Re: One more time - EVFDSPCH problems -- WDSCI-L




Vern,
I had the same thing happen before but in a different way.  It happened
when I was deciding between LPEX and CODE.  I did a recent test to recreate
what I was thinking and WDSC forced me to reboot and even break supposed
IBM rules.  Are there members in the LPEX Editor kept open for awhile?  I
think you mentioned that in a previous related posting about this same
problem.
There are several problems that hinge on closing WDSC while keeping members
open in LPEX for edit without saving members:
1. LPEX can hang.
Open a source member in the LPEX editor.  Don't do anything except exit
WDSC while member is open.  Make a change to that member in CODE or SEU
outside of WDSC.  Get back into WDSC and make a change to that member
already opened.  Now exit WDSC, select for backup, and "Yes" to load file
changes.  Now try to get back into WDSC and it says that another instance
of WDSC is up and if this is not the case, delete the ".lock" file in
".metadata" workspace directory.  Trying to delete this file shows it is
locked.  The best alternative I know of is to reboot and then you can get
back into WDSC.
2. If you open a member for edit in CODE and then open WDSC, you can have
the same member open for edit in LPEX and CODE.
Open a member in the LPEX editor and then exit WDSC.  Now open a member for
edit in CODE outside of WDSC.  Now open up WDSC and the same member will
now be open for edit in both CODE and LPEX.  This is not good, considering
there are checks in place to stop CODE and LPEX from opening the same
member up for edit twice.  It normally brings up the member for browse if
already being edited.  I think the WDSC workspace works with different
logic and keeps the same settings as before.  Now you can edit in CODE,
save, edit in LPEX while the same member is up for edit at the same time.
I did some experimenting and I was able to cause some exception errors this
way.  Not good at all.

I am on V5R2 WDSC 5.0 (haven't made that step to 5.1 yet).
Are these known problems, IBM?
Are you able to recreate your problem, Vern?  I think it would be good to
know exactly how you are opening the members in error and noting anything
that may be out of the ordinary.  I didn't mess with STRCODE this time.
Might it be a combination of #1 and #2 above?

Craig Strong

** Vern wrote:
I'm trying to open a source member from RSE with CODE editor. I do not have
a STRCODE session going. I get the MCH3401 that it can't find EVFDSPCH in
the CODE comm log. Nothing in the WDSC log.

V5R1 of OS/400, pretty current on PTFs
V5 of WDSC Advanced with latest updates (as of a couple weeks ago)

The downloading window sits there, finally times out (I assume). Then, when
trying again, I get "Host server is busy".

Looked in archives, saw something about library QCODE - no such animal on
this box.

Windows XP SP1

If I STRCODE, it seems to work fine from CODE but not from RSE - does not
find the CODE server. STRCODE *USRSPC has OS400 (default) in it, with lots
of values, and has size of 4096. DSPF on it says it is using record length
of 132 - 18 rows displayed altogether.

When I reboot Windows (to get rid of "busy" message) and start WDSC, then
try to open a member again, it still hangs, same messages in Comm log.
STRCODE *USRSPC now shows up in DSPF with row length of 3470, 3 of them,
although object size is still 4096. Probably something strange with DSPF.

Port is 4200, IP address is correct.

====

A little history - I have tried to open a member with CODE that was also
open in RSE, both for edit, IIRC.

Help
Vern