I was going through some of the talks at TED and I must confess I am generally amazed by the thoughts people present but the talk on "Use of internet.." went way beyond all . Clay has very nicely drawn parallels from two completely un-parallel stream. Especially, world over people are thinking about how internet (e-governance) can be used to improve the lives of people and bring transparency to the system ,this is excellent example. But with this I also want to draw attention two difference talks on TED, 1) Exposing corruption in Government 2) Chinese Firewall .. that inspires further and how people can tackle issues that will be faced once community wishes to use internet effectively.
I have been reading about 6.5 m "linkedin" accounts username/password posted on some Russian website by hackers in last day or two, gave 1 minute attention each and moved on. I thought I am not impacted even if my account is hacked.. at most what will happen ? I have not linked my bank account with it ! Hacker will get to know my profile and my contacts , so what ? there are other ways to get it any way. why would any one be so interested in my profile ? hacker can not sell it ...he can probably play with it and make it worse ..but I can still prove to "linkedin" and get it corrected.
Some what more thinking and realized it is not just about getting username and password , may be hacker's want to prove something ? . To some extend agree.. they want to prove that they can hack it and declare it to the world that they did it, which proves that social network are not safe ! but that did not convince me too as these things have happened in past and people still use it.
Next question, is there financial gain ? as I considered it to be utmost motivating :) . well ..not direct financial gain as "linkedin" is not financial organization but definitely there is indirect gain.
What really stumped me is when a day later that what can be done with my hacked account (not only of linkedin). I realized that there was an attempt to hack one of my personal e-mail account, which service provider thinks that was successful and I was asked to re-set password after successful validation of my identity. WOW , this is the account I hardly use and doubt if any one knows about it ..yet there wanted to hack it..and they did it successfully.
Then came "linkedin" account - When I logged in it said you are required to re-set your password via e-mail (not sure if this was hacked too) which is different from what is mentioned in one of the blogs here i.e. "linkedin is working on mechanism to re-set the password but not by clicking links". I also checked official blog of "Linkedin" which said it is a precautionary measure. Ah relieved ! I carefully followed steps to re-set password i.e. made sure non of the link given in e-mail points to a phishing site or site other than "Linkedin" .
So far so good ..I get e-mail notification on my registered e-mail ..good ..but what followed is stunning ..same e-mail confirmation to my all e-mail accounts (including the one that was hacked before ) . where did Linkedin get that information from ? I never provided it ? so ..Pandora's box is now opened ! Linkedin knows about me more then I told them , may be same is the case with other social network provider (Luckily I am not on Facebook ). who knows ?
Now , I am nervous and utmost cautious, it will be matter of time next time hackers can reach my bank account and do what ever they want ? (Exaggeration but possible). So changed all my password etc etc.. but will be wary about safety of my web identity , it is not that we look things in isolation any more everything is linked and any dent in any place can rock entire system.
On the other side, It remains to see what implications these has on enterprise world specially when moving onto cloud is next logical step for most corporates in case they are not doing already. And of course hopeful that this will be only one of instance.
Since RHEL has default personal agent installed on port 5555, it defaults to 5556 , this leads to situation where build forge would require to control agent on different ports (Windows on 5555 ). You have option of defaulting to 5555 after agent is installed which is mentioned here .
However if you still want to have it work on 5556 , there isn't direct configuration that lets you set port for each agent that you want to control , instead there is simple way to specify in the server console . One can navigate to "Servers" and click on "Add Server" , while you specify host in the details tab, mention fully qualified server host name along with port number in the format <fully qualified server name>:<host port> . Also with this you should even be able to control multiple agent on same server having different version of agent installed.
Rational build forge setup instructions guided me to install thin client to have rational tools working with oracle , jumped to install "instant client" which was nothing but extracting a zip file which server did not like and it kept cribbing about error stack with any of following error->
Error 6 Initializing SQL*plus
SP2-0750: You may need to set ORACLE_HOME to your oracle software directory.
message file sp1<lang>.msb not found.
Lot of people commented on this issue being one of setting the correct environment variable issue particularly "ORACLE_HOME".
However for me it was more than that, Oracle "instant client" does not have all capabilities that rational server required. So next logical step was to get "SQL plus" running on machine and most appropriate is to install full client rather than "top up" of instant client (options provided by think client downloads) else you will have same issue. Use Oracle's inbuilt installer to install the SQL client as it sets correct environment variables and also sets registry entry.
Once you finish installation via oracle installer and you are able to launch SQL plus client you are all set to go and Rational server will be happy too !
Different problems have different solutions but with facing customer you come up with something new ! I came across one "simple" to solve problem which turned out not so "simple" to solve.
Requirement was to get an e-mail notfication on completion of delivery while making sure that e-mail is sent out to only component owners and submitters and not to any one else. I would have expected an easy to configure options in RTC and send e-mail but unfortunately it is not so simple,it required relatively complex approach.
RTC allows configuration of e-mail notification via profile setting but it is limited to workitems and does not apply to code base. As shown in profile configuration additionally you could also configure format of e-mail i.e. HTML/TEXT . However it does not allow you to change content or template of e-mail.
To enable notification in code base you could do any of the following :
#1 Change Project/Team area process confiuration.
RTC allows you to customize process configuration for specific project/team, you could add a custom plug-in that sends e-mail notification in a custom content based on rules configured in operational behavior of team configuration. On enabling plug-in you could contribute to pre and post conditions.
This approach is much extensive and provides you fine grained control of activities that needs to be performed. However, on flip side it requires efforts to develop and maintain plug-in.
Additional information on how to develop and configure plug-in can be found in this article Creating Custom Operation Advisors .
#2 Enable "Dummy" build definition to send email.
Another approach, which is faster to develop but relatively complex is to leverage any existing functionality of RTC. Build engine has ability of sending e-mail.
You could create a "Dummy" build definition with only purpose to notify group about completion of delivery. You may additionally want to build the base but that is not important.
Since this build configuration is for entire stream it does not differentiate among components of stream and notification is triggered for any delivery to stream. To circumvert this issue and send notification to only specific group of people for specific component a two phased approach is adopted.
In first pass deliver only component and changes for component to an intermediate stream that is enabled for specific component. Which means you will need intermediate stream for each sub group of component(s).
On second pass which is has build definition configured on intermediate stream to poll at regular interval on any new changes an e-mail notification is triggered only for the components that are part of it to specific group of people.
To Enable spcific component you could configure "Load Options" and "Accept Changes" of "Jazz Source Control" tab.
This approach is quick , However on the other side if your build machine requires large number of build definition configured than you may have to take approach #1. Also, this approach does not allow you to send custom mail so a generic build completion e-mail is sent.
More information about creating build Engine definition can be found here - Creating build definition
We adopted approch #2.