequest aRequest
That is my problem -
In my regular logs I see nothing out of the ordinary. In fact this
instance didn't seem to have a lot of traffic on it either.
I am thinking that the javamonitor scheduled restarts are fubarring
it up. I have turned off graceful scheduling figuring
it had to kill the inst
A session can only be checked out on one thread at a time. If you
notice you are hanging waiting to check out a session. This means
that a previous session did not check in properly when it was done.
It's important to note that you don't have any session hanging doing
anything else, so i
OK-
I think I got the correct thread dump from a frozen instance this
morning:
What does it all mean?
Thanks
James Cicenia
Full thread dump Java HotSpot(TM) Server VM (1.5.0_06-68 mixed mode):
"Thread-1" prio=5 tid=0x00638a90 nid=0x1825800 waiting on condition
[0xb169d000..0xb169dd10]
On 03.03.2007, at 09:44, James Cicenia wrote:
Maybe I read that wrong.. however, I ran it.. looked at all my
logs.. and there is nothing!
This goes to webobjects.err - whereever that is, it is defined in the
startup scripts for the WebObjects services.
You have to change this file (make a
Did all that. Nothing of note in those logs either.
:-(
On Mar 3, 2007, at 10:59 AM, Mike Schrag wrote:
By default, SpawnOfWotaskd.sh redirects stderr to /dev/null, which
is a really annoying default behavior. Since kill -QUIT writes to
stderr, you won't be able to see this stack trace unt
By default, SpawnOfWotaskd.sh redirects stderr to /dev/null, which is
a really annoying default behavior. Since kill -QUIT writes to
stderr, you won't be able to see this stack trace until you modify
SpawnOfWotaskd.sh (there's a wikibook entry on this somewhere).
On Mar 3, 2007, at 11:44 A
More info -
Ok I just killed the offending instance with a -9 and now my
javamonitor is all happy.
sigh.
- James
___
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing list (Webobjects-dev@lists.apple.com)
Hel
Ok -
Maybe I read that wrong.. however, I ran it.. looked at all my logs..
and there is nothing!
And, even if I try to force my url to use that instance.. It gets
redirected. This is good
as otherwise my users would scream at me again.
This is a very baffling deployment problem.
- James
So I run, as per the wiki:sudo kill -QUIT pid(with the proper
number there).
kill -QUIT will not actually kill a java application. kill -QUIT
causes a dump of all thread stack traces to stderr of the java
process (very useful for debugging). Is there a wiki page that
claims that will
I am so confused with this JavaMonitor and this particular application.
I have had JavaMonitor running on my browser all day personally
monitoring it.
Well eventually, JavaMonitor seems to die because of one instance
going awry. This
means that the number of sessions, transactions, then nev
James,
If you ever get that exception, it means you have a problem with
locking your editing contexts. Do you use contexts you create
yourself, not just the default context on the session? If so, ANY
access to the context or the EOs in it should be wrapped in a try-
finally block and the
I had to step out...
Anyway..
I can't impress that these are really simple apps, which is really
perplexing. They only read data, there is no messaging, nothing...
My session class is very simple too.. I don't even track users.
It seems like at night they just don't terminate.. it used to be
On 27.02.2007, at 10:57, Chuck Hill wrote:
Are you sure the apps shut down correctly? I have seen this happen
when the apps hang after being told to shut down. Usually this
means there is a deadlocked session or a thread waiting to check
out an already terminated session.
We have seen t
Are you sure the apps shut down correctly? I have seen this happen
when the apps hang after being told to shut down. Usually this means
there is a deadlocked session or a thread waiting to check out an
already terminated session.
Chuck
On Feb 27, 2007, at 9:42 AM, James Cicenia wrote:
No.
I thought maybe it was a locking problem on these apps so I put
in the old multiclockmanager ... but still about 5 times a week
it won't restart the apps and then I get that problem.
I have about 8 instances running of 5 apps... all pretty much the same
except for the skins.
- James Cicenia
Are all three servers running the same applications?
Chuck
On Feb 27, 2007, at 8:15 AM, James Cicenia wrote:
Hello -
I just can't figure this out. Maybe I am doing something wrong with
this server.
My javamonitor refuses to restart applications at night..
intermittently. And then
it wil
On 27 Feb 2007, at 16:15, James Cicenia wrote:
Hello -
I just can't figure this out. Maybe I am doing something wrong with
this server.
My javamonitor refuses to restart applications at night..
intermittently. And then
it will show something like the following (not the real url) until
Hello -
I just can't figure this out. Maybe I am doing something wrong with
this server.
My javamonitor refuses to restart applications at night..
intermittently. And then
it will show something like the following (not the real url) until I
kill javamonitor.
However, today my apps are sti
21 matches
Mail list logo
