IDempiere/FullMeeting20150325
⇐ Table of Contents | Full Meeting Minutes | Full Meeting 2015-03-25
CarlosRuiz: Good Morning
nmicoud_: Bonjour
HideakiHagiwara: Konnichiwa : D
CarlosRuiz: I'll be working today in preparing the experimental environment
nmicoud_: Any chance you can review https://idempiere.atlassian.net/browse/IDEMPIERE-2235 ?
nmicoud_: ah... later then :)
aguerra: good morning @CarlosRuiz
Deepak_: Good Morning
Deepak_: Carlos, what is experimental environment?
Deepak_: Carlos, what is experimental environment?/
CarlosRuiz: Hi Deepak_
Not-311d: [IDEMPIERE] deepak updated IDEMPIERE-2539 Attachment set to "IDEMPIERE-2539.patch"
CarlosRuiz: there are some big tickets from hieplq that I've not been able to test and peer review - so we agreed to make an experimental environment to test and check ONE BY ONE
Not-311d: [IDEMPIERE] Find attached fix. It is small and easily understandable. Please integrate same.
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2539
Deepak_: Hello Carlos
CarlosRuiz: nmicoud_, about IDEMPIERE-2235 I think it must be something like refreshing the list when the wizard window is closed - probably via an event
nmicoud_: but how can i get it ?
Not-311d: [IDEMPIERE] deepak updated IDEMPIERE-2521 Attachment set to "IDEMPIERE-2521.patch"
Not-311d: [IDEMPIERE] This is Adaxa contribution we converted to iDempiere. Please use attached patch to integrate.
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2521
nmicoud_: as the WReportCustomization is not final, i can't use it in the form.addEventListener(DialogEvents.ON_WINDOW_CLOSE, new EventListener<Event>() {
Not-311d: [IDEMPIERE] deepak updated IDEMPIERE-2521 labels set to "+Patch"
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2521
CarlosRuiz: I mean in cmd_Wizard
CarlosRuiz: of zkReportViewer
nmicoud_: yes
CarlosRuiz: indeed the event is there - ON_WINDOW_CLOSE
nmicoud_: In the actual event, if you try to use av (WReportCustomization), Eclipse says : Cannot refer to a non-final variable av inside an inner class defined in a different method
Deepak_: Carlos, it looks whatever you committed for IDEMPIERE-2450 is overwritten by next commit for IDEMPIERE-2483
Not-311d: [IDEMPIERE] deepak updated IDEMPIERE-2540 Attachment set to "IDEMPIERE-2540.patch"
Not-311d: [IDEMPIERE] Find attached patch.
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2540
Not-311d: [IDEMPIERE] deepak updated IDEMPIERE-2540 labels set to "+Patch"
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2540
CarlosRuiz: moved below Deepak_
Deepak_: You mean logic moved below based on isZoom condition?
CarlosRuiz: nmicoud_, you say the listitem is not updated - so, I think recreating the listitem in that event
nmicoud_: deleting its content and recreating everytime ?
CarlosRuiz: yes Deepak, no need to show it underlined if cannot zoom
CarlosRuiz: yes nmicoud_ - refreshing the list when the WReportCustomization is closed
nmicoud_: means, eveytime ? (even if the print format name is the same)
hieplq: in finance report, i will select a period. i want report one month per year but i also want three month per year. i can archive it by multi Calendar? anything must consider?
Deepak_: Carlos, Found NPE issue on IDEMPIERE-2483. Fixing and sending patch on ticket
CarlosRuiz: must be related to a ticket recently opened by hieplq - NPE on accounting daily facts
CarlosRuiz: yes nmicoud_ - I think is ok to refresh on every close
nmicoud_: ah ok :) will be easier this way
CarlosRuiz: hieplq, what do you mean three month per year? a report showing one column per quarter?
hahmed: Hello everyone
hieplq: finance report or some other report request a perior paramter. perior normal set in calander by one month per one perior.
aguerra: hi hahmed
hieplq: sometime i want report in three month not just one month.
CarlosRuiz: hi hahmed
CarlosRuiz: yes - you can create a calendar for quarters
CarlosRuiz: in such calendar each year will have just 4 periods
hieplq: it's can create multi calander?
hieplq: just confirm for sure. but i will test by myself.
CarlosRuiz: yes - I just add a new calendar called "Quarter Calendar (just for reporting)"
Not-311d: [IDEMPIERE] deepak updated IDEMPIERE-2483 Attachment set to "IDEMPIERE-2483.patch"
Not-311d: [IDEMPIERE] Fixing Null pointer exception while table id column not on report.
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2483
hieplq: ok. thanks @CarlosRuiz. will test it.
Deepak_: Carlos, yes bug may be same. can you apply ptatch attached on IDEMPIERE-2483 and it will fix it
hieplq: want ask anyone know about any best wiki editor.
hahmed: I am pulling my hair because of one posting error. I can’t post, can’t void, can’t reverse, nothing can be done for a pair of Vendor Invoice and Material Receipt, its so ambiguos I don’t know what to ask here!
hahmed: basically I am getting an error when trying to post M_MatchInv for a very simple Vendor Invoice and Material receipt
hahmed: I am getting the error “Posting Error (Invoice not posted yet) - .Doc_MatchInv”
hahmed: BOTH invoice and receipt are posted successfully!
CarlosRuiz: hahmed, I think in this case you need to debug Doc_MatchInv - line 249 - to figure out why it arrives there
Deepak_: hahmed, Do invoice line has product only or charge only?
hahmed: I checked, its looking for a fact_acct posting for product inventory clearance account (*-51290) and its not there
hahmed: the invoice line has 1 product only
Deepak_: So how your invoice posting?
hahmed: invoice posts perfectly Dr 51210-Product Cost and Cr 21100-Account Payable
hahmed: just the matching doesn’t post
hahmed: happening only for one invoice, every other document posts fine
Deepak_: Can you try reposting invoice and see fact account lines
Deepak_: my worry is that you may changed product setup after creation on invoice
hahmed: nothing was done, it just came out of blue and has shaved tens of hours off my life lol
Deepak_: what is your invoice type?
Deepak_: * Document type
hahmed: AP Invoice
hahmed: I will have to try to trace every step of posting to see what is it that its trying to find thats not there
Deepak_: Hard to answer
Deepak_: Looks your product is of type service
Deepak_: or it is of item type?
hahmed: product is item, uom is each, everything is very *default*
Deepak_: Sorry need to leave
Deepak_: Carlos. can you integrate 2483 and 2539, Both are small bug fix
hahmed: @Deepak_ thanks for the effort, bye
CarlosRuiz: 2359?
Not-311d: [IDEMPIERE] nmicoud updated IDEMPIERE-2235 Attachment set to "IDEMPIERE-2235_v2.patch"
Not-311d: [IDEMPIERE] Hi Carlos, Please find an updated patch. When the panel close, the list of print formats is always refreshed. But we need to play with translation cache ; that's why i added new methods in PO. Thanks, Nicolas
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2235
tsvikruha: Hi community, does anybody struggled with following issue? https://groups.google.com/forum/?hl=cs#!topic/idempiere/YxRjErNynP4
tsvikruha: It looks like there are some postgres connections bugs with locking of records when many users access to the same server
Not-311d: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2539 Attachment set to "Freight_Category.pdf"
Not-311d: [IDEMPIERE] Good catch [~deepak], Two issues: - after applied the patch the PDF appears without data - see attached Freight_Category.pdf probably using 0.5 as stroke raises a different bug, I didn't test with other values - I think we would need to change the LineStroke=1 and HdrStroke=2 on the Default Print Table Format (ID=100) to keep the default reports with same layout as before the patch Regards, Carlos Ruiz
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2539
Not-311d: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to release-2.1 [+0/-0/±1] https://bitbucket.org/idempiere/idempiere/commits/
Not-311d: [iDempiere] logilite 429402f - IDEMPIERE-2483: Fixing NPE
Not-311d: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2528 status set to "Resolved" -assignee set to "Deepak Pansheriya" -resolution set to "Fixed"
Not-311d: [IDEMPIERE] Solved with patch from [~deepak] on IDEMPIERE-2483
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2528
CarlosRuiz: tsvikruha, most probably you can find the problem on custom code
CarlosRuiz: could be a connection leak if you say it's ok after restarting postgresql
CarlosRuiz: golden rule there - always close your rs and pstmt on finally block
tsvikruha: Yes I know that about closing pstmt in finally block
tsvikruha: a reviewed almost whole custom code
CarlosRuiz: note there are plugins breaking that rule
tsvikruha: too in community code?
CarlosRuiz: I think we fixed all of them in core - but we don't apply peer review on plugins
ocurieles: @CarlosRuiz Good Morning, some news about the new Server for new features ? @spina is waiting with the bat
tsvikruha: any idea how to easily find all places where golden rule about closing statement is broken?
CarlosRuiz: ocurieles, that server is basically for hieplq -)
CarlosRuiz: :-)
ocurieles: I know, but we can help to @hieplq :/ This is our intention
CarlosRuiz: tsvikruha, there are another source of connection leaks harder to see - is when you have a rs/pstmt within a loop and you leave connections open in the loop - just closing the last when exiting the loop
CarlosRuiz: you can help hieplq without that :-)
CarlosRuiz: :-) but to answer your question - yes - I'm working on that environment
ocurieles: ok
tsvikruha: CarlosRuiz I noticed that bug appears almost every time when MSequence is generating document no, but when our custom processes are calling procesess triggered this…rs/pstmt within a loop are not used in our custom code…
CarlosRuiz: it means another process generating a document sequence and kept the record lock
CarlosRuiz: pgadmin shows which process lock wich process - sometimes you can discover which tables are locked and that helps you to see which code is that
CarlosRuiz: long processes that don't commit are source of contention too - and also if you ask users things in the middle of a transaction
tsvikruha: what you mean by 'longs process that don't commit' ?
CarlosRuiz: a process that run for example 15 minutes
tsvikruha: or why this could happen? that some process don't commit its transaction?
tsvikruha: i see
CarlosRuiz: suppose a process that creates an invoice on minute 1 - and then run for another 14 minutes without committing
CarlosRuiz: the sequence record for invoice is locked for 15 minutes
CarlosRuiz: so any other process trying to number an invoice during those 15 minutes is locked
CarlosRuiz: I think we refactored in idempiere some long running processes to cope with that
CarlosRuiz: but there are lots that can be refactored - like InvoiceGenerate
hieplq: thanks a lot @ocuriele_ @CarlosRuiz @spina for support me with new environmet.
CarlosRuiz: hieplq, I'm preparing an email to you with the suggestions about the experimental workflow
CarlosRuiz: hieplq, I sent you the email
CarlosRuiz: so - you decide :-) which feature is the first to tackle there? process parameters?
hieplq: yes. process parameter with some ticket relate it.
CarlosRuiz: is important to communicate to community the changes so they know what to test
hieplq: yes. will notify what i change in group.
hieplq: hi @CarlosRuiz, about "your public key for ssh". i will use ssh command to create key and sent public part for you?
CarlosRuiz: yes please
Not-311d: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to release-2.1 [+0/-0/±1] https://bitbucket.org/idempiere/idempiere/commits/
Not-311d: [iDempiere] globalqss 4862dbf - IDEMPIERE-1900 iDempiere Debian Installer / thanks to V Brunner for reporting https://groups.google.com/d/msg/idempiere/WbaMAPfK6QM/OY0aa6Xh1yUJ
Not-311d: [IDEMPIERE] hahmed created IDEMPIERE-2541 Landed Cost Allocation does not honour delete/void
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2541
Not-311d: [IDEMPIERE] hahmed updated IDEMPIERE-2541 description set to "To reproduce: Create Material Receipt Create Invoice (Vendor) for Landed Cost, Create the invoice line, create landed cost and select the above receipt and receipt line, click distribute, verify that landed cost allocation is created. Now delete the Landed Cost. Ideally at this stage the landed cost allocation should also be deleted, however its not deleted. Now complete the invoice and vo
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2541
nmicoud_: hi @egil0902 ; i've updated https://groups.google.com/forum/?fromgroups=#!topic/idempiere/utwFQ2p2bfg - we talked this subject in last november. Could you have a look at it and say if you're still interested ? If yes, i will upload a first draft patch.
nmicoud_: gtg, will come back later
egil0902: Hi nmicoud how are you. ok i am going to look this
egil0902: Yes i am interested in this
egil0902: @nmicoud if you can send me a patch i can test
nmicoud_: egil0902 : ok will prepare that
CarlosRuiz: hieplq - server https://experimental.idempiere.org/webui/ is ready - I sent you an email with instructions about deploying there
hieplq: thanks will check now.
chdrsto: Hi there ... is there a posibility to use the java client over a encrypted connection? I mean ... the server will be accessible over the internet.
CarlosRuiz: swing client?
chdrsto: yes
CarlosRuiz: swing client connects connects directly to the database
CarlosRuiz: I guess you can configure it to connect to postgresql via SSL
CarlosRuiz: but usually that's too slow if the server is far - for such case is better to use zk
CarlosRuiz: I would think for any case - as the swing client has big security issues
chdrsto: what do you mean with security issue?
CarlosRuiz: as it connects directly to database - it requires the database password saved somewhere - that's saved in a properties file that is easy to decrypt
chdrsto: aha .. ok, but this might happen only if someone gets connected / access to the PC (client)
chdrsto: rignt?
CarlosRuiz: yep - the file is on the PC
CarlosRuiz: also note that zk client has evolved a lot with new features - meanwhile the swing client is just kept working - no new features in the last 3~4 years for swing
chdrsto: ok, ... thanks. I have an issue with the webclient ... means, I've got performance issues and hope i could encrease the performance with Swing Client
hieplq: @CarlosRuiz, i split IDEMPIERE-2389 to some patch, and commit to idempiere-experimental, maybe it help you easy review.
hieplq: i still doing around IDEMPIERE-2389 for some ticket, hope next day i can finish for deploy a test install.
Not-311d: [iDempiere] CarlosRuiz_globalqss pushed 2 commits to release-2.1 [+2/-0/±4] https://bitbucket.org/idempiere/idempiere/commits/
Not-311d: [iDempiere] logilite a08a297 - IDEMPIERE-2540:Allowing Vendor qty break to specify breaks for all customers
Not-311d: [iDempiere] globalqss 86b3fd3 - IDEMPIERE-2540 Allowing Vendor qty break to specify breaks for all customers / peer review
Not-311d: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2540 status set to "Resolved" -resolution set to "Fixed"
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2540
Not-311d: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-400
Not-311d: [IDEMPIERE] [~steveh], this is intended for server running the script sign-database-build.sh - in eclipse you need to set AD_System.IsFailOnBuildDiffer=N
Not-311d: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-400
Not-311d: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+2/-0/±12] https://bitbucket.org/idempiere/idempiere/commits/
Not-311d: [iDempiere] globalqss 7a40550 - hg merge release-2.1 (merge release2.1 into development)
Not-311d: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+0/-0/±3] https://bitbucket.org/idempiere/idempiere/commits/
Not-311d: [iDempiere] globalqss c54e104 - show login information on the new test and experimental servers
