I have created OFBIZ-11156 for this issue

Jacques

Le 14/08/2019 à 14:42, Jacques Le Roux a écrit :
This is due to the Solr component

Checking what I can do

Jacques

Le 14/08/2019 à 12:51, Jacques Le Roux a écrit :
Hi Girish,

Yes I think I'll go this way indeed.

I tried to run R17 and R18 locally no issues. Even reverting changes for 
OFBIZ-11151 does not help

Jacques

Le 14/08/2019 à 12:40, Girish Vasmatkar a écrit :
Hi Jacques

Hooking up OFBiz with a profiler should help as to which method (eventually
leading to the dependency/API) is taking a lot of time helping narrowing
down the cause. Not sure if you want to go down this road, but it may help.

Best,
Girish

On Wed, Aug 14, 2019 at 3:56 PM Jacques Le Roux <
jacques.le.r...@les7arts.com> wrote:

This is due to logging, here it a sample from trunk demo log (to be sure
it's not on my machine):

2019-08-14 03:06:57,734 |main
|ConfigXMLReader               |I| controller loaded: 0.009s, 207 requests,
81 views in
file:/home/ofbizDemo/trunk/plugins/scrum/webapp/scrum/WEB-INF/controller.xml
2019-08-14 03:06:59,831 |main
|log                           |I| Logging initialized @54684ms to
org.eclipse.jetty.util.log.Slf4jLog 2019-08-14 03:07:00,052
|main
|config                        |W| Trusting all certificates configured
for Client@4da6d664[provider=null,keyStore=null,trustStore=null]

After my change in OFBIZ-11151 I tried to revert to log4j-api:2.11.2

But still got this locally after

2019-08-14 11:31:34,409 |main
|log                           |I| Logging initialized @269930ms to
org.eclipse.jetty.util.log.Slf4jLog

Since there is not this problem in stable demo, I reverted to
log4j-api:2.6.2 but got the same type of issue

I then ran R16 locally and there was no issue. So it's something else than
just log4j-api version.

I continue...

Jacques

Le 14/08/2019 à 11:08, Jacques Le Roux a écrit :
The issue I initially reported (stop after loading scrum controller)
also happens to me in console now, still investigating...
Jacques

Le 13/08/2019 à 18:57, Mathieu Lirzin a écrit :
Hello,

Jacques Le Roux <jacques.le.r...@les7arts.com> writes:

Yes it was much faster before (but before what???), as fast as running
it in console.
Seems like a good scenario for a ‘git bisect’. ;-)



Reply via email to