• powermt check ( – select option –a ) ( This is needed when you reboot the server Mostly for patching do – Disk Management Rescan and run the command to clear dead paths )
• powermt update lun_names ( This is needed when LUN names is to be updated and you have to update in the system to refresh same)
Thursday, April 17, 2014
Friday, March 28, 2014
Reconfiguring Storage Processor (SP) IP address using Navisphere Manager
Reference
Reconfiguring Storage Processor (SP) IP address using Navisphere Manager
CLARiiON
CX Series,CLARiiON iSCSI Array.
Description:
Procedure
to reconfigure SP IP address using Navisphere Manager.
Resolution:
The storage processor (SP) IP address can only be set through the /setup page. Please follow the steps give
below.
If this procedure is done with active hosts attached to the array,
ensure that there are no single HBA-attached hosts and that failover software
is configured on each multi HBA attached host. SPs will reboot each time the IP
address change is submitted. Wait to change the other SP's IP
address until the first SP has rebooted.
1.
If
the storage system to be reconfigured is the domain master, select a new domain
master before proceeding.
2.
Make
sure both SPs for which you want to change IP addresses are connected to the
network
3.
If
the array on which you want to set or cannot set the IP address ("problem
array") is in a domain with other arrays, open Navisphere Manager on one
of the other arrays and remove it from the domain using the Configure Domain
dialog.
4.
Open
the setup page for SP A <SP
A IP address/setup> on the array.
5.
Reset
security and domain information for SP-A by pressing the "Reset
Security" button. Select "Yes" and "Submit" to
confirm.
6.
Re-open
the setup page for SP A on the array and restart the Management Server by
pressing the "Restart Management Server" button. Select
"Yes" and "Submit" to confirm.
7.
Open
the setup page for SP B <SP
B IP address/setup> on the array.
8.
Reset
security and domain information for SP B by pressing the "Reset
Security" button. Select "Yes" and "Submit" to
confirm.
9.
Re-open
the setup page for SP B on the array and restart the Management Server by
pressing the "Restart Management Server" button. Select
"Yes" and "Submit" to confirm.
10. Once the Management Server on
SP A and SP B has been restarted, security and domain information will be in an
uninitialized state. You should now be able to set the IP addresses using
the following steps
11. Open a browser and go to <Current SP A IP address/setup>
12. In the IP address field change
the current SP A IP address to the new SP A IP address and click "Apply
Settings". SP A will reboot which may take up to five minutes to complete
13. Go back to SP A's setup page
using the new SP A IP address <New
SP A IP Address/setup>
14. Open a browser and go to <Current SP B IP address/setup>
15. In the IP address field enter
SP B's new IP address and in the Peer IP Address field enter SP A's new IP
address and click "Apply Settings". SP B will reboot which may take
up to five minutes to complete.
16. Go back to the SP B setup page
<new SP B IP
Address/setup> and click "Restart Management
Server". Select "Yes" and "Submit" on the pop up
screen.
17. Go to SP A setup page <New SP A IP Address/setup>
(You may already be there due to step number 12) and click "Restart
Management Server". Select "Yes" and "Submit" on
the pop up screen. This will update the peer IP address on SP A.
18. If the array will be back in a
domain with other arrays, re-add it to the existing domain.
Security will automatically be initialized by obtaining the necessary
information from the domain master. If it was originally in the domain and it
was the domain master you can now select it as the domain master again.
If the array is not to be in a domain, open Navisphere Manager on SP A of the
array. You should now be prompted to initialize security. Select
"Yes" and add a global username and password.
Friday, March 14, 2014
How to migrate Clariion to VNX using SANCopy (naviseccli)
SANCopy syntax:
naviseccli -h -create -name name <-incremental>
[-srcwwn srcwwn|
-srclun srcspname srcNumOrSnapshotName <-device devicename="">|
-srcportwwn srcportwwn srclunnumber]
[-destwwn destwwns|
-destlunnumber destspname destlunnumber <-device devicename="">|
-destportwwn destportwwn destlunnumber]
<-changesonly> <-verify> <-start> <-throttle 1-10="">
<-linkbw linkbw=""> <-latency latency=""> <-o>
Example:
navicli -h sancopy -create -name -srcwwn xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx:xx destlunnumber -verify
Start SANCopy:
naviseccli -h sancopy -start -name
naviseccli -h
[-srcwwn srcwwn|
-srclun srcspname srcNumOrSnapshotName <-device devicename="">|
-srcportwwn srcportwwn srclunnumber]
[-destwwn destwwns|
-destlunnumber destspname destlunnumber <-device devicename="">|
-destportwwn destportwwn destlunnumber]
<-changesonly> <-verify> <-start> <-throttle 1-10="">
<-linkbw linkbw=""> <-latency latency=""> <-o>
Example:
navicli -h
Start SANCopy:
naviseccli -h
Sunday, March 2, 2014
How to see the EMC ECC Bundle Update version?
Wednesday, September 4, 2013
Memory usage is very high on Redhat Linux
# cat free
total used free
shared buffers cached
Mem:
24674784 24576836 97948 0
351244 21165268
-/+ buffers/cache:
3060324
21614460<<<<-----actual buffers="" cached="" free="" memory="" p="">
= + +
= - - -
= -
Swap:
25165808 1084452 24081356
Total Physical
Memory = 24674784 KB = 24096 MB
Physically Used
Memory = 24576836 KB = 24000 MB
Actual used
memory = 3060324 KB = 2988 MB = ~3 GB
buffers =
351244 KB = 343 MB
cached =
21165268 KB = 20669 MB = ~20 GB
Physically Free
Memory = 97948 KB = 95 MB
Memory free for
Applications = 21614460 KB = 21107 MB = ~20 GB
The items to note here are:
The philosophy in Linux is that an unused resource is a
wasted resource. The kernel therefore will use as much RAM as it can to cache
information from your local and remote filesystems and disks. This builds up
over time as reads and writes are done on the system trying to keep the data
stored in RAM as relevant as possible to the processes that have been running
on your system.
This caching is reported by the system as the sum of two
numbers, buffers and pagecache. The cache is reclaimed, not at the time of
process exit (you might start up another process soon that needs the same
data), but upon demand - i.e. When you start a process that needs a lot of
memory to run, the Linux kernel will reclaim memory that had been storing
cached data and give it to the new process.
Refer below article for detail information on this:
What is cache memory and why is memory utilization high
for cache memory?
Why is so much of my memory used by cache?
Saturday, August 24, 2013
ITIL
ITIL
* Companies want to concentrate on their core business. so non-core businesses are outsourced
to other vendors (service providers) to avoid those disturbances.
example, ICICI bank's core business is financial banking, ICICI does not want to have head ache
in managing their datacenter and the application which supports their core banking work.
so ICICI decides to outsource the backend infrastructure to other companies (service providers
like IBM, HP, TCS, Infosys, Unisys,...)
for example, now ICICI choose TCS to give support. what next?
The process and procedures should be in place.
both ICICI and TCS would sit and define the SLA (Service Level Agreement) which covers all the
topics like what are all to be supported by TCS. i.e, what OS and what applications?
at the same time, ICICI wants the every work done by TCS to be recorded like,
what is the work?
what is the date and time executed?
who performed?
and its result (success/failure)?
so this way, ICICI will have good history about the changes in their servers or applications.
This history can be used if required.
So here it comes ITIL.
* The Information Technology Infrastructure Library (ITIL) is a set of best practices that may
be used to deliver high quality IT services.
ITIL is best practice guidance and the only best practice framework for effectively managing
service management in IT, ITSM.
ITIL Benefits:
1. To align IT with the business
2. Reduce cost of operations
3. Improves quality of service
4. Delivers consistent IT services
5. Provides guidance for all types of organisation
There are 12 modules in ITSM. Important are,
1. Incident Management
2. Problem Management
3. Change Management
* An Incident is a system failure or error in the system. This may be a full disk (file system
full), a broken server (server down), or any application crash, etc. These are isolated one-time
events that have procedures for handling and fixing. These are managed under Incident Management.
* If the same type of incidents for a same application or for a same server are geneated, this
is to be considered as a Problem. every problem would have a reason what it caused and the
solution. i.e., root cause and the solution.
These are managed under Problem Management which helps to avoid further future same typr of
incidents.
* Managing an Incident means fixing the broken system to get it working as soon as possible.
While managing a Problem means finding the underlying root causes so that the Incidents do not
happen again or as often.
* Once we found the reason (root cause) of a problem and solution, the solution has to be
implemented. so there is a change in server configuration or application or OS.
These changes can not be done straight away. because the implementer of the change may not know
the impact of the change. Impact means application outage, future issues for other teams, etc.
So all the relevant teams would join together, discuss about the change and finally approve the
change. so Change would be implemented after the approval on a given date and time.
These processes are managed through change module.
* ITIL is not a tool, it is a set of rules and the best practices.
By considering these rules, some companies designed software to use these incident, problem,
change management best practices. Those softwares are,
1. BMC Remedy
2. Peregrine
3. osticket
4. Altris
* BMC Remedy Versions:
Remedy 6
Remedy 7.1
Remedy 7.6
* Companies want to concentrate on their core business. so non-core businesses are outsourced
to other vendors (service providers) to avoid those disturbances.
example, ICICI bank's core business is financial banking, ICICI does not want to have head ache
in managing their datacenter and the application which supports their core banking work.
so ICICI decides to outsource the backend infrastructure to other companies (service providers
like IBM, HP, TCS, Infosys, Unisys,...)
for example, now ICICI choose TCS to give support. what next?
The process and procedures should be in place.
both ICICI and TCS would sit and define the SLA (Service Level Agreement) which covers all the
topics like what are all to be supported by TCS. i.e, what OS and what applications?
at the same time, ICICI wants the every work done by TCS to be recorded like,
what is the work?
what is the date and time executed?
who performed?
and its result (success/failure)?
so this way, ICICI will have good history about the changes in their servers or applications.
This history can be used if required.
So here it comes ITIL.
* The Information Technology Infrastructure Library (ITIL) is a set of best practices that may
be used to deliver high quality IT services.
ITIL is best practice guidance and the only best practice framework for effectively managing
service management in IT, ITSM.
ITIL Benefits:
1. To align IT with the business
2. Reduce cost of operations
3. Improves quality of service
4. Delivers consistent IT services
5. Provides guidance for all types of organisation
There are 12 modules in ITSM. Important are,
1. Incident Management
2. Problem Management
3. Change Management
* An Incident is a system failure or error in the system. This may be a full disk (file system
full), a broken server (server down), or any application crash, etc. These are isolated one-time
events that have procedures for handling and fixing. These are managed under Incident Management.
* If the same type of incidents for a same application or for a same server are geneated, this
is to be considered as a Problem. every problem would have a reason what it caused and the
solution. i.e., root cause and the solution.
These are managed under Problem Management which helps to avoid further future same typr of
incidents.
* Managing an Incident means fixing the broken system to get it working as soon as possible.
While managing a Problem means finding the underlying root causes so that the Incidents do not
happen again or as often.
* Once we found the reason (root cause) of a problem and solution, the solution has to be
implemented. so there is a change in server configuration or application or OS.
These changes can not be done straight away. because the implementer of the change may not know
the impact of the change. Impact means application outage, future issues for other teams, etc.
So all the relevant teams would join together, discuss about the change and finally approve the
change. so Change would be implemented after the approval on a given date and time.
These processes are managed through change module.
* ITIL is not a tool, it is a set of rules and the best practices.
By considering these rules, some companies designed software to use these incident, problem,
change management best practices. Those softwares are,
1. BMC Remedy
2. Peregrine
3. osticket
4. Altris
* BMC Remedy Versions:
Remedy 6
Remedy 7.1
Remedy 7.6
Thursday, August 15, 2013
List of Unix/Linux Signals
A list of signals and what they mean
•These were all recorded from a Linux i386 system. Numbers may vary between platforms.
•Linux uses signals 34-64 for its real-time system which we are not interested in.
•“man 7 signal” gives the official manual page on signals.
•This is a fairly exhaustive list of signals. Only some of them will arise in the context of the make program.
1
SIGHUP
If a process is being run from terminal and that terminal suddenly goes away then the process receives this signal. “HUP” is short for “hang up” and refers to hanging up the telephone in the days of telephone modems.
2
SIGINT
The process was “interrupted”. This happens when you press Control+C on the controlling terminal.
3
SIGQUIT
4
SIGILL
Illegal instruction. The program contained some machine code the CPU can't understand.
5
SIGTRAP
This signal is used mainly from within debuggers and program tracers.
6
SIGABRT
The program called the abort() function. This is an emergency stop.
7
SIGBUS
An attempt was made to access memory incorrectly. This can be caused by alignment errors in memory access etc.
8
SIGFPE
A floating point exception happened in the program.
9
SIGKILL
The process was explicitly killed by somebody wielding the kill program.
10
SIGUSR1
Left for the programmers to do whatever they want.
11
SIGSEGV
An attempt was made to access memory not allocated to the process. This is often caused by reading off the end of arrays etc.
12
SIGUSR2
Left for the programmers to do whatever they want.
13
SIGPIPE
If a process is producing output that is being fed into another process that consume it via a pipe (“producer | consumer”) and the consumer dies then the producer is sent this signal.
14
SIGALRM
A process can request a “wake up call” from the operating system at some time in the future by calling the alarm() function. When that time comes round the wake up call consists of this signal.
15
SIGTERM
The process was explicitly killed by somebody wielding the kill program.
16
unused
17
SIGCHLD
The process had previously created one or more child processes with the fork() function. One or more of these processes has since died.
18
SIGCONT
(To be read in conjunction with SIGSTOP.)
If a process has been paused by sending it SIGSTOP then sending SIGCONT to the process wakes it up again (“continues” it).
19
SIGSTOP
(To be read in conjunction with SIGCONT.)
If a process is sent SIGSTOP it is paused by the operating system. All its
No. Short name What it means
state is preserved ready for it to be restarted (by SIGCONT) but it doesn't
get any more CPU cycles until then.
20 SIGTSTP Essentially the same as SIGSTOP. This is the signal sent when the user hits
Control+Z on the terminal. (SIGTSTP is short for “terminal stop”) The
only difference between SIGTSTP and SIGSTOP is that pausing is
only the default action for SIGTSTP but is the required action for
SIGSTOP. The process can opt to handle SIGTSTP differently but gets no
choice regarding SIGSTOP.
21 SIGTTIN The operating system sends this signal to a backgrounded process when it
tries to read input from its terminal. The typical response is to pause (as per
SIGSTOP and SIFTSTP) and wait for the SIGCONT that arrives when the
process is brought back to the foreground.
22 SIGTTOU The operating system sends this signal to a backgrounded process when it
tries to write output to its terminal. The typical response is as per
SIGTTIN.
23 SIGURG The operating system sends this signal to a process using a network
connection when “urgent” out of band data is sent to it.
24 SIGXCPU The operating system sends this signal to a process that has exceeded its
CPU limit. You can cancel any CPU limit with the shell command
“ulimit -t unlimited” prior to running make though it is more
likely that something has gone wrong if you reach the CPU limit in make.
25 SIGXFSZ The operating system sends this signal to a process that has tried to create a
file above the file size limit. You can cancel any file size limit with the
shell command “ulimit -f unlimited” prior to running make though it is
more likely that something has gone wrong if you reach the file size limit
in make.
26 SIGVTALRM This is very similar to SIGALRM, but while SIGALRM is sent after a
certain amount of real time has passed, SIGVTALRM is sent after a certain
amount of time has been spent running the process.
27 SIGPROF This is also very similar to SIGALRM and SIGVTALRM, but while
SIGALRM is sent after a certain amount of real time has passed, SIGPROF
is sent after a certain amount of time has been spent running the process
and running system code on behalf of the process.
28 SIGWINCH (Mostly unused these days.) A process used to be sent this signal when one
of its windows was resized.
29 SIGIO (Also known as SIGPOLL.) A process can arrange to have this signal sent
to it when there is some input ready for it to process or an output channel
has become ready for writing.
30 SIGPWR A signal sent to processes by a power management service to indicate that
power has switched to a short term emergency power supply. The process
(especially long-running daemons) may care to shut down cleanlt before
the emergency power fails.
31 SIGSYS Unused.
Courtesy: University of Cambridge
•These were all recorded from a Linux i386 system. Numbers may vary between platforms.
•Linux uses signals 34-64 for its real-time system which we are not interested in.
•“man 7 signal” gives the official manual page on signals.
•This is a fairly exhaustive list of signals. Only some of them will arise in the context of the make program.
1
SIGHUP
If a process is being run from terminal and that terminal suddenly goes away then the process receives this signal. “HUP” is short for “hang up” and refers to hanging up the telephone in the days of telephone modems.
2
SIGINT
The process was “interrupted”. This happens when you press Control+C on the controlling terminal.
3
SIGQUIT
4
SIGILL
Illegal instruction. The program contained some machine code the CPU can't understand.
5
SIGTRAP
This signal is used mainly from within debuggers and program tracers.
6
SIGABRT
The program called the abort() function. This is an emergency stop.
7
SIGBUS
An attempt was made to access memory incorrectly. This can be caused by alignment errors in memory access etc.
8
SIGFPE
A floating point exception happened in the program.
9
SIGKILL
The process was explicitly killed by somebody wielding the kill program.
10
SIGUSR1
Left for the programmers to do whatever they want.
11
SIGSEGV
An attempt was made to access memory not allocated to the process. This is often caused by reading off the end of arrays etc.
12
SIGUSR2
Left for the programmers to do whatever they want.
13
SIGPIPE
If a process is producing output that is being fed into another process that consume it via a pipe (“producer | consumer”) and the consumer dies then the producer is sent this signal.
14
SIGALRM
A process can request a “wake up call” from the operating system at some time in the future by calling the alarm() function. When that time comes round the wake up call consists of this signal.
15
SIGTERM
The process was explicitly killed by somebody wielding the kill program.
16
unused
17
SIGCHLD
The process had previously created one or more child processes with the fork() function. One or more of these processes has since died.
18
SIGCONT
(To be read in conjunction with SIGSTOP.)
If a process has been paused by sending it SIGSTOP then sending SIGCONT to the process wakes it up again (“continues” it).
19
SIGSTOP
(To be read in conjunction with SIGCONT.)
If a process is sent SIGSTOP it is paused by the operating system. All its
No. Short name What it means
state is preserved ready for it to be restarted (by SIGCONT) but it doesn't
get any more CPU cycles until then.
20 SIGTSTP Essentially the same as SIGSTOP. This is the signal sent when the user hits
Control+Z on the terminal. (SIGTSTP is short for “terminal stop”) The
only difference between SIGTSTP and SIGSTOP is that pausing is
only the default action for SIGTSTP but is the required action for
SIGSTOP. The process can opt to handle SIGTSTP differently but gets no
choice regarding SIGSTOP.
21 SIGTTIN The operating system sends this signal to a backgrounded process when it
tries to read input from its terminal. The typical response is to pause (as per
SIGSTOP and SIFTSTP) and wait for the SIGCONT that arrives when the
process is brought back to the foreground.
22 SIGTTOU The operating system sends this signal to a backgrounded process when it
tries to write output to its terminal. The typical response is as per
SIGTTIN.
23 SIGURG The operating system sends this signal to a process using a network
connection when “urgent” out of band data is sent to it.
24 SIGXCPU The operating system sends this signal to a process that has exceeded its
CPU limit. You can cancel any CPU limit with the shell command
“ulimit -t unlimited” prior to running make though it is more
likely that something has gone wrong if you reach the CPU limit in make.
25 SIGXFSZ The operating system sends this signal to a process that has tried to create a
file above the file size limit. You can cancel any file size limit with the
shell command “ulimit -f unlimited” prior to running make though it is
more likely that something has gone wrong if you reach the file size limit
in make.
26 SIGVTALRM This is very similar to SIGALRM, but while SIGALRM is sent after a
certain amount of real time has passed, SIGVTALRM is sent after a certain
amount of time has been spent running the process.
27 SIGPROF This is also very similar to SIGALRM and SIGVTALRM, but while
SIGALRM is sent after a certain amount of real time has passed, SIGPROF
is sent after a certain amount of time has been spent running the process
and running system code on behalf of the process.
28 SIGWINCH (Mostly unused these days.) A process used to be sent this signal when one
of its windows was resized.
29 SIGIO (Also known as SIGPOLL.) A process can arrange to have this signal sent
to it when there is some input ready for it to process or an output channel
has become ready for writing.
30 SIGPWR A signal sent to processes by a power management service to indicate that
power has switched to a short term emergency power supply. The process
(especially long-running daemons) may care to shut down cleanlt before
the emergency power fails.
31 SIGSYS Unused.
Courtesy: University of Cambridge
Subscribe to:
Posts (Atom)