Re: [pgadmin-support] 4th Try: pgadmin4 install, launch fail

FYI - this hasn't been forgotten. Devrim and I were travelling most of
last week, and have PostgreSQL releases to deal with this week. We're
planning to meet tomorrow to work on the RPMs.

On Wed, Feb 1, 2017 at 7:11 PM, stancs3 <scruise56@gmail.com> wrote:
> Just an additional note:
>
> I am not a mousey-clicky guy and insist on the Application launcher to
> work. Any CLI is great in fact much better. I am only harping on this
> because the sequence of use from a stock install will tell something
> about what is going on.
>
> And of course it will be nice to see the desktop version :).
>
>
> On Wed, 2017-02-01 at 12:07 -0700, stancs3 wrote:
>> Yes that's a typo (the only time I didn't do a copy/paste).
>>
>> The sequence is:
>>
>> 1. Make sure no pgAdmin4 processes are running.
>> 2. Make sure no python processes are running.
>> 3. Launch pgAdmin4 from Applications menu.
>> 4. pgAdmin4 splash appears - the nice blue one.
>> 5. The Fatal Error window appears with text below with pgAdmin4.py
>> 6. Note that a process is running called pgAdmin4 and showing a
>> command
>> of: /usr/pgAdmin4-v1/runtime/pgAdmin4. This aligns with what is coded
>> into /usr/share/applications/pgAdmin4-v1.desktop. This is stock from
>> the install.
>> 7. Note that no python process is running. (as one would expect given
>> the error message).
>> 8. Then I hit the OK button in the Fatal Error window.
>> 9. It all goes away, including the process in step 6 above.
>> 10. When this happens an abrt-hook-ccpp log is generated, which I can
>> send if it will help.
>> 11. I confirm that there is no prompt for paths.
>>
>>
>> My sincere apologies for the typo below; it is unhelpful to add
>> errors
>> into debug data.
>>
>> Anyway, I am at your service for further debug attempts.
>>
>>
>> Stan
>>
>>
>>
>> On Wed, 2017-02-01 at 09:25 +0000, Dave Page wrote:
>> >
>> > Hi
>> >
>> > On Tue, Jan 31, 2017 at 6:49 PM, stancs3 <scruise56@gmail.com>
>> > wrote:
>> > >
>> > >
>> > > Attached is the desktop launcher copied from
>> > > /usr/share/applications/.
>> > > This was created from the install.
>> > >
>> > > Error resulting:
>> > >
>> > > "An error occurred initialising the application server:
>> > >
>> > > Failed to locate pgAdmin5.py, terminating server thread."
>> > Does it really say pgAdmin5.py? If so, that's not our code.
>> >
>> > >
>> > >
>> > > This does not give you much further info. I can provide more
>> > > debug
>> > > info
>> > > if it will help; I am game to be coached on what/how to collect.
>> > When you click OK on the message above (assuming it does actually
>> > say
>> > pgAdmin4.py, not pgAdmin5.py), it should prompt you for the paths.
>> > Does it not do that?
>> >
>> > >
>> > >
>> > > Your previous comment re a WIP for repo install: if it would
>> > > simply
>> > > be
>> > > better if I was to go the compile route, I am fine with this. I
>> > > did
>> > > it
>> > > for pgadmin3 a few months ago ... and it was good. On the other
>> > > hand, a
>> > > nice repo install is likely the goal for us user dudes :).
>> > Devrim and I will be spending some time, probably next week, fixing
>> > the RPMs. We've had a small re-org at work which make it much
>> > easier
>> > to allocate some of his time for this :-).
>> >
>> > Unfortunately I doubt the compile route will help here. The Python
>> > code and the runtime are independent pieces of code - the runtime
>> > should locate the Python code based on it's configuration, not any
>> > build settings.
>> >



-- 
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company