Thursday, 31 May 2012

How to change Metrics in windows routing table, Change metric for multicast routing entry

To change the metric first add the entry and then change the entry. Like this, i have done for multicast routes. By changing this entry basically you specify if the multicast traffic originates from this PC, we should sent on which interface.

C:\Documents and Settings\Administrator>route print
  224.0.0.0        240.0.0.0     10.1.1.10         10.1.1.10            20
  224.0.0.0        240.0.0.0     192.168.20.10   192.168.20.10      20

 If same metric is present for both interfaces, i am not sure how it will decide to send on which interface. But it will be random, any interface it will choose and send. To make it send on only one interface change the metric like this :

C:\Documents and Settings\Administrator>route add 224.0.0.0 mask 240.0.0.0 192.168.20.10 metric 20

then change the metric


C:\Documents and Settings\Administrator>route add 224.0.0.0 mask 240.0.0.0 192.168.20.10 metric10

As metric for the interface 192.168.20.10 is less i..e 10 now the traffic will be sent only on this interface.





Wednesday, 30 May 2012

Top excuses by Developers for not Fixing Bugs !!!

I came across many bugs and the top resons for not fixing the bugs, even when they are valid are as follows:
1) Native OS doesn't support it. 
2) As we dont claim we are RFC XYZ compliant, so we won't fix this issue.
3) This bug doesnt affect the functionality in any way. 
4)  As this is very corner case and not observed in common scenarios.
5) This issue seems to be related to third party software not to our codebase.
6) This bug does not affect the performance of the device.
7)  Workaround is available so issue is resolved.
8) This is not recommended setting so wont fix it.
9) Hiding the configuration paramters . If issue is in paramters hide them and fix it.
Please add your comments , i'll be happy to know what other common reasons you come across for not fixing the bugs...

Saturday, 26 May 2012

How assert macro can help in embedded systems?

Have you seen any time these type of print messages?

Assertion NULL, failed at ../../abc.c 876
Assertion NULL, failed at ../../abc.c 876
Assertion NULL, failed at ../../abc.c 876
Assertion 0, failed at ../../abczy.c 987


Syntex is 


void assert ( int expression ) ;


The expression is evaluated and if it evaluates to zero, it causes assertion failure and that terminates the program.


In production generally it is not used and to disable this macro use 


#define NDEBUG before the inclusion of assert.h


There is a great discussion about which is better way to use this macro, if you know these messages you should go through this discussion once...




http://bytes.com/topic/c/answers/479154-how-does-assert-benefit-your-code-really/2

Friday, 25 May 2012

Why IxChariot allows EndPoints A and Endpoint B to be same?

I am using IxChariot for my testing and what i see is , I can create EndPoint A and EndPoint B with same IP address. So this means that there is no check or validation done when starting the tests.

If i give same IP for both the endpoint and run High Performance Throughput script i get max thorughput as Max 3200 Mbps and average of 2800 Mbps for TCP. the script is run for 5 min. If i run with TCP, Thorughput script i get average thorughput as 2373 Mbps.

Anyway , the main question is why there is no validation for IP addressess whther EPA is same as EPB ? As it doesnt make sense when you run the tests between same Endpoints.

Is it a bug ? You decide ?

How to merger two Wirehsark capture files or pcap files ?

I wanted to combine two wirehsark captures into one and see  the packets.

I searched for how to combine or merge two pcap files and this is how i did using mergecap. It allows you to combine two capture files and create a third one.


C:\Program Files\Wireshark>mergecap.exe -h
Mergecap 1.4.3 (SVN Rev 35482 from /trunk-1.4)
Merge two or more capture files into one.
See http://www.wireshark.org/ for more information.
Usage: mergecap [options] -w <outfile>|- <infile> ...
Output:
  -a                concatenate rather than merge files.
                    default is to merge based on frame timestamps.
  -s <snaplen>      truncate packets to <snaplen> bytes of data.
  -w <outfile>|-    set the output filename to <outfile> or '-' for stdout.
  -F <capture type> set the output file type; default is libpcap.
                    an empty "-F" option will list the file types.
  -T <encap type>   set the output file encapsulation type;
                    default is the same as the first input file.
                    an empty "-T" option will list the encapsulation types.
Miscellaneous:
  -h                display this help and exit.
  -v                verbose output.
Run Command :

C:\Program Files\Wireshark>mergecap.exe capture1.pcap cap2.pcap -w output.pcap


Now the output.pcap will have the contents merged of input files, capture1.pcap and cap2.pcap.

Thursday, 24 May 2012

3CX server, VOIP free SIP Server: Call limit reached Error in server !!

We are using 3CX Server as SIP Server for testing quality of Voice calls. It was working fine but suddenely it wasn't allowing even a single call and if we dial to another no, we hear voice like "Call limit has reached".

After we reboot the PC the problem was solved and it was working fine.

What is "DORA" process in Networking?

Hey if some one asks you "What is "DORA" process in Networking?" don't give surprised look. Its just a term now a days people are using, to refer to DHCP negotiation process i.e. DISCOVER, OFFER, REQUEST, ACK.

RFC 2131 is DHCP RFC, you should read at-least once .

I have written some posts on this, you can go through.

Let me know if you need any specific topic on DHCP.

Wednesday, 23 May 2012

"Error: rlm_eap: SSL error :02001002:system library:fopen:No such file or directory " How to solve in freeradius Server.

Free Radius Server SSL error :



I was using Fedora 14 we downloaded the source code of free radius. did following things:

1) ./configure
2) make
3) make install

Now everything got installed , but when i started the server following error came:

radiusd -X

Error: rlm_eap: SSL error :02001002:system library:fopen:No such file or directory
 Error: rlm_eap_tls: Error reading
certificate file /etc/raddb/certs/server.pem
 Error: rlm_eap: Failed to initialize type tls
: Error: /etc/raddb/eap.conf[17]:
Instantiation failed for module "eap"
 Error:/etc/raddb/sites-enabled/inner-tunnel[223]: Failed to find module
"eap".
Error:/etc/raddb/sites-enabled/inner-tunnel[176]: Errors parsing
authenticate section.
Error: Errors initializing modules



=========================

I searched in forum and the solution which worked

1. go to certs folder [/etc/raddb/certs]
2. start bootstrap [./bootstrap]
3. chown root:radiusd /etc/raddb/certs/*]


Now again restart the server "radiusd -X " , it worked fine.


Ix Chariot VoiP test, MOS scores are not displayed , How to get the MOS score in IxChariot VoIP Pair.

I was running IxChariot between two endpoints as VoIP pairs. The idea was to get the MOS score between the host behind the access Point and the Wireless client.

So I installed IxChariot in one PC and seperate WinXP Endpoint in Wirless client.

Note that this is licensed software and is not available free. Qcheck is another software which is available free, given by IXIA and can be used for basic performance tests.

When i ran the tests, the test finished successfully but the MOS scores were not displayed. I ran again the test for 4-5 times with no success.

The support site of IXIA says this
=========================================

    Symptoms

    A VoIP test in IxChariot completes, but no MOS scores or one-way delay are being reported.

    Cause

    VoIP tests require UDP clock syncing packets to flow between EP1 and EP2 over Port 10115. Also, a restrictive firewall can block types of

test traffic to flow between endpoints.

    Solution

    Open Port 10115 for UDP in both directions through the firewall.

==========================================

In my case there was no firewall involved so nothing could be done. OK so next step was to restart the Endpoint. I have observed that the Chariot performance reading depends a lot on PC configuration. Also if you run contious tests on same PC for long duration at some point you see throughput drop.I did not investigate the case but if we restart the EP ( Endpoint) the problem will go.

So i restarted the endpoint and Whola !! i started gettting the MOS scores :-)


Tuesday, 22 May 2012

Moving from Testing to Development, is it OK?

So you are planning to move from testing to development ? Is it a good move? Please answers below questions:

1) How many years of testing experience you have?

2) Are you into manual or automation testing?

3) How many hours you can spend in office?

4) Are you aware of online forum and communities?

5) How comfortable you are in Linux w.r.t Windows?

6) Are you comfortable working late nights and weekends?

7) Is salary is main reason, because of which you are thinking of this career change?

8) Hows your testing team , are you a team player?


Lets see how it goes:

1) Generally i have seen people moving from testing to development in early years of career i.e. between 1-3 years.  If you have crossed 4 rth year in testing, its preferable to be in testing only. if you move into development it will be like you have started with zero experience even if you have 4 years of testing experience.

According to me , development is always a subset of testing. In testing you will have system knowledge but in development its in modules. For development you need coding  skill set and for testing you need attitude.
I know many people wont agree to this, but this is what i think.

So the main point is if you move from "testing to development", people tend to zero your experience of testing. But if you move from "development to testing" its like moving from one specialized set towards whole system knowledge so still your development experience hold true.

2) If you are into manual testing moving to coding will be little difficult. But nothing is impossible, if you are determined. IF you are already into automation , then its all about logic, whether it is C, C++ or Perl, Python and TCL/TK. The transition will be easy for you.

3)  IF you think that the job should be of 9-5 time, then development is not for you. When the deadline approaches i have seen developers sleeping in office :-)

4) Do you know half of the work done in development is only porting of software. Generally 80 % companies take code/ module from open source platform and implement in their platform. The biggest drawback is if an issue exists in open source it is quite possible that you will also face that problem. So you should be comfortable in asking in forum and open source communities. I have seen developers getting fired for not solving issues which exists in open source platform.

IF you face such situation you should be smart enough to convince people/management that its not your fault, if you fail to do so it may affect you appraisal process. Don't get scared just be smart.
 
5) You should be as comfortable in Linux as you are in windows.

6) Testing and development both has their own share of late nights and weekends, but development has more, be prepared for this change.

7) Salary in product company for testing and development is almost same. But in Services based company it may vary. Generally its twice your valid experience.

8)  I have seen a very good bond between team members of testing team. I have seen only 1-2 two guys/gals in a team of 25 who are "Different". By "Different" means the typical type of people with ego, i-am-better-than-you attitude, i-know-lot-but-i-wont-tell attitude . Otherwise the testing team share a very good bond between themselves. But in development everyone is "Different" , so be prepared for it. Generally its all alone job in development, so don't rely on others to solve the logic for you. If he/she does also he/she will take the credit so be prepared for it.

Happy Transition !!!   

How to uninstall a program in Linux?

1) if installed using rpm use
        rpm -e package-name

2) if installed from source code
       make uninstall

3) If "make uninstall" not working remove manually all the files. Read Makefile to understand where it is installing the files.

4) yum remove package-name

5) sudo apt-get remove package-name ( for debian/ubuntu).

6) Use command "make -n install" and note down the steps, try to reverse the steps.

Before trying above, try using built in application "package manager" first, using this you can install/ uninstall the program.


Thursday, 17 May 2012

Development to Testing, is it good career move ?

I have seen people thinking over and over again about this question.

The reason for this change can be

1) Too much work pressure. They need to work on Saturday and Sundays. Evryday its like 12-14 hrs they spend in office.

2) Not good team member/ manager/lead

3) Marriage and or Children : AS they are getting married or planning to have babies they think of moving to testing.

Reasons stopping them (Myth)  :

1) Good salary ( Most cases the only reason !!!)

2) The pride they feel when they are into development.

So is it a good decision to move from development to testing? Yes and NO. IT always depends again on few factors. Every individual is different and so every situation is. So you should think about few things before making move:

1) IS it something you wants to move away from work pressure ? 


   The fact can be stated like: Lets say the developer works 15 days in a month, give the code to tester , the tester tests the code in a week and developer will again fix the code in a week. So effectively if you see the developer works for 3 weeks in a month and tester only one week. But this one week the tester has to work very hard to break the code. So if the developer put X % effort in writing code the tester has to put 2X% effort to break the code. So the remaining 3 weeks will be spent by the tester in analyzing standards, learning about tools, knowing about use cases etc.
     The developer will always be in pressure to make the code and for tester it is always break the code.

 Generally for breaking the code there is no set time lines but for making code it is , so developer will always be tensed and working.

 If you want to move away from that pressure "Testing" is the job for you.


2)   Are you running from your boss?
      If you want to change because of your team/ lead or manager, then may be this is not the right move for you. In QA or in development there are "Best Managers" and "Worst Managers" , whom you are reporting or whom will report you dont know, both departments has its own share of both types of managers.
    Better will be, for you, is to involve your managers' manager and  let him know what is happening.

3) Marriage or Children : Yes , please move to testing. "Family" and  "Money" both are important. Once money gone you can earn it again but "Family" gone , you can never get back. So you should see if you are spending good time with your kid and family members. To work once-twice on weekends is ok but if it is a routing you should look for change. It can be either company change or career change to testing.

4)  Salary difference: Salary in company to company varies, but in general, for product based company the salary of testers are equivalent to developers. For services companies still there is difference but its closing fast. Now for same experience, the salary for tester and developer is almost same.

5) Pride and EGO issues : I have seen QA people getting more respect than development people. Its all about knowledge, if you have good knowledge of your subject, you will get respect. It does not matter whether you are in development or testing.

So if you are sure and know the reason don't think too much . Try to make "Testing as Career" , it is worth.

Believe me !!!!



   

 


Reverse engineering for firmware!

Recently i came across this great post about reverse engineering.

http://www.devttys0.com/2011/05/reverse-engineering-firmware-linksys-wag120n/



Wednesday, 16 May 2012

What are different switching modes? How switch forward data?

There are basically four types of  modes in which switch can operate:

1) Store and forward : Stores full packet in buffer, do CRC validation and if passed forward the packet.

2) Cut thorugh : Only 14 bytes receive, check destination forward the packet. Faster than Store and Forward.

3) Fragment-free : Checks for first 64 byte, if proper forwards the data.

4) Adaptive : Changes to any of the above modes based on CRC error count and runt count (possible collision)

Eg :    Store and forward switched : Dlink DES-1008PA 8-PORT, HP E6600  Cut through :        NETGEAR XSM7224S, Cisco Nexus 5010  Fragment free :   Cisco catalyst 2100 ( supports all ) 

What is MAC layer ACK in wireless ? ACK in wireless is generated by MAC or PHY layer

This is one basic question of wireless. Which layer generates  ACK for unicast packets?

Is it physical layer or the MAC layer. When we say MAC layer, is it controlled by software or hardware?

When i asked this question to few team members, they were like confident that it is "physical layer". As the software does not seem to control this functionality, and it is done at hardware level.

I dig through the standard and no where in physical layer specification, i found the ACK generation. So it has to be "MAC layer". But how?

The answer after researching and searching through Google was : it is generated at MAC level but done at hardware level.

Generally if it is hardware people tend to think its only physical layer. But its not the case now a days many chip sets has advanced capability of doing things at hardware level. The main advantage of this is, it is very fast. The same thing, you do, at hardware level and software level. The hardware level implementation will always be fast, so to make things easy and fast, few things are implemented at hardware level. This functionality is one of them.

IF you go through TI  chipset specification you get below points

http://focus.ti.com/pdfs/bcg/ti_acx100.pdf


"The MAC hardware can directly generate ACK, RTS, CTS, probe response and beacon packets. It also provides wired equivalent privacy (WEP). The MAC can perform encryption and decryption on frames using keys of up to 256 bits in length"


This clearly states that MAC hardware is generating the ACK packet. So if someone ask you this question you should say :

"The generation of ACK is done at MAC layer, but this MAC layer is part of hardware which is physical layer."


 Don't get confused !!!


Difference between Frame, Packet Segment and Datagrams ?

These terms are used to refer at which OSI layer we are referring to, the data remain same but the headers and trailers are getting added or removed.

In Phy layer i.e. physical layer the actual data is in "bits" i.e. 0 and 1.

When it reaches to data link layer or Layer 2 it becomes "frame" with source and destination mac address getting added.

When it reaches the third layer or the networking layer it becomes a "packet" with source and destination IP address attached to it.

Finally it become "segment" at Layer 4 or TCP/IP layer. Basically "segment" term is used for TCP connections and "datagrams" for UDP.

So what about other layers like : session, presentation and application ? In these layers the data is termed as "PDU" or "protocol data unit".

So when someone says, that a "frame" is being received by a networking device, then you know that the device is switch.

Router receives a packet and process it.




Thursday, 10 May 2012

Code of conduct when you join some open source forum!

First time when you join some forum its always good to follow some basic things like:

1) Please be patient while asking question, thanks them when they are trying to help you. And do some basic research before asking question. No body is free, and its their wish if they want to help you or not.

2) Always remember ,if u do something wrong , its not only you , it affects all people like you. I have seen if its not anonymous question , people tend to form opinion for whole race or community based on the way you are asking questions. So don't repeat mistakes.

2) Basic rule is "state clearly what you intend to do and what is not happening" .

3) Don't spam the list with same questions. If you don't get answer, please don't post again and again the same question , begging for their help. Look for other forum, other places where you can get answer.

4) Always try to give more and more information  about the question, all logs, commands, your tried steps. Link to steps followed, reference to manuals etc. It should be like you tried your best to find the answer and as you couldn't , you are asking in forum.

5) Never forget to search the forum for the answer, before asking the question. Most probably you'll have someone who has asked similar question before.

This is very good link which explains "how to ask questions" . I recommend everyone who wants to join forum to go through this once.

http://www.catb.org/~esr/faqs/smart-questions.html



What is the career path for experienced tester ?

This  is something i have been asking myself for quite sometime. What is career path for experienced tester ? Every year so many people are getting graduated and looking for jobs.

Still, in India its more like getting job is priority rather than thinking about questions like development or testing. Here i see a pattern, if you are good ( percentage 75 % or more ) after finishing college, first priority is development, if you didn't get job in first year , you look for testing jobs. Still no job, ok lets do some testing course. Finally, join somewhere.

Testing, as a career is still not  popular , and sometimes i am surprised, when they compare testing job with "Data operator jobs".  Anyway, that's another topic altogether.

Well, in my case i started career in JAVA, got SCJP, very popular in 2000, joined some company as software engineer and worked there looking for codes. It was more like when you are in development you know everything about your module, but if i ask some basic question about full system, you don't have an answer. Exceptions are always there, but this is general pattern i see.if you ask the same question to tester, he can tell you from different ways , explain perfectly what you are looking for. So my views, my personal view is "Testers" has more knowledge about the system as a whole rather than a "Developer". Slowly i got chance and i moved to testing.

Now, it is a beautiful journey, testing new things, new protocols , new  devices, but now what? after almost 12 + years i have mentored many people, saw them moving to development. Many people moving from dev to testing ( surprisingly numbers of people moving to testing is more than the number of people moving from testing to dev, as i know) .

Now, every year intern join, within a year or two they can do almost the same things which you will do. So, how do you differentiate yourself? What options do you have when you move up the ladder ?

Well, we have many options !!!

1) Become a manager and manage people. : In India becoming manager is one option which is true as you become senior.

2) Contribute more towards open source projects.

3) Mentor new people.

4) Start a website and write about your experience.

5) Learn more things, new things . Learning never stops and technology evolves very fast.


Wednesday, 9 May 2012

ICICI Prudential Website Errors ( Login error : Failure of Server APACHE bridge)

I had to pay insurance premium so i tried to access the website , http://www.iciciprulife.com . I logged in after two attempts. It was very slow.

A) When i clicked on home page i got this error :


 
I did some Googling for this problem and from ONT forums, found like:

"The error usually means that the web server can't communicate with the other web applications such as shared services/ workspace."

Below answers to fix it :

  • Increase the WLIOTimeoutSecs in the filles - HYSL-Weblogic.conf, iisproxy.ini
  •  Edit <instance home>\config\OHS\ohs1\moduleconf\forms.conf (or the corresponding .conf file where you have defined the <Location> with the URI of your J2EE application) and set DynamicServerList On (it will be Off by default) . 
  •   The issue was the web server is not having enough memory to allocate multiple process running in parallel. You can look at JVM optimzation on web server process.
  • Stop and start the OHS component, by default ohs1


B) When tried to click on other links got below error messages:





Hope they improve their website and get it tested properly with load, stress tests as well.

Tuesday, 8 May 2012

Different types of VLAN , VLAN types and VLAN modes?

What are different "Types" of VLAN?

What are different "Modes" of VLAN ?

Are "Types" and "Modes" are same terms or has some difference ?

Where is "Access/trunk/transparent" ports defined?

What  is "IP/Protocol/Mac based VLAN?

How to configure VLAN on different switches ?



Monday, 7 May 2012

[Project-Open] How to restore back up file in Project-open

In ]po[ we wanted to do some experiments, so we installed it in new machine.
We have daily backups for the database for production server  , so i took the latest one and put it in the directory "backup" in newly installed PC.

When we tried to restore the database we could't see the file in Admin-> backup through webpage.

So I created fresh backup ,  it created a new file at the same location. Now i renamed the "old database" file as the "new database" file . Now, i restored it, and it worked fine.

It means that we cannot restore backup, if it is not in the same format as generated one from project-open.

Sounds like "Bug" but i need to confirm again before declaring this :-) 

Sunday, 6 May 2012

Generate OOPS and kernel panic in kernel ( module )

From some mailing list :

For OOPS :
--------------------------------
create a module.... and in the init function.....

int init_module(void)
{
     /* deferencing a null pointer */
     *(int*)0 = 1 ;
     return 0;
}

the kernel will generate an oops when you insert this module.

---------------------------------

For Kernel panic , need to search still.  

Difference between JFFS2 and Sqashfs

I was trying to read about the differences between Jffs2 and Sqashfs filesystems. I could not find differences in bullets so i created this for my reference.




Squashfs is a highly compressed read-only filesystem for Linux. Squashfs compresses both files, inodes and directories, and supports block sizes up to 1Mbytes for greater compression. It is implemented as a kernel module under VFS.



Journalling Flash File System version 2 or JFFS2 is a log-structured file system for use with flash memory devices. It is the successor to JFFS. JFFS2 has been included in the Linux kernel since the 2.4.10 (2001-09-23) release. Most prominently JFFS2 is used in OpenWrt.
At least three file systems have been developed as JFFS2 replacements; LogFS, UBIFS, and YAFFS.


Squashfs:

  • Data, inodes and directories are compressed.
  • Squashfs stores full uid/gids (32 bits), and file creation time.
  • Files up to 2^64 bytes are supported. Filesystems can be up to 2^64 bytes.
  • Inode and directory data are highly compacted, and packed on byte boundaries. Each compressed inode is on average 8 bytes in length (the exact length varies on file type, i.e. regular file, directory, symbolic link, and block/char device inodes have different sizes).
  • Squashfs can use block sizes up to 1Mbytes (the default size is 128K). Using 128K blocks achieves greater compression ratios than the normal 4K block size.
  • File duplicates are detected and removed.
  • Both big and little endian architectures are supported. The mksquashfs program can generate filesystems for different endian architectures for cases where the host byte ordering is different to the target. This is useful for embedded systems.
  • SquashFS supports variable block size 0.5KB to 64KB
  • SquashFS supports compression of both the metadata and block fragments 
  • SquashFS  comes as a set of kernel patches and the driver module
  • SquashFS is a read-only file system

JFFS2 (Journaling Flash file system) :
  • Designed specifically for use with both NOR and NAND flash devices. 
  • Compression. Three algorithms are available: zlib, rubin, and rtime.
  • Better performance. JFFS treated the disk as a purely circular log. 
  • It has mechanisms for plugging in different compression algorithms, including not using any compression at all.
  •  JFFS2 is integrated into the kernel.
  • JFFS2 is a read/write file system


Source :





[Project-Open] : How to increase file storage size ?

We are using Project-Open , project management system for time sheet and absence management. We are thinking to use it as master documentation system as well . It has file storage section which can be used to store files.

To enable the limit on single file upload we can modify it as below :

In  "C:\project-open\servers\projop\etc\config.tcl" change the maximum upload size to 200 MB. The default is defined as 20 MB. Shutdown and restart the server.

For me, I was able to upload the file approx 197 MB after the change in parameters. For 400-500 MB file size i think we need to modify other session time out parameters also, as given in below link.

As of now we have max size requirement of 30 MB, so no issues.


As i understand "File storage" section through its online documentation :

What ]po[ file storage doesn't have:

1) It has no version control system of files.

2) It will not initiate mails when files are uploaded, 

3) Folders have permission not the files.

What ]po[ file storage can do:

1) It can link files with projects. So all project related files can be kept at one place.

2) Permissions can be given, read/write/view/admin for viewing files.

3) Files can be searched with in the file storage ( need to see how ?)

4) Upload / Download can be tracked ( need to see how ?)

5) Status of folders by users ( closed or open ) can be tracked ( need to see how ?)

Will see how it goes !!!

Friday, 4 May 2012

How to remove PostgreSQL in Windows machine

These steps i found in one of the forums , and is very useful.

In Windows:

Via Uninstaller:

i) Go to the Add/Remove Programs and uninstall postgresql server from there or go to the postgresql server installation directory and run the uninstaller.

Manual Uninstallation:

i) Remove the postgresql server installation directory. (rd /s /q "C:\Program Files\PostgreSQL\8.3") Assuming default location.

ii) Delete the user 'postgres' (net user postgres /delete)

iii) Remove the Registry entries. (HKEY_LOCAL_MACHINE\SOFTWARE\PostgreSQL\Installations\postgresql-8.3) and (HKEY_LOCAL_MACHINE\SOFTWARE\PostgreSQL\Services\postgresql-8.3)

iv) Remove the postgresql-8.3 service. (sc delete postgresql-8.3) 



I like Revo uninstaller the best, it removes and invokes built in uninstaller tool, clear  all registry settings. But for the postgres user, we need to delete it manually only as above.

Thursday, 3 May 2012

How to start Absence Workflow in Project-Open?

We are using Project-Open 3.5 on Windows XP platform. We wanted to use it to apply absences and get approval.

For this i went through Workflow process documents of Project-open, and it seems like an ocean. The great part about it is, even if I read through all the documentation, i could not find any way, to do it.

So i went through forum and i got the answer. To enable workflow for absence approval, we need to

Go to -> Admin-> Categories->Intranet Absence Type->

Select the create Leave type  Like "Casual Leave" -> In string 1 give "vacation_approval_wf"

Restart the server. Now when you add any new Absence , you can see in workflow inbox , approve button will come.

Still we were not able to initiate "emails" when any leave is applied, searching in forums, hope to get some day  :-) :-)
Related Posts Plugin for WordPress, Blogger...