28 Jun
2009
28 Jun
'09
3:55 p.m.
I've got a very strange behaviour of my image...
It's all ok with Image Persistency on locally on my windows PC. But when I
switch to it at hosting (it's webfactional.com with ubuntu) our site stops
to answer (after saving the image, if I understand correctly). When I click
on "snapshot" in "System Management" I get the following walkback in
browser:
Internal ErrorError: subscript is out of bounds: 2
MethodContext(Object)>>errorSubscriptBounds:
MethodContext(ContextPart)>>at:
MethodContext>>tempAt:
[] in [] in WAWalkback>>renderStackFrame:on:
Array(SequenceableCollection)>>withIndexDo:
[] in WAWalkback>>renderStackFrame:on:
BlockClosure>>renderOn:
[] in WAGenericTag(WATagBrush)>>with:
BlockClosure>>renderOn:
WARenderCanvas(WACanvas)>>nest:
WAGenericTag(WABrush)>>with:
WAGenericTag(WATagBrush)>>with:
WARenderCanvas(WAHtmlCanvas)>>definitionList:
WAWalkback>>renderStackFrame:on:
[] in [] in [] in WAWalkback>>renderStackOn:
BlockClosure>>renderOn:
[] in WAGenericTag(WATagBrush)>>with:
BlockClosure>>renderOn:
WARenderCanvas(WACanvas)>>nest:
WAGenericTag(WABrush)>>with:
Without any buttons, nor anything visible in Pharo (using VNC). But "Process
Browser" shows no process for Seaside... How can I find the cause of the
problem and/or fix it?
--
Dennis Schetinin
28 Jun
28 Jun
4:06 p.m.
2009/6/28 Dennis Schetinin <chaetal(a)gmail.com>
I've got a very strange behaviour of my image...
It's all ok with Image Persistency on locally on my windows PC. But when I
switch to it at hosting (it's webfactional.com with ubuntu) our site stops
to answer (after saving the image, if I understand correctly). When I click
on "snapshot" in "System Management" I get the following walkback in
browser:
May be it can be the cause: we start WAComEncoded on port 2530 ...and we do
it using "port 2530" in command line.
--
Dennis Schetinin
7:37 p.m.
It looks like you are using a recent Pharo image and
Seaside2.8a1-lr.590. In this combination sometimes printing a walkback
causes another error. This is the stack trace you are seeing. Please
update to Seaside2.8a1-lr.591 to see the real cause of the problem.
Lukas
On Sun, Jun 28, 2009 at 4:06 PM, Dennis Schetinin<chaetal(a)gmail.com> wrote:
2009/6/28 Dennis Schetinin <chaetal(a)gmail.com>
--
Lukas Renggli
http://www.lukas-renggli.ch
I've got a very strange behaviour of my image...
It's all ok with Image Persistency on locally on my windows PC. But when I
switch to it at hosting (it's webfactional.com with ubuntu) our site stops
to answer (after saving the image, if I understand correctly). When I click
on "snapshot" in "System Management" I get the following walkback in
browser:
May be it can be the cause: we start WAComEncoded on port 2530 ...and we do
it using "port 2530" in command line.
--
Dennis Schetinin
_______________________________________________
Magritte, Pier and Related Tools ...
https://www.iam.unibe.ch/mailman/listinfo/smallwiki
8:08 p.m.
It looks like you are using a recent Pharo image and
Seaside2.8a1-lr.590. In this combination sometimes printing a walkback
causes another error. This is the stack trace you are seeing. Please
update to Seaside2.8a1-lr.591 to see the real cause of the problem.
Are there instruction for update?
And I have some news about the situation. ...Still strange but after some
work I have 'a service is already running on port ' exception during
snapshot. But manually starting WAKomEncoded on that port is ok... And
clicking "proceeding" also works.
Seems like port is still "busy" during or just after snapshot, but becomes
"free" few moments later. Isn't it some kind of synchronization problem?
--
Dennis Schetinin
8:15 p.m.
On Sun, Jun 28, 2009 at 8:08 PM, Dennis Schetinin<chaetal(a)gmail.com> wrote:
You just load the latest version through Monticello from the Seaside repository.
It looks like
you are using a recent Pharo image and
Seaside2.8a1-lr.590. In this combination sometimes printing a walkback
causes another error. This is the stack trace you are seeing. Please
update to Seaside2.8a1-lr.591 to see the real cause of the problem.
Are there instruction for update? Seems like port is still "busy" during or
just after snapshot, but becomes
"free" few moments later. Isn't it some kind of synchronization problem?
All sockets are closed during a snapshot, thus the web server has to
be restarted. I think I heard before that people run into problems
when setting the port environment variable and saving their images. I
don't remember the solution though.
Lukas
--
Lukas Renggli
http://www.lukas-renggli.ch
8:41 p.m.
It looks
like you are using a recent Pharo image and
Seaside2.8a1-lr.590. In this combination sometimes printing a walkback
causes another error. This is the stack trace you are seeing. Please
update to Seaside2.8a1-lr.591 to see the real cause of the problem.
Are there instruction for update? Seems like port is still "busy" during
or just after snapshot, but
becomes
"free" few moments later. Isn't it
some kind of synchronization problem?
All sockets are closed during a snapshot, thus the web server has to
be restarted. I think I heard before that people run into problems
when setting the port environment variable and saving their images. I
don't remember the solution though.
8:53 p.m.
Btw, seems like unassigning before assigning helps :)
TcpService >> start
self verifyOkToStart ifFalse: [^self].
self unassignPort. "This line is new!"
self assignPort.
[super start] on: Error do:
[ :ex | self unassignPort. ex pass]
Though I'm not sure it will not have some bad consequences in some other
place...
--
Dennis Schetinin
5415
days inactive
5415
days old
6 comments
2 participants
participants (2)
-
Dennis Schetinin -
Lukas Renggli