IDempiere/FullMeeting20141029

From WikiQSS
Revision as of 18:50, 29 October 2014 by CarlosRuiz (talk | contribs) (full meeting)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Table of Contents | Full Meeting Minutes | Full Meeting 2014-10-29

CarlosRuiz: Good Morning
cool: ok let me try this
nmicoud: Bonjour
JanThielemann: hi
a42niem: hi all, back from hospital, getting up to normality again
CarlosRuiz: Hi Dirk
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+0/-0/±3] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] jthielemann b758c94 - IDEMPIERE-2259 Add SysConfig keys to avoid using hardcoded values on login page
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2259 status set to "Resolved" -assignee set to "Jan Thielemann" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2259
ocurieles_DCS: Good Morning
CarlosRuiz: Morning
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+0/-0/±1] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] globalqss 1ac024c - cleanup MSysConfig of unused keys - documenting missing entries in http://wiki.idempiere.org/en/System_Configurator_(Window_ID-50006)
allgood: good morning folks
CarlosRuiz: Hi Claudemir
allgood: did you received my pull request?
allgood: very small bug-fix!
allgood: noted that you associated my name with my nick! :-D
CarlosRuiz: all good :-)
CarlosRuiz: about the import translated references? I have that pending to review
allgood: no... there is one simpler
allgood: the associated references the pull request is no good
allgood: isn't good
allgood: IDEMPIERE-2273 - pull request #92
allgood: CSV importer have a call to field.isMandatory() on the preprocess routie
allgood: routine
allgood: but field is the field of the current selected record on GUI... so, isMandatory() can't do it's job at that time
allgood: did you associated my nick with my last name (since it is similar in spanish)
allgood: ?
CarlosRuiz: ah - Todo Bom :-)
allgood: yep... it really is my last name!
allgood: lot of jokes around here
allgood: and a lot of responsibility to keep up to it!
CarlosRuiz: :D
CarlosRuiz: you must be optimist all the time
allgood: I try to
nmicoud: Hi CarlosRuiz, do you think you will have time to review and integrate the following little tickets : 2279, 2275, 2271, 2260, 2089 ?
CarlosRuiz: I'll try
nmicoud: thanks
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+0/-0/±2] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] globalqss 24fecc3 - IDEMPIERE-2278 Entity Types are not cached - but static array
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2278 status set to "Resolved" -assignee set to "Carlos Antonio Ruiz Gomez" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2278
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2279 status set to "Resolved" -assignee set to "Nicolas Micoud" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2279
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+2/-0/±0] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] nmicoud 6617acb - IDEMPIERE-2279 C_ContactActivity SalesRep_ID has no AD_Reference_Value_ID
nmicoud: gtg, bye bye
CarlosRuiz: bye
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2275 status set to "Resolved" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2275
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+0/-0/±3] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] globalqss 2e4ff71 - IDEMPIERE-2275 Process button of toolbar doesn't handle access / based on patch from Nicolas Micoud (nmicoud)
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2089 Attachment set to "IDEMPIERE-2089_CR.patch"
Not-2f44: [IDEMPIERE] Hi [~nmicoud], can you please test the attached IDEMPIERE-2089_CR.patch It covers more points where the hardcoded window is defining behavior.
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2089
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2260 status set to "Resolved" -assignee set to "Nicolas Micoud" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2260
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to development [+2/-0/±2] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] globalqss ee261bb - IDEMPIERE-2271 iDempiere is hardcoded EMail to support subject / IDEMPIERE-2260 iDempiere is hardcoded in MClient.testEMail / based on patch from Nicolas Micoud (nmicoud)
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2271 status set to "Resolved" -assignee set to "Nicolas Micoud" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2271
norbertbede: hi all
CarlosRuiz: hi norbertbede
norbertbede: any news with ZK7 reported issues fixing ? im able to teste again :)
norbertbede: hi carlosruiz
CarlosRuiz: hieplq have fixed some
Not-2f44: [IDEMPIERE] norbert.bede created IDEMPIERE-2281 Alert Processor and Scheduler ignoring Client Storage Provider Setup
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2281
Not-2f44: [IDEMPIERE] norbert.bede created IDEMPIERE-2282 Print format footer - prints database server details - security issue
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2282
Not-2f44: [IDEMPIERE] norbert.bede created IDEMPIERE-2283 Print format - footer hardcoded idempiere instead installation name
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2283
allgood: hi folks, back again with some doubts
allgood: is there a way to remote debug an idempiere installation with eclipse?
CarlosRuiz: yes - you start the idempiere server adding debug as parameter
CarlosRuiz: idempiere-server.sh debug
CarlosRuiz: and then you can connect to port 4554
allgood: wow... it is already made...
allgood: i added some parameters to the command line, but eclipse doesn't obey the break points
CarlosRuiz: ah - is tricky - when you define the remote application you must choose a project
CarlosRuiz: I think just breakpoints within that project are caught
allgood: i can only debug one project?
CarlosRuiz: maybe there is another way - but I haven't researched
allgood: i am trying to look why my ACTIVE plugin isn't registering the document event handler... it happens only on the deployed installation.. on eclipse it works!
allgood: could it be because the service.ranking value?
CarlosRuiz: yes - that's necessary
allgood: but the problem would appear on eclipse also, right?
allgood: i am connecting to the ieventmanager service
allgood: looks like there is no service.ranking on the client side
Not-2f44: [IDEMPIERE] carlosruiz_globalqss created IDEMPIERE-2284 MProductCategory/MAssetGroup s_cache useless
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2284
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2284 status set to "Resolved" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2284
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 2 commits to development [+0/-0/±3] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] globalqss 788b5fb - IDEMPIERE-2116 Cannot correct Inventory OnHand ORA-01861 / thanks to David Garcia for the hint
Not-2f44: [iDempiere] globalqss 8e403d4 - IDEMPIERE-2284 MProductCategory/MAssetGroup s_cache useless
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to zk7 [+15/-16/±14] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] globalqss cc12bac - IDEMPIERE-2245 Platform Upgrade for r3
norbertbede: CarlosRuiz are you tested idempiere on amazon EC2/RDS ?
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2245
Not-2f44: [IDEMPIERE] https://bitbucket.org/idempiere/idempiere/commits/cc12bac Upgrade libraries: from postgresql-9.2-1004.jdbc4.jar to postgresql-9.3-1102.jdbc41.jar from c3p0-0.9.2.jar to c3p0-0.9.5-pre9.bin.zip from ojdbc6.jar to ojdbc7.jar from cron4j-2.2.1.jar to cron4j-2.2.5.zip from jcommon-1.0.16.jar to 1.0.23 from jfreechart-1.0.13.jar to 1.0.19 from super-csv-2.0.0-beta-1.jar to 2.2.0 from vt-password-3.1.1.jar to 3.1.2 from jasperreports-5.1.2.jar
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2245
Not-2f44: [IDEMPIERE] carlosruiz_globalqss updated IDEMPIERE-2116 status set to "Resolved" -assignee set to "Carlos Antonio Ruiz Gomez" -resolution set to "Fixed"
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2116
CarlosRuiz: I did basic tests to create the database - but not deep
CarlosRuiz: I think Chuck Boecking have tested RDS more
norbertbede: we are running live 8GB (large instance) EC2 and RDS large (8GB)
norbertbede: there are 3 clients running live 12 hours/day
norbertbede: generally i want to share our experiences
allgood: I have 2 classes that connects to the IEventHandler service, is there a limitation on this? should I use only one per bundle?
norbertbede: we running pgbadger and VisualVM for tune.
allgood: this is consequence of a refactor, we merged 2 bundles into 1
norbertbede: overall in comparision with a dell server with 16GB ram - running java+pg looks me worst response for enduser
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to zk7 [+8/-0/±19] https://bitbucket.org/idempiere/idempiere/commits/
norbertbede: usually when opening windows tabs etc
Not-2f44: [iDempiere] globalqss 1568d91 - Merge development (8e403d4382f8) into zk7 branch
Not-2f44: [iDempiere] CarlosRuiz_globalqss pushed 1 commit to zk7 [+0/-0/±2] https://bitbucket.org/idempiere/idempiere/commits/
Not-2f44: [iDempiere] globalqss c0b8256 - IDEMPIERE-2245 Platform Upgrade for r3 / drop the maxtime getting the files - it didn't allow to finish correctly in jenkins
Not-2f44: [IDEMPIERE] nmicoud updated IDEMPIERE-2089
Not-2f44: [IDEMPIERE] Hi Carlos, The patch is ok for me, thanks I think we could then open other tickets to manage other hardcoded windows this way (eg: DPActivitites or AMenu) WDYT ? Nicolas
Not-2f44: [IDEMPIERE] http://idempiere.atlassian.net/browse/IDEMPIERE-2089
mand1nga: norbertbede: how large is the database on that instance?
mand1nga: just curious about it
norbertbede: hi mand1nga
norbertbede: do you mean in mb ?
norbertbede: or bckup size
mand1nga: hello
norbertbede: backup size is approx 1,4GB
mand1nga: 1st option, backup might be compressed ;)
norbertbede: after i remove files stored in database
norbertbede: its an issue where e.g. in adempiere i found at customer 4GB just ad_attachements
CarlosRuiz: norbertbede, do you mean rds is slower than the dell?
norbertbede: im comparing now pgbadger reports
norbertbede: RDS looks OK
mand1nga: I believe it's a good practice to store ad_attachments on a dedicated tablespace, pointing to a different disk
norbertbede: we had some issues with indexes
norbertbede: but after proper indexing our CPU with 8GB runninh 20-40% on peaks
norbertbede: the worst is SAN networj
norbertbede: k
mand1nga: it's a nice setup though, looks like it can handle a much bigger database
norbertbede: means EC2 and RDS latency
norbertbede: on network
norbertbede: in last month RDS was updated to SSD disks
norbertbede: and each GB storage reserve for you 3 PIOPS
norbertbede: means in case 100GB you have reserverd 300 PIOPS/sec
norbertbede: its quite enough
norbertbede: max now is around 50
norbertbede: but ec2 instance looks me slow
norbertbede: instance type is m3large 8GB ram and 2 ecu
norbertbede: zk has slow responses unpredictable
norbertbede: that is im waiting Zk7 with new server instead tomcat
mand1nga: is it slow serving static files?
norbertbede: we are not using filesystem well. soon CPU and RAM
norbertbede: so Fs doenst make sense IMHO
mand1nga: I'm not thinking about hardware, more about server tuning, tomcat in this case
norbertbede: the problem is ec2 hardware behind is shared with others
norbertbede: that make unpredictable system
norbertbede: what experts recommend to reserve over 1 physical hardware
norbertbede: then you are the "boss" - then its eventually much morebetter
mand1nga: gotcha
norbertbede: cant imagine how can i tune better actual tomcat/osgi
norbertbede: we have setup 3GB for osgi in startup script
norbertbede: that is what we done also on dell
norbertbede: i have start also VisualVM for monitoring JAVA Machine inside
CarlosRuiz: what I'm using in a production installation is - production machine is a dedicated server with 64GB - and amazon instance is a contingency server (replicated via pg replication and attachments in fs replicated using rsync)
norbertbede: so we need to learn lot :)
norbertbede: CarlosRuiz but its dedicated not amazon aws
CarlosRuiz: yep - I use amazon as contingency - not for production
norbertbede: ah ok. i see
CarlosRuiz: in that installation
mand1nga: I see that opening a sales order window triggers about 50 http requests
norbertbede: amazon is very easy to setup and maintain
CarlosRuiz: is a big one - in my metrics I think the performance of a single dedicated against two amazon (one for DB and one for adempiere) is 3:1
norbertbede: i like that
CarlosRuiz: but in the previous installation we still didn't have RDS - probably with RDS can improve to 2:1
norbertbede: what you mean about setup running multiple clients as SaaS on 1 server :)
CarlosRuiz: still you cannot get with amazon the performance of a dedicated - I think is not feasible - and if you try to scale up the amazon server - then the cost becomes too high compared with a dedicated
norbertbede: yes costs are higher...
mand1nga: norbertbede: have you taken a look at https://tomcat.apache.org/articles/performance.pdf ?
mand1nga: that might make some difference, given the somewhat high amount of http requests, imho
norbertbede: CarlosRuiz do you mean jetty improve perfornace ?
norbertbede: thanks mand1nga
CarlosRuiz: it must - but maybe we need an unbiased report :-)
CarlosRuiz: look this page
CarlosRuiz: https://webtide.com/why-choose-jetty/
mand1nga: anytime
norbertbede: i met this site. that is why believe could help
norbertbede: jetty
mand1nga: sounds nice
norbertbede: maybe an indiscrete question.trek global running amazon or dedicated servers. multitenant ?
norbertbede: maybe its secret :)
norbertbede: just for curious
CarlosRuiz: OK, this is the plan - from the actual development branch I'm going to create release 2.1 - and 2.1 will become the next maintenance branch (2.0 will be closed for maintenance)
CarlosRuiz: after 2.1 released - then I'll integrate zk7 into development and it will become r3.0
CarlosRuiz: I created the templates for new features 2.1 wiki pages
CarlosRuiz: I would like to get community help to document the new features described in this category
CarlosRuiz: http://wiki.idempiere.org/en/Category:To_Be_Documented
CarlosRuiz: whatever you can contribute to
CarlosRuiz: I'll post this in forums too - and publish this also as weekly meeting
CarlosRuiz: gtg now - c u then