Tuesday 6 November 2012

SSH (SECURE SHELL)



       SSH (SECURE SHELL)

Definition of SSH:  SSH is known as SECURE SHELL 



-----> It is a Unix based Command interface and protocol for securely getting access to remote computer.

-----> It is widely used by network Administrators to control web and other kinds of servers Remotely.



-----> SSH is actually a suite of three utilities



1) SLOGIN

2) SSH


3) SCP




----> SSH commands are encrypted and secured in several ways.

---->  Both ends of the client/server connection are authenticated using a digital certificate, and passwords are protected by being encrypted.



----> SSH uses RSA public key cryptography for both connection and authentication. Encryption algorithms.


-----> Include Blowfish, DES, and IDEA. (IDEA is the default.)



----->  SSH2, the latest version, is a proposed set of standards from the Internet Engineering Task Force (IETF).




ISSUE  1  : we got response from users saying that they not able to receive messages from their
application.



SOLUTION   :   we identified that message receiver server not able to receive messages from MQ, because of File storage failed, then we informed the same to system infrastructure team they added SAN, (STORAGE AREA NETWORK) even though the problem not got resolved.

----->  I am seeing some transaction timeout errors in the logs



Q) What is SAN?



A) ------>SAN means STORAGE AREA NETWORK MANAGER.

     ----->  Storage Area Network management is the set of tools, policies, processes, and organization that provide information about monitor the devices in a Storage Area Network (SAN). 

  ------>  To manage the physical infrastructure, the IT organization would have to individually manage each component of the SAN infrastructure

  
  1) 4 * 32 + 8 * 16 = 256 switch ports
  2) 2 different switch management packages
  3) 40 storage frame ports, approximately
  4) 200 disks
  5) 600 shares or mount points
  6) 600 HBAs
  7) 300 instances of 4 different operating systems
    TOTAL NUMBER OF OBJECTS TO MANAGE = 1996



ISSUE  2   :   (1.2) Issues: JMS transaction timeout messages like this in WAS.



SOLUTION   :   there are bulk number of JMS messages stored in MQ Server, when ever san got added at a time all the messages in MQ, hit the server at a time, so server is not able process that many number of requests at a time, then we increase the JTA (JAVA TRANSACTON API) transaction time out value from 30 to 100, this will allow more time for transaction to complete. 



----> After process all the requests we changed the transaction time out value back to  30.




ISSUE 3   : The WAS server logs are not getting generated in both the nodes. The last time stamp in the logs file is 31/10/2011 15:36.The same issue has occurred few times and we have to restarted both the managed servers for the logs to be generated.



SOLUTION    :    The log rotation was not set properly, which I have set now. Also the log stopped in the middle of printing some debugs, which suggests that it ran out of disk space. 



-----> The IBM_profile is mounted on root partition.




ISSUE  4   :    (TAM Test Tool not working) which came today morning to APLSUP54 and request them to check why the ? java.lang.OutOfMemoryError: unable to create new native thread? ocurred.



SOLUTION  :    This is again a native memory issue. The JVM heap allocated is 1.5 Gb (which is necessary otherwise app starts giving heap errors), which leave 512mb for native memory out of the possible 2Gb max.



-----> This native memory is used by all the native modules like MQ or application codes creating native memory. If the native memory is not sufficient then you get OutofMemory:unable to create a new native thread error.


ISSUE  5   :   we are facing connection failure error, due to what causes?



SOLUTION   :    Finally we found, driver there is one DB driver corrupted in it causes the issue.

Firewall -> loadbalancer -> Webserver -> app.serv isa plugin -> hits application -> contact db to ds 

request - > firewall -> loadbalancer(split the load into multiple Webserver) -> (sitemider/ssl) webserver ->via plug in which is in the application server -> application

In side cluster jvm will respond.




Q)   What are the two basic steps that admin have to do after deploying the application & before running the application?



A)  we have to regenerate the web server plug-in, copy it over to the web server machine and do a quick restart of it


Q)   If the app server crashes in the middle of application deployment , wht could be the reason and what steps we have to follow?



A) If the server resources are less while deployment, when application requests are hitting application server due to less resources the application server will crash. If the application server crash then that time we can get thread dump.




Q)  During WAS ND installation, one default server "server1" creates right?can we change its name during installation?

A)  If we are installing, in silent we can change the default server name  in GUI its not possible.




Q)  How To Connect To WSadmin Console through SOAP ?

A)  Generally we connect wsadmin console directly with soap port or rmi port.
specially if we want to check whether a particular SOAP port of dmgr is working or not,  we need whenever we are federating a node to a dmgr.


wsadmin -conntype SOAP -port 8879

wsadmin -conntype RMI -port 9809


wsadmin -conntype RMI -port 2809 -user u1 -password secret1

0 comments:

Post a Comment