Curiosité de vie

Windows 10 - Améliorer la vitesse

noreply@blogger.com (Unknown) — Fri, 22 Nov 2019 21:21:00 +0100

Analyser les performances

Ouvre le gestionnaire des tâches: CTRL + MAJ + ECHAP

performances: vérifiez visuellement quel composant sature: CPU, Disque, Mémoire
Détails: lister processus en cours d'exécution: tri possible sur CPU ou Mémoire en décroissant
Historique des applications : tri sur temps processeur en décroissant
Démarrage : vérifier ldes services au démarrage. Désactiver en cas de doute

Disque dur

Analyse / Defrag: "Optimiser les lecteurs"
Cleanup: "Nettoyage de disque"
Régler l’apparence et les performances de Windows
- cocher Ajuster afin d’obtenir les meilleures performances

Applications imposées

Lister

Ouvrir PowerShell (en tant qu'administrateur)

 Get-AppxPackage | Format-List -Property Name

Exemple pour rechercher Facebook ou Mirosoft

 Get-AppxPackage -Name fa* | Format-List -Property Name
 Get-AppxPackage -Name Microsoft* | Format-List -Property Name

Désinstallation

Ouvrir PowerShell (en tant qu'administrateur)

 Get-AppxPackage *windowscommunicationsapps* | Remove-AppxPackage
 Get-AppxPackage *xboxapp* | Remove-AppxPackage
 Get-AppxPackage *officehub* | Remove-AppxPackage
 Get-AppxPackage *solitairecollection* | Remove-AppxPackage
 Get-AppxPackage *bingfinance* | Remove-AppxPackage
 Get-AppxPackage *bingnews* | Remove-AppxPackage
 Get-AppxPackage *bingsports* | Remove-AppxPackage
 Get-AppxPackage *bingweather* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.Office.OneNote* | Remove-AppxPackage
 Get-AppxPackage *Facebook.Facebook* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.WindowsMaps* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.WindowsCamera* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.People* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.Windows.Photos* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.BingTravel* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.BingFoodAndDrink* | Remove-AppxPackage
 Get-AppxPackage *Microsoft.FreshPaint* | Remove-AppxPackage

Réinstaller une application

Si par erreur vous avez supprimé une application utile, reprenez son nom dans la ligne suivante (ex. pour la première^):

Get-AppxPackage -AllUsers  -Name *windowscommunicationsapps* | Foreach {Add-AppxPackage -DisableDevelopmentMode -Register "$($_.InstallLocation)\AppXManifest.xml"}

Sources utilisées

GPG: minimal how to

noreply@blogger.com (Unknown) — Thu, 7 Jun 2018 13:20:00 +0200

GPG minimal how-to !

GPG is a free and easy way to encrypt and/or verify your exchanges.
This HowTo was done with "gpg (GnuPG) 1.4.22" a GNU GPLv3 tool.

This post includes GPG basic commands:
- generate your key
- send/receive encrypted content
- sign or verify a content
- export/import your secret key

on windows host, you could install and use git bash to follow this steps

Initialize GPG tool using 'list keys' ?

$ gpg --list-keys
gpg: répertoire « /home/osboxes/.gnupg » créé
gpg: nouveau fichier de configuration « /home/osboxes/.gnupg/dirmngr.conf » créé
gpg: nouveau fichier de configuration « /home/osboxes/.gnupg/gpg.conf » créé
gpg: le trousseau local « /home/osboxes/.gnupg/pubring.kbx » a été créé
gpg: /home/osboxes/.gnupg/trustdb.gpg : base de confiance créée

First time a gpg command is executed, gpg produces base gpg init: a ~/.gnupg directory

How to generate my key ?

$ gpg --gen-key

Keep default options, and answer question about real name, and email, and passphrase.

That's all.

Now you can list your key:










$ gpg --list-secret-keys

You could distribute your public key:

$ gpg --keyserver hkp://keyserver.ubuntu.com:80 --send-keys KEYIDHERE

How to send encrypted content ?

Search (or ask) your recipient KEYID using the following command:

$ gpg --keyserver hkp://keyserver.ubuntu.com:80 --search-keys robert

Receive the key (require KEYID)

$ gpg --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys KEYIDHERE

Encrypt file using a given recipient (require related key)

$ gpg -r jojo@yoyo.fr -o encrypted.gpg -e clear_content.txt

this command will encrypt "clear_content.txt" into "encrypted.gpg" using "jojo@yoyo.fr" key.
you could add multiple recipient by repeating -r option
you coul ascii encode by using -a option

Use your favorite email client to send encrypted file.

How to receive encrypted content from somebody

You will need the following requirements:

generate and distribute your own key (step above)
communicate about the way to encrypt content (step above) to you recepient

Use your favorite email client to receive encrypted file.
Decrypt received file using the following command:

$ gpg -o decrypted.txt -d encrypted.gpg

How to add detached signature?

You would like to prove that YOU send clear information (this command doesn't encrypt the message itself)

$ gpg -o doc.sig --clearsign doc

You will need to answer you passphrase.
This will add a clear signature from doc and generate doc.sig as result

How to verify message with detached signature?

You would like to prove clear information associated signature identity (this command doesn't decrypt the message itself)

Assume you message with detached signature is in a file called "doc.sig"

$ gpg -o doc -d doc.sig

How to backup your secret key?

export it (in a very very secure place)

$ gpg --list-secret-keys

$ gpg --export-secret-keys -a MYKEYID > username_gpg_export_secret.asc

import it

$ gpg --import -a username_gpg_export_secret.asc

"-a" means ascii export/import. You could omit this option to export/import binary.

Galaxy Tab 10.1: HowTo move to CyanogenMod 10

noreply@blogger.com (Unknown) — Sat, 28 Apr 2018 15:02:00 +0200

1) Install TWRP (from this tutorial)
(suppose you already have adb installed)

BOOT with VOL UP + POWER
go to DOWNLOAD on right with VOL-UP
(wait odin)
on PC
Start ODIN
In AP field select TWRP => twrp311-20171009-p4wifi.tar.md5
clic "start"
(tablet reboot after install) unconnect USB and close odin
BOOT with VOL UP + POWER
go to UPDATE on left with VOL-LOW

connect tablet with USB

2) check ADB

ADB should dectect

\GalaxyTab10_1>c:\Adb\adb.exe devices
List of devices attached
C4F125509CFF91F recovery

2) boot onto TWRP and push ROM files

\GalaxyTab10_1>C:\adb\adb.exe push cm-10-20130825-UNOFFICIAL-p4wifi.zip /tmp
1300 KB/s (160472620 bytes in 120.502s)

\GalaxyTab10_1>C:\adb\adb.exe push gapps-jb-20121011-signed.zip /tmp
1295 KB/s (92706064 bytes in 69.888s)

Push SuperUser file
\GalaxyTab10_1>C:\adb\adb.exe push superuser.zip /tmp
1325 KB/s (3922325 bytes in 2.889s)

Install ROM
Install Gapps
Install SuperUser
REBOOT

ROM Wiko Ridge Fab 4G - retour d'expérience

noreply@blogger.com (Unknown) — Sat, 23 Dec 2017 18:03:00 +0100

Wiko Ridge Fab 4 ROM: partage d'expérience

Après de nombreux problèmes avec la ROM officielle RIDGE FAB 4G_V14 (publicité sur le launcher ou clavier TouchPal, reboot intempestifs, blocages au démarrage, problèmes de détection de storage, desfois des problèmes avec la charge, et facebook..), et vu que nous ne sommes pas les seuls, nous décidons de changer de ROM.

Avec notre téléphone, nous n'avons pas trouvé de ROM non-officielle qui apporte satisfaction (notre exigence: DATA + GPS + CAMERA + STABLE): cf. Mini HowTo (B)

Avec la ROM officielle en supprimant TouchPal: cf. Mini HowTo (A).
Notre expérience: après 3 jours dans cette configuration, le téléphone est plutôt stable, il n'a fait que 2 reboot (non désirés) lors d'installations d'application. Après 20 jours, des problèmes de plantages d'applications google apparaissent. Très vite le téléphone ne boot plus.
Conclusion: changement de téléphone!

Wiko Ridge Fab 4 ROM: our feedback

Following multiples issues with official ROM (advertising from the laucher or TouchPal keyboard, unwanted reboots, startup hangs, storage detection issue, sometimes recharging issue or facebook issue..) and as we are not alone in this case, we decided to change ROM.

With our phone, we didn't found unofficial ROM which brig us satisfaction (DATA+GPS+CAMERA+STABLE): cf. Mini HowTo (B)

Other way: re-install official ROM and remove the TouchPal application: cf. Mini HowTo (A).
Our feedback: following 3 days in this configuration, phone seems stable. We just encounter 2 unwanted reboots while installing apps. Following 20 days, we encounter some hang issue with google app, and then few days after the phone were not able to boot anymore.

Conclusion: change the phone!

Resources

adb+fastboot ( eg. https://www.getdroidtips.com/how-to-install-adb-and-fastboot-on-windows/ )

adb doc: http://adbshell.com/commands/adb-devices

TWRP recovery for your phone (eg. https://www.getdroidtips.com/twrp-recovery-wiko-ridge-4g/#Download_Files )
Unofficial ROM (+patch) from XDA-Developers:

Nom	Experience in my case: keeping in mind that each phone have it's own hardware issue ;)
cyanogenmod-unofficial	Boot OK, Wifi OK, Camera OK but need to disable GPS Location to make it working. With GPS enabled, the phone hang when taping on application icon. Day2: the mobile is not able to find 2G/3G/4G ! :(
maxicm-5-1-1-unofficial	boot OK, wifi OK, but camera and GGMaps hangs the phone after fiew seconds!! :(
lineageos-13	(need to retest with gapps!) seems stable but there is some matter: no camera + batterie issue
resurrection-remix	not tested because of forum feedback: Mobile data doesn't work. Camera works but for a few seconds. Then device hangs.
SlimLP	not tested following ASAZING answer: "Mobile data and gps do not work"

Google Apps: http://opengapps.org/
SuperSU (v2.82 ou sup.)

HowTo pre-requisites

check adb is detecting your device:
(from mobile)

power + volume UP until vibration

(else try volume DOWN on some device)

"apply update from adb"

(from workstation)

  c:\adb\adb.exe devices

your device must be listed

HowTo: boot on TWRP recovery

(from workstation)

  c:\adb\fastboot.exe boot TWRP_L5320.img

  (command is now waiting the phone...)

(from mobile)

power + volume UP until vibration
"reboot to bootloader" (line 5)

Mini HowTo (A) re-install Official ROM from scratch onto WIko Ridge Fab 4G 5320 (with Snapdragon) and remove TouchPal

If you follow this steps, it's at your own risks. We can't be responsible if you brick your phone

download official ROM from support-fr.wikomobile.com
download SuperSU (v2.82 ou sup.)
unzip RIDGE_FAB_4G_LL_V14.zip and execute RIDGE FAB 4G_V14.exe, then follow the step described on the screen.
boot on TWRP

push SuperSu on the phone:

c:\adb\adb.exe push SuperSU-v2.82-201705271822.zip /tmp

install SuperSU using TWRP

reboot the phone
via Android parameters ("Langue et saisie"), move from Touchpal keyboard to Google keyboard.
install Désinstalleur App System(ROOT) (SuperSu required)
with this application, uninstall "TouchPal 2015" (and associated langage packs)!!

Mini HowTo (B) install Custom Unofficial ROM onto WIko Ridge Fab 4G 5320 (with Snapdragon)

If you follow this steps, it's at your own risks. We can't be responsible if you brick your phone

From following samples: (getdroidtips.com) lineage-os-13-wiko-ridge-fab-4g and (phonandroid.com) Root Ridge4G

What "patch" means ? (5320 only)

download ROM + patch
push ROM (ex. for L5510) on the phone using adb
push patch (ex. from 5320) on the phone using adb
onto TWRP :
install ROM (dont reboot)
(got back)
install patch (dont reboot)
(got back)
install GAPPS
reboot

Install ROM from TWRP:

boot on TWRP recovery (already described)
lets do a Backup ! (use menu option and follow instructions, check system and data partitions)
push your ROM (and patch) file(s) on the phone (adapt zip path+filename)

(from workstation for MaxiCM ROM)
c:\adb\adb.exe push MaxiCM-5.1-20170209-UNOFFICIAL-l5510.zip /sdcard
c:\adb\adb.exe push kernel_ASAZING-L5320.zip /sdcard
c:\adb\adb.exe push open_gapps-arm-5.1-super-20171224.zip /sdcard

(from workstation for Cyanogem ROM)
c:\adb\adb.exe push cm-12.1-20170211-UNOFFICIAL-l5510.zip /sdcard
c:\adb\adb.exe push kernel_ASAZING-L5320.zip /sdcard
c:\adb\adb.exe push open_gapps-arm-5.1-super-20171224.zip /sdcard

( optionnal ) if you need more space, remove older ROM

(from workstation)
c:\adb\adb.exe shell
cd /sdcard
ls
rm OLD_ROM_SAMPLE.zip
exit

HowTo flash ROM on the phone

(from phone / TWRP recovery)
- "wipe" data (fr: formatter)
  - "advanced Wipe"
  - check cache + System + data
  - swipe to wipe
- back back back
- "install" rom ("flash" with TWRP):
  - select pushed file: ROM
     - dont check zip signature and dont check reboot
     - swipe to confirm Flash
     - at the end, back
  - (option) select pushed file: PATCH (eg. kernel_ASAZING..)
  - dont check zip signature and dont check reboot
     - swipe to confirm Flash
  - select pushed file: GAPPS
  - dont check zip signature and dont check reboot
     - swipe to confirm Flash
- at the end, Wipe cache/dalvik, swipe to Wipe
- then "Reboot System"

EMUI 5.0 Android 7 fiche pratique

noreply@blogger.com (Unknown) — Wed, 18 Oct 2017 20:23:00 +0200

EMUI 5.0 Android 7 fiche pratique

Retrouver une application:
- sur l'accueil, glisser le doigt vers le bas et saisir le début du nom de l'application.

- dans la liste des résultats taper sur l'application pour la lancer ou bien sur l'icône position de droite pour localiser son raccourci.

Applications masquées
- pour voir les applications masquées il faut faire un zoom sur l'écran d'accueil (séparer 2 doigts collés à l'écran).

- pour ajouter ou retirer (démasquer) une application masquée, il faut taper sur le [+] après la dernière application masquée:

Cochez les applications pour les masquer ou les décocher pour démasquer et les retrouver sur l'accueil.

HowTo: Swig C to Java: functions that manage an array of int, simple or complex struct (eg. array of objects)

noreply@blogger.com (Unknown) — Fri, 17 Feb 2017 15:00:00 +0100

You would like to map a C code from your Java application and you have heard about Swig: this post is for you!

This post is a really simple example to understand how to manage (from Java application) a C array of complex structure.

After reading Swig 3 documentation and searching on the net, I've finally succeed to create a little poc on how to return a list of struct.

proof of concept sample includes 4 functions:

a function sumitems that accept an array of integer as parameter to calculate a sum (part of Swig3 documentation)
a function populateSampleItem that just write on a given simple structure
a function populateItems that update an existing array of struct
a function buildItems that create from scratch a result array of struct

source of this poc is available here : https://github.com/boly38/pocswig
this poc is widely inspirated from similar example from "Samuel Jacob's Weblog" post (thanks to him !)

First file to write is the C header poc.h:

/* (in) simple int array as parameter */
int sumitems(int *first, int nitems);

/* simple structure */
typedef struct MyItem_t {
      int  id;
      char *name;
} MyItem;

/* (in/out) simple struct as parameter (updated by the function) */
void populateSampleItem(MyItem *item);

/* array of structure */
typedef struct MyItems_t {
      int  count;       // elements count
      MyItem *elements; // array of MyItem
} MyItems;

/* (in/out) array of structure as parameter (updated by the function) */
void populateItems(MyItems *items);

/* (out) array of structure (generated by the function) */
MyItems *buildItems();

The you will then have to write the C implementation. Here is a sample poc.c:

#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include "poc.h"

int sumitems(int *first, int nitems) {
  int i, sum = 0;
  for (i = 0; i < nitems; i++) {
    sum += first[i];
  }
  return sum;
}

void populateSampleItem(MyItem *item) {
  item->id = 1234;
  item->name = strdup("getSampleItem");
}

void populateItems(MyItems *items) {
    int nb = items->count;
    items->elements = malloc(nb * sizeof(MyItem));
    for (int j=nb-1;j>=0;j--) {
        items->elements[j].id = j;
        char elementName[80];
        sprintf(elementName, "populateItems %d", j);
        items->elements[j].name = strdup(elementName);
    }
}

MyItems *buildItems() {
    printf("buildItems 14 elements");
    int nb = 14;
    MyItems *items= malloc(sizeof(MyItems));;
    items->count = nb;
    items->elements = malloc(nb * sizeof(MyItem));
    for (int j=nb-1;j>=0;j--) {
        items->elements[j].id = j;
        char elementName[80];
        sprintf(elementName, "buildItems %d", j);
        items->elements[j].name = strdup(elementName);
    }
    return items;
}

And now to access this function from java, you will have to use Swig.

Swig use a specification file to setup how to map function/types/etc... This file is a .i file. Here is the file poc.i:

%module swigpoc

%include "arrays_java.i";
%apply int[] {int *};

%{
 #include "poc.h"
%}

%include "poc.h";
%extend MyItems_t{
  MyItem * getElement(int i) {
      return &$self->elements[i];
  }
}

Line 1 define the module name,
Line 3&4 define how to handle int array using swig facility,
Line 6 to 8 to tell to Swig to output include line into the target wrapper file.
Line 10 reuse as is the header file as specification (Swig MUST wrap all header file methods and structs to Java).
Line 11 to 15 to tell to Swig to append an extra function to help Java user to access to array element.

Now you will have to generate Java files! Use Swig :

rm -f *Item.java *.o *.dll swigpoc*
swig -java poc.i

You could look at your directory, there is some new C and Java files: poc_wrap.c, MyItem.java, MyItems.java, swigpoc.java swigpocJNI.java Next step is to build up the DLL (shared library) (exemple under Cygwin):

#!/bin/bash
JAVA_HOME=/cygdrive/c//Programmes/Java/jdk1.8.0_112/
INCLUDES="-I$JAVA_HOME/include/ -I$JAVA_HOME/include/win32/"

x86_64-w64-mingw32-gcc.exe -c poc.c poc_wrap.c $INCLUDES
x86_64-w64-mingw32-gcc.exe $INCLUDES -shared -o poc.dll poc_wrap.o poc.o

Now you can play with your new library from Java; exemple PocExample.java:

public class PocExample {

  static {
    System.out.println("load poc ...");
    System.loadLibrary("poc");
    System.out.println("load poc ... OK ");
  }

  public static void  main(String args[]) {
    System.out.println("poc");

    System.out.println("sumitems:");
    int[] arrayB = new int[10000000];          // Array of 10-million integers
    for (int i=0; i<arrayB.length; i++) {      // Set some values
      arrayB[i] = i;
    }
    int sum = swigpoc.sumitems(arrayB, 10000);
    System.out.println("SumB = " + sum);


    System.out.println("MyItem:");
    MyItem myt = new MyItem();
    swigpoc.populateSampleItem(myt);
    System.out.println("myt.name = " + myt.getName());

    MyItems myts = new MyItems();
    myts.setCount(10);
    swigpoc.populateItems(myts);

    for (int j=0; j<myts.getCount(); j++) {
      System.out.println(String.format("myts.element[%d].name = '%s'",j, myts.getElement(j).getName()));
      System.out.println(String.format("myts.element[%d].id = '%s'"  ,j, myts.getElement(j).getId()));
    }


    MyItems rez = swigpoc.buildItems();
    for (int k=0; k<rez.getCount(); k++) {
      System.out.println(String.format("rez.element[%d].name = '%s'",k, rez.getElement(k).getName()));
      System.out.println(String.format("rez.element[%d].id = '%s'"  ,k, rez.getElement(k).getId()));
    }
  }
}

I let you execute that:

1
2
3

#!/bin/bash
JAVA_HOME=/cygdrive/c//Programmes/Java/jdk1.8.0_112/
$JAVA_HOME/bin/javac *.java && $JAVA_HOME/bin/java PocExample

If you see something wrong, please tell me. Else hopes this helps!

Yet another MUSE Best-Of

noreply@blogger.com (Unknown) — Wed, 31 Aug 2016 14:14:00 +0200

Let me introduce you to the best-of MUSE songs :p

Muse - 2006 - Starlight [Official Music Video]

Muse - 2009 - Uprising [Official Video]

Muse - 2010 - Hysteria [Official Music Video]

Muse - 2010 - New Born

Muse - 2010 - Plug In Baby

Muse - 2010 - Sunburn

Muse - 2010 - Time Is Running Out (video)
Muse - 2012 - Resistance

Muse - 2013 - Supremacy (Official Video)

Muse - 2015 - Dead Inside [Official Music Video]

Muse - 2015 - Reapers [Official Lyric Video]

Muse - 2015 - The Handler [Official Lyric Video]

Concerts

Muse Live - 2000 - Presqu'île de Malsaucy - Eurockéennes (Belfort)

Muse Live - 2015 - Main Square Festival

NB: year is youtube video publication date. With helps of thetoptens

How to Jenkins : chain jobs with parameters

noreply@blogger.com (Unknown) — Fri, 5 Feb 2016 12:07:00 +0100

This post describes How To chain jobs with parameter(s).
- Choose optionA if you're using pipeline plugin.
- OptionB is a little bit deprecated, but usefull if you would like to script your build execution or trigger it remotely.

(option A) Define a post action in first job

First recommended and simple way if your are using "Build Pipeline". Go to the first job configuration to append a post-action :

(option B) Chain it using shell scripts

NB: this option will make loosing your pipeline chain.
I recommend to use 2 small dedicated temp jobs to test it first. That would avoid to run unnecessarely long build...

Steps :

Install Build Token Root Plugin
Define first job with parameter
Define second job ("follower") with parameter
Define a trigger for the follower
Append a shell script to chain the two jobs
Try!

Install Build Token Root Plugin

Ask to your Jenkins administrator to install this plugin : Build Token Root Plugin
Wait that Jenkins reboot (required).

Define first job with parameter

Add "branch" parameter to your first job.

Add a shell script to echo the variable.

Define follower job with parameter

Add "branch" parameter to your second job (same as previously).
Add a shell script to echo the variable(same as previously).

Define a trigger for the follower

Add a trigger for the second job. Set a secret token here.

Append a shell script to chain the two jobs

Open the first job configuration to add the following shell script.
You will need to adapt the jenkins host value corresponding to your environment.

# show job parameter
echo branch=${branch}

#  resources
#  https://wiki.jenkins-ci.org/display/JENKINS/Parameterized+Build
#  https://wiki.jenkins-ci.org/display/JENKINS/Build+Token+Root+Plugin
#  http://curiositedevie.blogspot.nl/2016/02/how-to-jenkins-chain-jobs-with.html
export JENKINS_HOST=https://myjenkins.domain.net
export JOB_TOKEN=THISISWONDERFULL
export JOB_NAME=git_config_follower
export JOB_PARAM1=branch=${branch}
export JOB_CAUSE=Launching_follower_with_branch_set_to_${branch}
export JOB_PING="$JENKINS_HOST/buildByToken/buildWithParameters?job=$JOB_NAME&token=$JOB_TOKEN&$JOB_PARAM1&cause=$JOB_CAUSE"
echo $JOB_PING
curl --verbose --insecure $JOB_PING

Try!

Launch the first job with a parameter.

Go to the Job console to check curl http status.

Go to the follower (second) job logs.

Check the follower result!

Log4j 1.x with maven: minimal howto

noreply@blogger.com (Unknown) — Thu, 7 Jan 2016 12:02:00 +0100

This post is a minimal howto for log4j version 1.

What a simple task ! but who never search a simple sample on the web send a little stone to me ;)

Feel free to copy paste anyway ;))

What are the dependencies

You will need to add log4j, slf4j and lombok (to use @Slf4j) :

 <properties>
  <!--General project configuration -->
  <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
  <project.reporting.outputEncoding>UTF-8</project.reporting.outputEncoding>

  <lombok.version>1.16.6</lombok.version>
  <log4j.version>1.2.17</log4j.version>
  <slf4j.version>1.7.10</slf4j.version>
  <junit.version>4.12</junit.version>
 </properties>
 <dependencyManagement>
  <dependencies>
   <!-- @Slf4j annotation -->
   <dependency>
    <groupId>org.projectlombok</groupId>
    <artifactId>lombok</artifactId>
    <version>${lombok.version}</version>
    <scope>provided</scope>
   </dependency>
   <!-- ## LOGS ## -->
   <dependency>
    <groupId>log4j</groupId>
    <artifactId>log4j</artifactId>
    <version>${log4j.version}</version>
   </dependency>
   <dependency>
    <groupId>org.slf4j</groupId>
    <artifactId>slf4j-api</artifactId>
    <version>${slf4j.version}</version>
   </dependency>
   <dependency>
    <groupId>org.slf4j</groupId>
    <artifactId>slf4j-log4j12</artifactId>
    <version>${slf4j.version}</version>
   </dependency>

How to use it

Simply add "@Slf4j" to your class, and then you could use "log" as logger.

import java.security.InvalidParameterException;

import lombok.extern.slf4j.Slf4j;

@Slf4j
public class MyClass {
    public MyClass() {
        log.debug("constructor YEP");
    }
    
    public void doIt(long value) {
        log.info("doIt {}", value);
        try {
            throw new InvalidParameterException("Zobby");
        } catch (Exception ee) {
            log.error("I'm so stupid: {} => {}", ee.getClass().getSimpleName(), ee.getMessage(), ee);
            log.warn("same without stack trace !: {} => {}", ee.getClass().getSimpleName(), ee.getMessage());
        }
    }
}

How to configure appenders

To configure log4j, add a log4j.properties file to your resources. Start by defining a console appender. This appender should appear in the rootLogger definition too. You could define the following content to start:

log4j.rootLogger=DEBUG, console

## Console appender 
log4j.appender.console=org.apache.log4j.ConsoleAppender
log4j.appender.console.threshold=DEBUG
log4j.appender.console.layout=org.apache.log4j.PatternLayout
log4j.appender.console.layout.ConversionPattern=%d %5p [%c{1}] %m%n

# package log levels
#
# Log4j bootstrap
#
log4j.logger.org.apache.http = INFO
log4j.logger.com.myapp = DEBUG
#

How to configure file appender

Update your rootLoger to refer to the logfile appender, and add a logfile appender definition:

log4j.rootLogger=DEBUG, console, logfile

## File appender
log4j.appender.logfile=org.apache.log4j.RollingFileAppender
log4j.appender.logfile.threshold=DEBUG
log4j.appender.logfile.layout=org.apache.log4j.PatternLayout
log4j.appender.logfile.layout.ConversionPattern=%d{ISO8601} - %-5.5p - %t - %c - %m%n
log4j.appender.logfile.file=MY_CUSTOMFILEHERE.log
log4j.appender.logfile.maxBackupIndex=5
log4j.appender.logfile.maxFileSize=2048KB
log4j.appender.logfile.append=true

How to configure graylog (splunk-like) appender

Update your rootLoger to refer to the graylog appender, and add a gelf appender definition:

### Graylog appender
log4j.appender.graylog2=org.graylog2.log.GelfAppender
log4j.appender.graylog2.graylogHost=tcp:mygraylogserver.net
log4j.appender.graylog2.graylogPort = 12201
#log4j.appender.graylog2.originHost=(default to the local hostname)
log4j.appender.graylog2.facility=gelf-java
log4j.appender.graylog2.layout=org.apache.log4j.PatternLayout
log4j.appender.graylog2.extractStacktrace=true
log4j.appender.graylog2.addExtendedInformation=true
log4j.appender.graylog2.additionalFields={'environment': '${projectEnvironment}', 'application': 'MyPoc','version': '${buildVersion}_${buildTimestamp}'}

How to configure NT event appender

Pre-requisites : add NTEventLogAppender.dll and NTEventLogAppender.amd64.dll to the current directory or to the java ld library path. You could append "-Djava.library.path=PATH" to your application if needed. Get this dll files from log4j zip distribution ^{doc : https://wiki.apache.org/logging-log4j/NTEventLogAppender}

Update your rootLoger to refer to the ntappender appender, and add a new ntappender definition:

# NTEventLogAppender
log4j.appender.ntappender=org.apache.log4j.nt.NTEventLogAppender
log4j.appender.ntappender.Source=MyKillerApp
log4j.appender.ntappender.layout=org.apache.log4j.PatternLayout
log4j.appender.ntappender.layout.ConversionPattern=%d{ISO8601} - %-5.5p - %t - %c - %m%n

For this last usecase you will be able to see the app logs into Windows Event Log service :

Swagger UI + Auth0 Bearer

noreply@blogger.com (Unknown) — Wed, 6 Jan 2016 10:14:00 +0100

When you would like to integrate Swagger UI with a token based authentication (like Auth0), this is the way you need to update Swagger index.html :

      function addApiKeyAuthorization(){
        var key = encodeURIComponent($('#input_apiKey')[0].value);
        if(key && key.trim() != "") {
            var apiKeyAuth = new SwaggerClient.ApiKeyAuthorization("Authorization", "Bearer " + key, "header");
            window.swaggerUi.api.clientAuthorizations.add("key", apiKeyAuth);
            log("added key " + key);
        }
      }

Instead of adding api_key as part of the query, this code is adding your token in an "Authorization" header and with "Bearer " as prefix.

enjoy

GMail: gestion de vos emails aux petits oignons

noreply@blogger.com (Unknown) — Thu, 6 Aug 2015 15:14:00 +0200

Lorsque vous vous abonnez à une liste de diffusion / mailing-liste (exemple: archive des ML Apache ) ou alors lorsque vous recevez des mails automatisés (exemple: résultats de build Jenkins);

ce que vous aimez c'est :

recevoir les mails en temps réel ou selon vos préférences sur vos différents supports clients mails,
(ML) pouvoir y répondre de suite.

ce que vous détestez c'est :

avoir à classer/archiver ces emails,
avoir à supprimer les anciens mails inutiles (redondants avec les archives des ML ou avec votre dashboard Jenkins).

L'idéal serait de souscrire aux ML sans en avoir les inconvénients. Peut être des services en lignes ou d'autres méthodes existent (ex. souscription à un RSS) ; certains même me diront mais pourquoi classer ses mails :), .. mais personnellement j'ai décidé d'opter pour l'automatisation de la gestion de ce type de mails

Première étape : appliquer un label aux emails reçus : je ne vais pas faire l'affront de décrire cette étape car je pense que l'aide GMail est auto suffisante.

Seconde étape : utiliser un script google script pour automatiser la gestion de ces mails. Pour cette étape, rendez vous sur cette page de code. Vous y trouverez le code à copier-coller sur votre google script. Le reste n'est que paramétrage. Ce script utilise notamment les filtres avancés pour déterminer les fils de discussion (thread;groupe de mails) utiles.

Le code est assez lisible en l'état.

Dernière étape : planifier l'exécution du script selon vos besoins. L'aide en ligne de google script est très bien structurée.

Personnellement avec un filtre de réception, je positionne mes mails automatisés sous un label fils du label "ML".

Puis avec ce code google script, je peux archiver automatiquement les mails de ML vieux de 2 jours et supprimer automatiquement les mails archivés de ML vieux de 60 jours minimum et n'ayant pas de label "keep".

Et vous comment faites vous ? des remarques positives/négatives sur ce type de gestion ..?

Microsoft Windows: minimal howto

noreply@blogger.com (Unknown) — Wed, 3 Jun 2015 17:19:00 +0200

This post is a kind of minimal 'how to' for Micro$oft Windows.

Windows : basic command (to execute from [WINDOWS + R])

cmd             : classic console
powershell      : power shell command console
msinfo32        : host info
devmgmt.msc     : devices management (fr: Gestionnaire de périphériques)
certmgr.msc     : certificates managers
services.msc    : windows OS services
javaws -viewer  : windows java services
eventvwr.exe    : Event viewer
shell:Startup   : show startup programs (if any)
procmon.exe     : (download) show process actiity (really useful)
compmgmt.msc    : Computer Management

NB: for ssl cf ssl howto

god mode (Windows 10)

create a folder on windows 10 desktop and name it

GodMode.{ED7BA470-8E54-465E-825C-99712043E01C}

network related commands

list all LISTENING ports

netstat -ano

netstat -ano | find "LIST"

find an application from port^src

netstat -ano | find "LIST"
==> Pick the wanted process id from the last column. Example pid "8112"
tasklist  | find "8112"
tasklist /svc | find "8112"

DNS lookup

nslookup www.google.fr

nslookup
server 8.8.8.8
www.google.fr

other usefull common tools

FileZilla (secure FTP)
Notepad++ (light editor)
Cygwin (unix like shell)
Process Hacker 2 (improved task list manager)

ssl certificates and truststore: minimal howto

noreply@blogger.com (Unknown) — Tue, 27 Jan 2015 16:21:00 +0100

This post is a kind of minimal 'how to' for SSL, certificates and truststore :

Tomcat HTTPS application : how to use a valid and trusted self-signed certificate for localhost
SSL Client : "PKIX path building failed" error

Tomcat HTTPS application : how to use a valid and trusted self-signed certificate for localhost

Context

As WebDevelopper, I use Tomcat and I need to work with localhost with a secure web application. The issue with the latest browser update (ex. ggchrome) is that an invalid or untrusted self signed certificate could block the navigation or AJAX exchanges.
For example, you could get the folloowing error :

NET::ERR_CERT_AUTHORITY_INVALID
:: certificate validation chain is not trusted
ERR_INSECURE_RESPONSE
:: unable to trust a server answer

to workaround this issue, you could follow these steps:

generate a self-signed certificate for localhost
tell to tomcat to use this certificate
append this certificate to your workstation certificates manager

Generate a self signed certificate

$ keytool -genkey -keyalg RSA -alias tomcat \
  -keystore $HOME/.keystore -storepass changeit -validity 360 -keysize 2048

Warning: answer "localhost" to the first question

more documentation: tomcat8 ssl howto - sslshopper create self signed cert

Tomcat ssl configuration

Connector's attributes example from server.xml:

SSLEnabled="true" clientAuth="false" 
 keystoreFile="${user.home}/.keystore" keystorePass="changeit" 
 maxThreads="150" port="8443" 
 protocol="org.apache.coyote.http11.Http11NioProtocol" 
 scheme="https" secure="true" sslProtocol="TLS"

Append this certificate to your workstation certificates manager

navigate to your tomcat application ; example : https://localhost:8443
(example using google chrome), right click on address bar lock / click on "certificate informations (...)" link / "details" tab
Choose "copy certificate into file" - keep default format X509 DER.
right click on the just created file to "Install certificate"
Select the following target : "Trusted root certification authorities"("Autorités de certification racines de confiance")
You could verify the certificate installation : Windows Start / Execute / certmgr.msc ; this will open Windows certificates manager
Restart your browser (use CTRL + ALT + Q for Google Chrome instead of closing the window).
navigate to your tomcat application ; example : https://localhost:8443 : your certificate should be trusted now

SSL Client : "PKIX path building failed" error

Java error

javax.net.ssl.SSLHandshakeException: 
sun.security.validator.ValidatorException: PKIX path building failed: 
sun.security.provider.certpath.SunCertPathBuilderException: unable to find valid certification path to requested target

Java workarounds

(bad and quick way) Disable all ssl check by configuration cf. UnsafeSSLHelper (from javabox github project)
(right way) Update the ssl verification chain. For example by importing new valid certificate(s) to your truststore.

Windows

Windows / List certificates:

Start / Execute / certmgr.msc

Google Chrome / List certificates:

go to Parameters, the search "ssl" (chrome://settings/search#ssl)

Import a certificate in a truststore file:

Make a backup :

cp %JAVA_HOME%/jre/lib/security/cacerts %JAVA_HOME%/jre/lib/security/cacerts.orig

Import a certificate :

keytool -import -alias MyCert -keystore %JAVA_HOME%/jre/lib/security/cacerts \
     -trustcacerts -file MyCert.cer

(cf. commandes+keytool ...)

Implort a certificate in JDK truststore:
create a little batch like this one :

set JAVA_HOME=C:\Program Files\Java\jdk1.8.0_65
set PATH=%JAVA_HOME%\bin;%PATH%
REM set JRE_CACERTS=%JAVA_HOME%\lib\security\cacerts
set JDK_CACERTS=%JAVA_HOME%\jre\lib\security\cacerts
set PASSWORD=changeit
set DIR=certsdir

for %%f in (%DIR%\*.cer) do keytool -keystore "%JRE_CACERTS%" -storepass %PASSWORD% -noprompt -importcert -alias "%%~nf" -file %%f
for %%f in (%DIR%\*.cer) do keytool -keystore "%JDK_CACERTS%" -storepass %PASSWORD% -noprompt -importcert -alias "%%~nf" -file %%f

This script install all certificates under certsdir to the JDK (or JRE).

shared libraries "lib.so and lib.dll" file: minimal how to

noreply@blogger.com (Unknown) — Mon, 1 Dec 2014 15:31:00 +0100

This post is a kind of minimal 'how to' for the shared libraries : ex. under linux mylib.so or under windows my.dll

Windows

List library exported symbols (then functions):
- Tool : Dll Export Viewer (nirsoft.net)
- (visualstudio required) dumpbin /exports my.dll [not tested]

Linux

List library exported symbols (then functions)

readelf -s /usr/lib/libspreadsheet.so |grep workbook_sheet

nm -D /usr/lib/libspreadsheet.so |grep workbook_sheet

objdump -T /usr/lib/libspreadsheet.so |grep workbook_sheet

List library dependencies

readelf -d /usr/lib/libspreadsheet.so|grep NEEDED

prosyst OSGi: minimal how to

noreply@blogger.com (Unknown) — Wed, 26 Nov 2014 15:39:00 +0100

This post is a kind of minimal 'how to' for Prosyst OSGi.

console (or telnet) commands

(telnet only) toggle command to get stdout / stderr on the telnet output

dump

Bundle installation

install mybundle.jar

Bundle start

start mybundle.jar

Bundle installation and start alias

i -S mybundle.jar

Bundle restart alias

rs mybundle.jar

rs 42

Bundle update := uninstall, install, start

update mybundle.jar

update 42

whole prosyst OSGi framework restart (you will loose the telnet connection)

rs 0

for more information : runtime_console 7.5 doc

npm et bower: fiche pratique

noreply@blogger.com (Unknown) — Fri, 7 Nov 2014 09:40:00 +0100

Cet article est une fiche pratique (commandes de base) pour npm et Bower package managers.

Commandes utiles pour configurer NPM:

npm config set proxy http://proxy:8080
npm config set https-proxy http://proxy:8080
npm config set strict-ssl false

Installation

npm install -g bower

Bower derrière un proxy : utiliser les variables d'environnement http_proxy et https_proxy ou le fichier .bowerrc

Installation d'une dépendance

bower install jquery

Mise à jour d'une dépendance

bower update jquery

Info et version d'une dépendance

bower info jquery

Rechercher une dépendance: via la commande ci-dessous ou via le site officiel http://bower.io/search/

bower search angular%bootstrap%

Bower repose sur un fichier de configuration bower.json dont voici un exemple:

{
  "name": "myproject-webgui",
  "version": "0.0.0",
  "dependencies": {
    "angular": "1.3.1",
    "json3": "~3.2.6",
    "es5-shim": "~2.1.0",
    "jquery": "~1.11.0",
    "bootstrap": "~3.0.3",
    "angular-resource": "1.3.1",
    "angular-cookies": "1.3.1",
    "angular-sanitize": "1.3.1",
    "angular-route": "1.3.1",
    "ng-table": "0.3.3"
  },
  "devDependencies": {
    "angular-mocks": "1.3.1",
    "angular-scenario": "1.3.1"
  }
}

git : fiche pratique

noreply@blogger.com (Unknown) — Fri, 3 Oct 2014 14:43:00 +0200

Cet article est une fiche pratique (commandes de base) pour Git.

Client Windows : perso j'utilise Msysgit. Il y en a plein évidemment...

Git derrière un proxy ? 2 solutions :

 export http_proxy=http://user:pass@host:port

 git config --global http.proxy http://user:pass@host:port

Un problème de certificats SSL ?

 git config --global http.sslVerify false

Visualiser l'historique d'un dépôt (avec msysgit)

 gitk --all &

récupérer un dépôt distant

 git clone https://domain.com/repo.git

connaître l'état du dépot

 git status

indexer l'ensemble des modifications : supprimer les fichiers, ajouter les fichiers manquants ou modifiés

 git add --all .

annuler des modifications locales non commitées

 git reset --hard

résoudre un conflit (lors d'un merge ou cherry-pick..)

 git mergetool

creer un tag

 git tag -a monTag -m 'monTag Il est beau'

 git push --tags

déplacer un tag sur la version actuelle

 git tag -f -a monTag

 git push origin :refs/tags/monTag

 git push --tags

positionner des alias de commandes git : dans le fichier de config global (~/.gitconfig)

 [alias]
        #Basic
        co = checkout
        st = status -s
        ci = commit
        br = branch
        df = diff
        cp = cherry-pick
        cl = clone
        diff = diff --word-diff
        dc = diff --cached
        who = shortlog -s --
        r  = !git ls-files -z --deleted | xargs -0 git rm
        #revert last commit
        rollback = reset --soft HEAD~1
        #Commit using previous message
        reapply = commit -c ORIG_HEAD
        #Logs
        lg = log --graph --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset %an - %s %Cgreen(%cr)%Creset' --abbrev-commit --date=relative
        #Les logs du user
        mylg = log --graph --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset %an - %s %Cgreen(%cr)%Creset' --abbrev-commit --date=relative --author="Vincent\\ Bostoen"
        ll = log --pretty=format:"%C(yellow)%h%Cred%d\\ %Creset%s%Cblue\\ [%cn]" --decorate --numstat
        lol = log --graph --decorate --pretty=oneline --abbrev-commit
        lola = log --graph --decorate --pretty=oneline --abbrev-commit --all
        #Lister les commits pas encore pushés
        ctp = log --graph --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset %an - %s %Cgreen(%cr)%Creset' --abbrev-commit --date=relative origin/master..master
        #Mettre un fichier en argument pour avoir ses modifications
        fl = log -u
        #Avoir le contenu du dernier commit
        dl = "!git ll -1"
        #Avoir le diff avec un commit
        diffr = "!f() { git diff "$1"^.."$1"; }; f"
        #Lister tous les chemins pour un nom de fichier
        f = "!git ls-files | grep -i"
        #Chercher tous les fichiers contenant une chaine de caractère
        gr = grep -Ii
        #Lister les alias
        la = "!git config -l | grep alias | cut -c 7-"
        #Remove all locally deleted file (git rm on all 'D' files)
        rma = "!git diff -z --name-only --diff-filter=D | xargs -0 git rm"
        #Stash
        sl = stash list
        sa = stash apply
        ss = stash save
        amend = !git log -n 1 --pretty=tformat:%s%n%n%b | git commit -F - --amend
        ignore=!([ ! -e .gitignore ] && touch .gitignore) | echo $1 >>.gitignore
        up = !sh -c 'git pull --rebase --prune && git log --pretty=format:\"%Cred%ae %Creset- %C(yellow)%s %Creset(%ar)\" HEAD@{1}..'
        bt = for-each-ref --format='%(refname:short) <- heads="" pre="" refs="" upstream:short="">

gradle: fiche pratique

noreply@blogger.com (Unknown) — Wed, 1 Oct 2014 17:01:00 +0200

Cet article est une fiche pratique pour Gradle.

Depuis Windows

Derrière un proxy : éditer gradlew.bat et modifier la ligne suivante pour positionner votre proxy

set DEFAULT_JVM_OPTS=-Dhttp.proxyHost=myproxy.com -Dhttp.proxyPort=8080

liste des tâches

./gradlew.bat tasks

build

./gradlew.bat build

build sans test (traduire: exécuter la tâche 'build' sans exécuter la tâche 'test')

./gradlew.bat build -x test

build sans test ni javadoc

./gradlew.bat build -x test -x javadoc

heroku : fiche pratique

noreply@blogger.com (Unknown) — Fri, 19 Sep 2014 16:26:00 +0200

Cet article est une fiche pratique pour Heroku.

Depuis Windows

Pour les commandes heroku^download, je conseille d'installer et d'utiliser une console msysgit

Connexion

Positionner un proxy (windows)

set http_proxy=http://myproxy:8080
set https_proxy=http://myproxy:8080

Positionner un proxy (nux)

export http_proxy=http://myproxy:8080
export https_proxy=http://myproxy:8080

Pour utiliser une clé SSH : démarrer ssh-agent.exe puis utiliser ssh-add.exe
Pour se connecter ^doc

heroku login

Use "json-schema-validator" as an OSGi bundle

noreply@blogger.com (Unknown) — Tue, 12 Aug 2014 11:38:00 +0200

Hello

This post describes how to integrate json-schema-validator 2.2.5 in an OSGi container.

json-schema-validator home page redirect to the binary download site to get OSGi ready library. But with this last one I was unable to make json-schema-validator working... I got an unsatisfied package jackson-databind.

So I decide to integrate by hand json-schema-validator and related dependencies. 14 jar to get and check (see into the MANIFEST.MF the "Export-Packages" directive..)

Here is the way :

use json-schema-validator in your Eclipse Java project by adding the pom dependency.
make a sample unit test working (see example package of the json-schema-validator project).
open your project "pom.xml" file using eclipse to know the current dependency tree (view called "Dependency Hierarchy").

After getting one-by-one each mandatory dependendency of the json validator, here is the final list I get :

btf-1.2.jar
jackson-annotations-2.4.1.jar
jackson-core-2.4.1.1.jar
msg-simple-1.1.jar
jackson-databind-2.4.1.3.jar
guava-16.0.1.jar
uri-template-0.9.jar
jackson-coreutils-1.8.jar
rhino-1.7R4.jar
libphonenumber-6.0.jar
mailapi-1.4.3.jar
jopt-simple-4.6.jar
joda-time-2.3.jar
json-schema-core-1.2.4.jar
json-schema-validator-2.2.5.jar

You can get each jar from a given maven repository; example joda-time-2.3.jar.

For each jar, I have to check the OSGi Export package directive. Here is the way to do that :

open the jar file using zip tool (ex. 7zip)
edit the file "META-INF/MANIFEST.MF"
check that the file contains an Export package directive like the following example for yoda time :

Export-Package: org.joda.time;version=2.3,org.joda.time.base;version=2
 .3,org.joda.time.chrono;version=2.3,org.joda.time.convert;version=2.3
 ,org.joda.time.field;version=2.3,org.joda.time.format;version=2.3,org
 .joda.time.tz;version=2.3

If this is not the case, I had to add it by hand. I had to do that for the following library : rhino, libphonenumbers, joptsimple. See under the related export package :
- rhino-1.7R4.jar
- get an updated version here
- libphonenumber-6.0.jar
- EDIT: it's now useless to do this update since this jar is embedded in the bundle's project (cf. under)
- related issue : code.google.com libphonenumber issue #205
- joptsimple
```
Export-Package: joptsimple;version="4.6"
```
- EDIT: it's now useless to do this update since issue #65 has been fixed; So you could use a recent version : jopt-simple-4.8-SNAPSHOT.jar

After all these steps, when I want to validate a json content with a schema, I got an error while loading meta-schema "/draftv4/schema" (problem of class loading).
Until now, I have not figured out how to fix it.

java.lang.ExceptionInInitializerError
        at com.github.fge.jsonschema.SchemaVersion.(SchemaVersion.java:65)
        at com.github.fge.jsonschema.SchemaVersion.(SchemaVersion.java:44)
        at com.github.fge.jsonschema.core.load.configuration.LoadingConfigurationBuilder.(LoadingConfigurationBuilder.java:
17)
        at com.github.fge.jsonschema.core.load.configuration.LoadingConfiguration.byDefault(LoadingConfiguration.java:151)
(...)
Caused by: java.io.IOException: resource /draftv4/schema not found
        at com.github.fge.jackson.JsonLoader.fromResource(JsonLoader.java:91)
        at com.github.fge.jsonschema.SchemaVersion.(SchemaVersion.java:63)

This file is present into "json-schema-core-1.2.4.jar" but not loaded as resource by OSGi container...

Following this error, I just open an issue json-schema-validator issue #111.

As workaround, with the help of Gábor Bakos and as first draft, I just push a "OSGi'fyed" version of the "json-schema-validator" here : github.com boly38 json-schema-validator-bundle.

Of course any feedbacks are welcome

gem : fiche pratique

noreply@blogger.com (Unknown) — Thu, 22 Aug 2013 15:14:00 +0200

Cet article est une fiche pratique pour gem (alias RubyGems).

Commandes de base

Afficher l'environnement

gem environment

Lister les gem installés

gem list

gem q --L

gem query --local

installer le gem "json en version 1.4.6

gem install json -v '1.4.6'

Désinstaller le gem json

gem uninstall json

Lister les gems disponibles

gem list --remote

Rechercher un gem comportant "sdk"

gem search "*sdk" --remote

Ressources utiles : ruby-toolbox.com : recherche de gem et stats consolidées sur les gems (à partir des dépôts public et github)

Créer une gem : un squelette avec un ensemble de fichier par defaut

 gem install bundle
 bundle gem mongemmaison
 cat  mongemmaison/mongemmaison.gemspec
 cat  mongemmaison/bin/mongemmaison.rb

find : fiche pratique

noreply@blogger.com (Unknown) — Tue, 20 Aug 2013 14:01:00 +0200

Cet article est une fiche pratique pour find.

Commandes de base

Lister les fichiers comportant la chaîne "Tutu"

find . -type f -exec grep -l Tutu {} \;

Lister les fichiers modifiés dans les 15 dernières minutes

find . -type f -mmin -15

Lister les fichiers modifiés il y a moins de 1 jour (dans les dernières 24heures)

find . -mtime -1

Lister à partir de /usr les dossiers comportant 'maven' sans remonter les erreurs liées aux permissions refusées

find /usr -type d -name maven 2>&1 | grep -v 'Permission non accordée'

How to resize a VirtualBox Ubuntu vm hard drive

noreply@blogger.com (Unknown) — Fri, 9 Aug 2013 10:18:00 +0200

HowTo : resize ubuntu vm hard drive hosted on Virtual Box

Pre-requisites

on host (host is the computer having virtualbox) :

VirtualBox version should be greater than 4.0.0
you should have enough space to clone your VM and available space (using to extend your hard drive)
you should have enought space to get gparted livecd (130Mo)

Step by step

original Ubuntu will be called "oU" and the clone "oUextended"

on oU menu halt ubuntu (or stop oU using VirtualBox)
on virtualbox : right clic on oU, then (context menu) clone "oU"

name the clone "oUextended" (or wathever you want)

on host under a shell : go under oUextended directory and use this command to extend for example to 25GB the hard drive :

 c:\Users\myuser\VirtualBox VMs\UBUNTU_EXTENDED_25G>"c:\Program Files\Oracle\Virtual Box\VBoxManage.exe" modifyhd UBUNTU_EXTENDED_25G.vdi --resize 25625  
 0%...10%...20%...30%...40%...50%...60%...70%...80%...90%...100%

on virtualbox : oUextended/configuration you should show the hard drive size : 25G (==25625MB)
on virtualbox : oUextended/configuration you should assert booting on CD before hard drive

go to http://gparted.sourceforge.net/download.php and download the last stable gparted ISO
on virtualbox : oUextended/configuration/storage : add a CD and select gparted iso
on virtualbox : start oUextended
on oUextended : boot menu : select the first and default choice : "GParted Live"
on oUextended : anwer gparted question (depending of your preferences) : eg. keyboard azerty / fr / PC Keyboard / 08 French / (0)
on oUextended : select the parent ("extended") of the partition you would like to extend : right clic on it / resize / extend to the full free size

on oUextended : the parent ("extended") of the partition should have 24Gio size
on oUextended : select the partition you would like to extend : right clic on it / resize / extend to the full free size

on oUextended : the partition should have 24Gio size
on oUextended : clic on "Apply" / wait the operations to finish / clic "Close"
on oUextended : clic on "Exit" on the gparted desktop; then select "shutdown" / enter
on virtualbox : oUextended/configuration/storage : remove gparted iso CD (should be an empty CD)
on virtualbox : start oUextended

Assert the result on ubuntu xterm :

df -kh

Resources

gparted : official website
gparted : gparted french ubuntu documentation
storage related question : virtualbox documentation
a youtube video on how to resize VirtualBox hard drive
codeformatter help me to format this article

Ajout d'un proxy au démarrage d'une instance OpenStack

noreply@blogger.com (Unknown) — Mon, 5 Aug 2013 17:50:00 +0200

Pour compléter la configuration d'une instance OpenStack et ajouter un proxy voici ce que l'on peut faire.

Au démarrage d'une instance il est possible d'ajouter un script personnalisé (Customization Script).
Nous allons utiliser cette option pour :

vérifier que les paramètres de proxy sont correctement positionnés
ajouter les paramètres de proxy s'ils sont absent

Le script qui permet de faire ceci est décrit ci-dessous.
Il suffira de l'ajouter lors de la création d'une instance (dernier onglet) comme "Customization Script" :

#!/bin/bash
INSTANCE_USER=osadmin
PROFILE_SCRIPT=/home/$INSTANCE_USER/.bash_profile
PROXY_PRESENT=`grep http_proxy $PROFILE_SCRIPT | wc -l`
if [[ $PROXY_PRESENT == "1" ]]
 then
   echo "proxy present dans $PROFILE_SCRIPT"
 else
   echo "proxy absent de $PROFILE_SCRIPT ; on l'ajoute"
   echo "export http_proxy=proxy.mycompany.fr:8080" >> $PROFILE_SCRIPT
   echo "export https_proxy=proxy.mycompany.fr:8080" >> $PROFILE_SCRIPT
   echo "export no_proxy='127.0.0.1,localhost,169.254.169.254'" >> $PROFILE_SCRIPT
fi

Explication du script :

PROFILE_SCRIPT comporte le nom du script profile joué à chaque connexion de l'utilisateur INSTANCE_USER.
PROXY_PRESENT permet de détecter si oui (1) ou non (0) ce fichier comporte déjà un paramétrage relatif à "http_proxy".
Dans l'instruction conditionnelle suivante si ce n'est pas le cas (else : PROXY_PRESENT== "0"), on ajoute les export nécessaires et utiles pour positionner les proxies.

Il est possible de constater l'exécution de ce script en affichant (via le portail OpenStack) les logs complets de l'instance
ou en tapant la commande suivante :

env|grep proxy

Cloud Foundry Ubuntu 12.04 quick start instance

noreply@blogger.com (Unknown) — Tue, 2 Jul 2013 17:42:00 +0200

Here is a instance on how to setup CloudFoundry client using Ubuntu (even behind a proxy).

references

install source documentation : getting-started.html#install-cf

CF install without using proxy

(pretty simple) this should work :

 sudo apt-get install ruby rubygems
 sudo gem install cf --no-ri --no-rdoc
 cf target api.run.pivotal.io

CF install behind a proxy

Behind a proxy it seems like there is a problem into CF scripts.
first we need to use ruby version 2 (wich is not the default ruby version).
and then we must patch an url validation step

So first we will use rvm (ruby version managment).

(to avoid) avoid using default ruby-rvm package (sudo apt-get remove ruby-rvm)
(optionnal) clean previous install :

 sudo gem uninstall cf
 sudo apt-get remove rubygems ruby ruby-rvm

(optionnal) to avoid ruby doc task (^src)

  echo 'gem: --no-rdoc --no-ri' >> ~/.gemrc

RVM installation

RVM : Ruby Version Manager (src: https://rvm.io/rvm/install)

 curl -L https://get.rvm.io | bash -s stable --auto-dotfiles

in case of error ("curl returned status '22'.") try it one more time..
next steps:

 . .bashrc
 rvm reload
 rvm install 2.0.0

Ensure running terminal as login shell : cf. https://rvm.io/integration/gnome-terminal
then selecte 2.0.0 version :

 rvm use 2.0.0
 rvm --default use 2.0.0
 $ ruby -v
ruby 2.0.0p247 (2013-06-27 revision 41674) [i686-linux]

CF install using gems

 sudo apt-get install rubygems
 gem install cf

patch CF to use proxy

To launch cf behind proxy, set http_prxy env var; example :

 export http_proxy=myproxy.foo:3128

Try to target cf :

 cf target api.run.pivotal.io
  # with stack trace :
  cf -t --debug target api.run.pivotal.io

In version 3; we are facing the following error :

 Errno::ETIMEDOUT: Connection timed out - connect(2)  
 For more information, see ~/.cf/crash  
 /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:288:in `initialize': Connection timed out - connect(2) (Errno::ETIMEDOUT)  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:288:in `new'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:288:in `sane_target_url'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:309:in `client_target'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:377:in `client'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:163:in `save_token_if_it_changes'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:155:in `block in execute'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:101:in `wrap_errors'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb:151:in `execute'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/mothership-0.5.1/lib/mothership.rb:45:in `start'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/bin/cf:13:in `<top required="">'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/bin/cf:23:in `load'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/bin/cf:23:in `<main>'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/bin/ruby_noexec_wrapper:14:in `eval'  
     from /home/myuser/.rvm/gems/ruby-2.0.0-p247/bin/ruby_noexec_wrapper:14:in `<main>'

We should patch cli.rb script wich doesn't use proxy to validate url

  vi /home/myuser/.rvm/gems/ruby-2.0.0-p247/gems/cf-3.0.0/lib/cf/cli.rb
  /sane_target_url

comment the following line 288 :

  #          TCPSocket.new(url, Net::HTTP.https_default_port)

then try it again :

  export http_proxy=myproxy.foo:3128
  cf target api.run.pivotal.io

this should work now!

Nb: discussion is always in progress on the push command : discussion an an issue has been submitted cf issue#20