Solaris 10 shutdown

Hello.
I have Oracle installed on 2 Solaris 10 machines and when i shutdown the primary i want it to signal to the standby to perform a switchover. A switchover involves the standby and the now "shutting down" primary exchanging a token. So i edited the S10oracle file that is in rc3.d, in the "stop" portion of the scripts i called another script that scp'd a file to the standby database. When the standby receives that file it starts the switchover. My question is this....
When a script is run after shutdown, has it got a limited ammount of time to run? will it just stop execution after 2 minutes or some other arbitary time? or will it run until it is complete and then the computer shutsdown?
Thanks in advance for any help you can give me

I'm pretty sure there's a timeout on the service that does this. For startup/shutdown, I think it's around 30 minutes. I'm not sure what happens if a timeout occurs, if shutdown continues or halts.
Darren

Similar Messages

How to find out the info regarding Solaris, server hang or shutdown.

Hi
As I am handling the Solaris 9 Server remotely.
As in last week the server has stopped to respond suddenly, we have tried to ping, ssh which was not working. As ultimately we have asked the Data Centre Team to hard reboot the same, to resolve the issue.
In the same ref I would like to know how we come to know the reason, what caused to reboot the same, i.e . any log file, etc..
I have check /var/adm/messages, but not found any detail for the same.
Thanks
Rajan
I

No core files?
No hope for an answer.
As you learned in the other Internet forum.
http://www.linuxquestions.org/questions/solaris-opensolaris-20/how-to-find-out-the-info-regarding-solaris-server-hang-or-shutdown.-621500/
To get such corefiles analyzed,
you would need to use your service contract and log a support case with Sun.
They have the special software tools to do that.

Init 0 or shutdown -i0 on solaris 8 automatically reboots

I have a solaris 8 server and when I issue init 0 or shutdown -i0 it automatically reboots. I was able to issue init 5 and it shutdown/power down, but not init 0.
I looked at inittab and it shows
s0:0:wait:/sbin/rc0 >/dev/msglog 2<>/dev/msglog </dev/console
Any suggestions?
thanks

Hi
Thank you. I have 2 SunFire v440s. Both have auto-boot?true and I only have this problem with one of them. I was able to issue init 5 and the server shut down, but both init 0 or shutdown -i0 rebooted.
Unfortunately the server with the problem is remote so I can't do allot of testing with it.
Thank you for your suggestions.

Solaris 10 05/09 Update 7 - Zone shutdown issue

I have just clean installed several servers with solaris 10 update 7 (both sparc and x64, no other software installed) and I am getting the following svc error messages whenever I shutdown a zone.
Aug 19 18:26:22 svc.startd[25328]: svc:/network/ipsec/policy:default: Method "/usr/sbin/ipsecconf -F" failed with exit status 1.
Aug 19 18:26:22 svc.startd[25328]: svc:/network/ipsec/policy:default: Method "/usr/sbin/ipsecconf -F" failed with exit status 1.
Aug 19 18:26:22 svc.startd[25328]: svc:/network/ipsec/policy:default: Method "/usr/sbin/ipsecconf -F" failed with exit status 1.
Aug 19 18:26:23 svc.startd[25328]: network/ipsec/policy:default failed: transitioned to maintenance (see 'svcs -xv' for details)
The log file for svc:/network/ipsec/policy:default shows the following error.
+[ Aug 19 18:26:22 Stopping because service disabled. ]+
+[ Aug 19 18:26:22 Executing stop method ("/usr/sbin/ipsecconf -F") ]+
ipsecconf: (loading pf_policy) socket:: Permission denied
ipsecconf: unable to open policy socket: Permission denied
+[ Aug 19 18:26:22 Method "stop" exited with status 1 ]+
The errors occur in both sparse and whole root zones with a shared ip-type (does not occur in exclusive ip-type). The zonecfg is shown below.
set zonepath=/export/zones/rdo23wok
set autoboot=false
set ip-type=shared
add inherit-pkg-dir
set dir=/lib
end
add inherit-pkg-dir
set dir=/platform
end
add inherit-pkg-dir
set dir=/sbin
end
add inherit-pkg-dir
set dir=/usr
end
add net
set address=xxx.xxx.xxx.xxx/xx
set physical=bge0
end
Looking at the release notes for 05/09 update 7 it mentions that "IP security (IPsec) is now managed by the following Solaris Management Facility (SMF) services" which seems to fit with the error I am getting.
Although I can prevent the error messages by disabling the service in each zone with svcadm it is a bit annoying and would like to have a better solution. Does anyone know how I can prevent this service from being enabled when I create a new zone?
Thanks
Jools
Edited by: joolshomer on Aug 19, 2009 10:57 AM

I have just clean installed several servers with solaris 10 update 7 (both sparc and x64, no other software installed) and I am getting the following svc error messages whenever I shutdown a zone.
Aug 19 18:26:22 svc.startd[25328]: svc:/network/ipsec/policy:default: Method "/usr/sbin/ipsecconf -F" failed with exit status 1.
Aug 19 18:26:22 svc.startd[25328]: svc:/network/ipsec/policy:default: Method "/usr/sbin/ipsecconf -F" failed with exit status 1.
Aug 19 18:26:22 svc.startd[25328]: svc:/network/ipsec/policy:default: Method "/usr/sbin/ipsecconf -F" failed with exit status 1.
Aug 19 18:26:23 svc.startd[25328]: network/ipsec/policy:default failed: transitioned to maintenance (see 'svcs -xv' for details)
The log file for svc:/network/ipsec/policy:default shows the following error.
+[ Aug 19 18:26:22 Stopping because service disabled. ]+
+[ Aug 19 18:26:22 Executing stop method ("/usr/sbin/ipsecconf -F") ]+
ipsecconf: (loading pf_policy) socket:: Permission denied
ipsecconf: unable to open policy socket: Permission denied
+[ Aug 19 18:26:22 Method "stop" exited with status 1 ]+
The errors occur in both sparse and whole root zones with a shared ip-type (does not occur in exclusive ip-type). The zonecfg is shown below.
set zonepath=/export/zones/rdo23wok
set autoboot=false
set ip-type=shared
add inherit-pkg-dir
set dir=/lib
end
add inherit-pkg-dir
set dir=/platform
end
add inherit-pkg-dir
set dir=/sbin
end
add inherit-pkg-dir
set dir=/usr
end
add net
set address=xxx.xxx.xxx.xxx/xx
set physical=bge0
end
Looking at the release notes for 05/09 update 7 it mentions that "IP security (IPsec) is now managed by the following Solaris Management Facility (SMF) services" which seems to fit with the error I am getting.
Although I can prevent the error messages by disabling the service in each zone with svcadm it is a bit annoying and would like to have a better solution. Does anyone know how I can prevent this service from being enabled when I create a new zone?
Thanks
Jools
Edited by: joolshomer on Aug 19, 2009 10:57 AM

Solaris 8 x86 Shutdown

Hello,
I have installed the Solaris 8 Intel x86 edition. Now I have a (maybe silly) problem: I don't know how to shut down Solaris.
I am working with the CDE desktop. When I am logging out from there, the desktop is closed and the entrance panel for another CDE session appears. From this point, I don't find a way how to get a command prompt or a menu item for shutting down the solaris system for switching off the computer in a safe way.
When I am switching off the computer at the point of the entrance panel, this causes damages of the file system.
Inside CDE desktop, I cannot find a menu item for closing the CDE session and shutting down.
When I open a console within CDE and enter SHUTDOWN or HALT command, this ends the running Solaris but also damages the file system.
Can anybody tell me what the correct procedure is?
Thank you very much for your help!
Thomas

Hello,
I reinstalled Solaris 8 the same way as before, but now I always shutdown the system with init 5. So far, no problems with the file system occured. So I think the problem was the wrong shutdown methods I have used before.
--Thomas

Oracle 8.1.6 autimatic startup/shutdown on sparc solaris 8

I tried to setup a script in /etc/init.d that should call dbstart to auto-start my DB when I boot my sparc solaris 8 server. I tried to run it in rc3 and then in rc2 to no avail. I'd like to know how to setup the server to automatically startup oracle on boot and automatically stop it on shutdown... Can anyone help please?

Could you tell us how to do that????
We all met the ORA-01034 problem!
<BLOCKQUOTE>quote:<HR>Originally posted by Wile E. Coyote:
Yes, that's no problem
I've installed 8.1.6 on several machines running Solaris 8 without any problem<HR></BLOCKQUOTE>
null

Local zones during global shutdown in Solaris 10

Hello,
I am trying to determine if issuing a shutdown command from the global zone, e.g. shutdown -g0 -i5 -y performs graceful shutdowns on any running local zones, or does it issue a halt to the local zones without running shutdown scripts in them. Searching the Solaris zones forum brings up some older discussions (pre 2010) where users reported a halt was done, but the expectation is this would be changed in later Solaris 10 patches or revisions. Has this ever been addressed? Is there a way to determine if a local zone has been halted or shutdown? What is the recommended way to perform a shutdown of a global zone, when local zones are running?
Thanks,

The global zone will run a "svcadm disable zones" for each zones and this corresponds to a "zoneadm shutdown" or "zoneadm halt". Both zoneadm command will use a "init 0" in each zone.
If you have a zone that requires a shutdown script to stopping for example a database gracefully, then you should run the shutdown inside each zone directly.

How to show the shutdown progression of Solaris 10

Hi,
I am just wondering how to show shutdown progress of Solaris 10, the default messages on the console are too little.
Thanks
Bin

i have tried to change /etc/default/init
first of all it is read only fie.
then this command /user/dt/bin/dtconfig -kill
is command is not working.............
is thr any other process to change the language.
i have mistaken to chage my language to spanise.
i want to revert back to english or default.

A simple question: think of shutdown in solaris

i'm a beginner in Solaris. i installed it as a guest OS in virtualbox.
i found these commands: "shutdown, halt, poweroff, init." can't really shutdown the machine as a non-root user.
However, i can just simplely click the button "Shut Down" to shutdown machine on graphic interface as a non-root user.
What did the button do in background? Does it work in terminal?

What version of Solaris or OpenSolaris are you using?
cat /etc/release and post.
alan

Muxer error on WLS shutdown on Solaris 8

Hi,
We're running WLS 5.1.0 (sp9) on Solaris 8 with JDK 1.2.2_07. When we shut WLS
down we get the following muxer ThreadDeath errors at the end of the shutdown:
Tue Oct 30 20:16:08 GMT+01:00 2001: <WebLogicServer> Waiting: '20' seconds
for threads to complete shutdown
Tue Oct 30 20:16:08 GMT+01:00 2001: <SSLListenThread> Shutting down Thread[SSLListenThread,5,]
Tue Oct 30 20:16:08 GMT+01:00 2001: <ListenThread> Shutting down Thread[ListenThread,5,]
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: Before free/total=62504072/67108864
(93%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: After free/total=64015832/67108864
(95%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: Before free/total=64004336/67108864
(95%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: After free/total=64020208/67108864
(95%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Shutting down execute
threads
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Shutdown completed
Tue Oct 30 20:16:28 GMT+01:00 2001:<E> <Performance Pack> Muxer got error: null
java.lang.ThreadDeath
Tue Oct 30 20:16:28 GMT+01:00 2001:<E> <Performance Pack> Muxer got error: null
java.lang.ThreadDeath
Tue Oct 30 20:16:28 GMT+01:00 2001:<E> <Performance Pack> Muxer got error: null
java.lang.ThreadDeath
WLS doesn't then actually shut down so we have to kill the process to get rid
of it. When we try to restart WLS we get the following error (note that I've inserted
asterisks where our IP address was displayed):
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <ListenThread> **** Unable to create server
socket for: ********, port: '7001'. Perhaps the address is incorrect or another
process is using the port.
java.net.BindException: Address already in use
 at java.net.PlainSocketImpl.socketBind(Native Method)
 at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:398)
 at java.net.ServerSocket.<init>(ServerSocket.java:173)
 at weblogic.t3.srvr.ListenThread.newServerSocket(ListenThread.java:215)
 at weblogic.t3.srvr.ListenThread.run(Compiled Code)
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <ListenThread> **** The WebLogic Server
is no longer listening for connections.
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <ListenThread> **** You should probably
restart it.
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <SSLListenThread> **** Unable to create
server socket for: ********, port: '7002'. Perhaps the address is incorrect or
another process is using the port.
java.net.BindException: Address already in use
 at java.net.PlainSocketImpl.socketBind(Native Method)
 at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:398)
 at java.net.ServerSocket.<init>(ServerSocket.java:173)
 at weblogic.security.SSL.SSLServerSocket.<init>(SSLServerSocket.java:67)
 at weblogic.t3.srvr.SSLListenThread.newServerSocket(SSLListenThread.java:473)
 at weblogic.t3.srvr.ListenThread.run(Compiled Code)
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <SSLListenThread> **** The WebLogic Server
is no longer listening for connections.
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <SSLListenThread> **** You should probably
restart it.
Tue Oct 30 19:47:43 GMT+00:00 2001: <WebLogicServer> Server shutdown by system
Tue Oct 30 19:47:43 GMT+00:00 2001: <WebLogicServer> Server shutdown by system
Tue Oct 30 19:48:44 GMT+00:00 2001: <WebLogicServer> WebLogic Server started
Tue Oct 30 19:48:44 GMT+00:00 2001: <WebLogicServer> Server shutdown commencing
NOW
Tue Oct 30 19:48:45 GMT+00:00 2001: <HTTP> Destroying servlets in default servlet
context.
Tue Oct 30 19:48:45 GMT+00:00 2001: <WebLogicServer> Requesting thread shutdown
Tue Oct 30 19:48:45 GMT+00:00 2001: <WebLogicServer> Waiting: '20' seconds
for threads to complete shutdown
Tue Oct 30 19:48:45 GMT+00:00 2001: <SSLListenThread> Shutting down Thread[SSLListenThread,5,]
Tue Oct 30 19:48:45 GMT+00:00 2001: <ListenThread> Shutting down Thread[ListenThread,5,]
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: Before free/total=64963208/67108864
(96%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: After free/total=65159344/67108864
(97%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: Before free/total=65147696/67108864
(97%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: After free/total=65159944/67108864
(97%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Shutting down execute
threads
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Shutdown completed
WLS is the onle process using these ports so it shouldn't hit this problem. Our
only recourse is then to reboot the server itself after which WLS starts okay.
Obviously we don't want to have to reboot the server everytime we shut down WLS.
Has anyone else encountered this problem? Any help greatly appreciated.
Thanks,
Garry

Hi.
Please open a case with support for this issue.
Thanks,
Michael
Garry Baird wrote:
Hi,
We're running WLS 5.1.0 (sp9) on Solaris 8 with JDK 1.2.2_07. When we shut WLS
down we get the following muxer ThreadDeath errors at the end of the shutdown:
Tue Oct 30 20:16:08 GMT+01:00 2001: <WebLogicServer> Waiting: '20' seconds
for threads to complete shutdown
Tue Oct 30 20:16:08 GMT+01:00 2001: <SSLListenThread> Shutting down Thread[SSLListenThread,5,]
Tue Oct 30 20:16:08 GMT+01:00 2001: <ListenThread> Shutting down Thread[ListenThread,5,]
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: Before free/total=62504072/67108864
(93%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: After free/total=64015832/67108864
(95%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: Before free/total=64004336/67108864
(95%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <GC> GC: After free/total=64020208/67108864
(95%)
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Shutting down execute
threads
Tue Oct 30 20:16:28 GMT+01:00 2001: <WebLogicServer> Shutdown completed
Tue Oct 30 20:16:28 GMT+01:00 2001:<E> <Performance Pack> Muxer got error: null
java.lang.ThreadDeath
Tue Oct 30 20:16:28 GMT+01:00 2001:<E> <Performance Pack> Muxer got error: null
java.lang.ThreadDeath
Tue Oct 30 20:16:28 GMT+01:00 2001:<E> <Performance Pack> Muxer got error: null
java.lang.ThreadDeath
WLS doesn't then actually shut down so we have to kill the process to get rid
of it. When we try to restart WLS we get the following error (note that I've inserted
asterisks where our IP address was displayed):
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <ListenThread> **** Unable to create server
socket for: ********, port: '7001'. Perhaps the address is incorrect or another
process is using the port.
java.net.BindException: Address already in use
at java.net.PlainSocketImpl.socketBind(Native Method)
at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:398)
at java.net.ServerSocket.<init>(ServerSocket.java:173)
at weblogic.t3.srvr.ListenThread.newServerSocket(ListenThread.java:215)
at weblogic.t3.srvr.ListenThread.run(Compiled Code)
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <ListenThread> **** The WebLogic Server
is no longer listening for connections.
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <ListenThread> **** You should probably
restart it.
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <SSLListenThread> **** Unable to create
server socket for: ********, port: '7002'. Perhaps the address is incorrect or
another process is using the port.
java.net.BindException: Address already in use
at java.net.PlainSocketImpl.socketBind(Native Method)
at java.net.PlainSocketImpl.bind(PlainSocketImpl.java:398)
at java.net.ServerSocket.<init>(ServerSocket.java:173)
at weblogic.security.SSL.SSLServerSocket.<init>(SSLServerSocket.java:67)
at weblogic.t3.srvr.SSLListenThread.newServerSocket(SSLListenThread.java:473)
at weblogic.t3.srvr.ListenThread.run(Compiled Code)
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <SSLListenThread> **** The WebLogic Server
is no longer listening for connections.
Tue Oct 30 19:47:43 GMT+00:00 2001:<!> <SSLListenThread> **** You should probably
restart it.
Tue Oct 30 19:47:43 GMT+00:00 2001: <WebLogicServer> Server shutdown by system
Tue Oct 30 19:47:43 GMT+00:00 2001: <WebLogicServer> Server shutdown by system
Tue Oct 30 19:48:44 GMT+00:00 2001: <WebLogicServer> WebLogic Server started
Tue Oct 30 19:48:44 GMT+00:00 2001: <WebLogicServer> Server shutdown commencing
NOW
Tue Oct 30 19:48:45 GMT+00:00 2001: <HTTP> Destroying servlets in default servlet
context.
Tue Oct 30 19:48:45 GMT+00:00 2001: <WebLogicServer> Requesting thread shutdown
Tue Oct 30 19:48:45 GMT+00:00 2001: <WebLogicServer> Waiting: '20' seconds
for threads to complete shutdown
Tue Oct 30 19:48:45 GMT+00:00 2001: <SSLListenThread> Shutting down Thread[SSLListenThread,5,]
Tue Oct 30 19:48:45 GMT+00:00 2001: <ListenThread> Shutting down Thread[ListenThread,5,]
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: Before free/total=64963208/67108864
(96%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: After free/total=65159344/67108864
(97%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Invoking garbage collection
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: Before free/total=65147696/67108864
(97%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <GC> GC: After free/total=65159944/67108864
(97%)
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Shutting down execute
threads
Tue Oct 30 19:49:05 GMT+00:00 2001: <WebLogicServer> Shutdown completed
WLS is the onle process using these ports so it shouldn't hit this problem. Our
only recourse is then to reboot the server itself after which WLS starts okay.
Obviously we don't want to have to reboot the server everytime we shut down WLS.
Has anyone else encountered this problem? Any help greatly appreciated.
Thanks,
Garry--
Developer Relations Engineer
BEA Support

How to detect shutdown Solaris within driver?

How within driver to detect rebooting system, but not oriented on standart scrips (rc5, rc6) and standart command (like init 5, init 6) ?

I am very big on following the DDI exactly. I'll tell you a way to cheat, but I'll also tell you not to do it. If you use this, note that your driver could unexpectedly break with any new Solaris release. Watch out with each release to see that the dev_ops definition has not changed in a way that breaks your driver.
The devo_reset function in dev_ops must be set to nodev according to the DDI, but I do know of several drivers that use devo_reset to do cleanup before shutdown.
<PRE>
int
xx_reset(register dev_info_t *const dip, const ddi_reset_cmd_t cmd)
if (cmd != DDI_RESET_FORCE)
return (DDI_FAILURE);
</PRE>

Init 5 doesn't shutdown Solaris 8 Sparc system

Hi,
I am trying to bring down the Solaris 8 Sparc system using the command # init 5 but it doesn't do anything whereas the command # shutdown brings down the system. What could be the problem of init 5 command not functioning.
Thanks & Regards

ok, i should have caught this earlier
looks like, your program is not able to start because of not finding a symbol called __res_state. This symbol is found in /usr/lib/libresolv.so.2
the most likely reason is that u probably do not have some required patches installed
if u r open to install the sun patch cluster , then that is probably the best way to go
visit the below url
http://sunsolve.sun.com/pub-cgi/show.pl?target=patches/patch-access
and look for solaris 9 (around 180 mb of size) download this patch cluster and install them in single user mode
btw, when u have installed the jes version of web server,it is already running only in 32 bit.
hope this helps

Want to shutdown solaris 10 using java application

Greetings all!
I have been unsuccessful at creating my application to shut down a Solaris 10 server with a java application. I am hoping someone might be able to point me in a good direction here.
I want to log on as a regular user and run a java application that monitors remote information (UPS battery voltage) and will gracefully shutdown the server if there is a problem. If I run my application as root all is well, except that I am running as root :)
I can write code to access the shell:
Process proc = Runtime.getRuntime().exec("ls");
BufferedReader Input = new BufferedReader(new InputStreamReader(proc.getInputStream()));
while ((s = Input.readLine()) != null) {out.println(s);}
But I can't do a "su"?
I've tried multiple examples out there of code that allows shell communication, and good examples of using Runtime.getRuntime()
I think the conclusion is that you can't issue a su command via java application.
I've also tried to telnet from a console to localhost and log in as root - no joy there either.
Would someone have information that might point me in the right direction to solve this problem? Even better, if there is a sample snippet to go along with any help.
Thank You very much!
Ray

Wow,
Thank you for the very detailed response! I read the referenced web page last night, I'll read it again today too. RBAC sounds like a pretty complicated administrative tool. Coming from the Win32 domain, most of this security stuff is sadly uncharted territory for me (except for my mac).
Let me see if I understand your method:
Create a role, javarole, this role needs Primary Administrator priviliges (in order to invoke "shutdown -y -i5 -g0") only available to root
Create a powerdown profile, essentially a script:
this line creates the file:
# echo "powerdown:::profile to powerdown server:help=powerdown.html" >> /etc/security/prof_attr
I think this is the actual added script: (instead of the shutdown command just use init 5)
# chmod 755 /usr/sbin/powerdown
# cat > /usr/sbin/powerdown <<EOF
#!/bin/sh
/usr/sbin/init 5
EOF
This line is a bit confusing to me, is it also part of the script?
# echo "powerdown:suser:cmd:::/usr/sbin/powerdown:uid=0" >> /etc/security/exec_attr
I see the uid=0 (root)
It seems obvious to me that I log in a root to create this script, which is the reason it has permission to shutdown...right?
Also, I tried the su command using runTime.exec("su poweroff"); // a user I created to try poweroff scripts
It seems that su is generally not available, I will create the role and try to telnet to it.
What about creating a generic user script that includes the javarole name and password, and then run the script "/usr/sbin/powerdown"
Yes, it leaves a role password exposed, however it is a very limited role, that is not accessible by a user login, but the turn off script is available to every user?
Regards,
Ray

Solaris 9 Automated Remote Shutdown

We have mixed server platforms at our company and not as many ups units as servers. We need to have a process of automatically notifying non-ups connected servers that they must shut down, once ups power is running low. We have found freeware for accomplishing remote shut-down for windows boxes but have seen nothing for solaris that can either be run from a windows box or a solaris box. Any suggestions?

Well... A lot of the new Sun servers have what is called "Lights Out Management" or a "Remote System Controller" (or was it "Console"? I always get that mixed up!)... These will allow you to powerdown the system remotely.
There is a little "but" to this though: you'll have to find a way to make your system's console port accesible through the network. Usually this is done through what is called a "terminal server" or "terminal concentrator". The firm Cyclades is a well known company who makes systems like these (although there are many many other companies as well).
By connecting your Sun's console port to the terminal server (which itself is connected to the network) you will be able to log in directly to your Sun from your Windows workstation. Most of these systems even let you use SSH for extra safety!
From the console you can hop through to the LOM or RSC prompt and tell the system to "power-off", which literally turns it off. Off course, this foregoes any form of proper shutdown, so you'll lose data and so on. Which leads me to this question: why do you not simply log in to the Sun system through the network and run a "shutdown" or "init" command?
Regards,
Thomas

Shutdown normal|immediate slow on new install of 11R2.0.3.{2|3|4} on Solaris

Just installed a new pair of SPARC T4-4 (each w/2x CPUs, 128GB mem) with Solaris 11.1 sru 3.4.1 to run a Data Guard solution for 3x instances using 3x diff. Oracle 11R2 versions (11.2.0.3.2-4)
But all three instances on both pro & sec server randomly indicates Time Drift warnings in their alert logs and vktm traces, but both solaris boxes are running ntp sync fine.
Also various operations like instance startup/shutdown takes a long time, as well as a simple launch of sqlplus can take minutes. Also spawning a simple CLI like 'ps -ef' can sometimes take up to 20-30 sec. Neither swap nor cpu is loaded.
Could it be a known bug seen before or a faulty init ora parameter and/or combination?
Also posted this under Solaris 11 Forum:
simple processes are very slow at intervals
TIA for any hints!
snip of a VKTM trace:
Trace file /global/oraadmin/diag/rdbms/<redacted>/trace/<redacted>_vktm_27415.trc Oracle Database 11g Enterprise Edition Release 11.2.0.3.0 - 64bit Product ion ORACLE_HOME = /global/orabase/product/11.2.0.3.4/db_1 System name: SunOS Node name: <redacted> Release: 5.11 Version: 11.1 Machine: sun4v Instance name: <redacted> Redo thread mounted by this instance: 0 <none> Oracle process number: 3 Unix process pid: 27415, image: oracle@<redacted> (VKTM) *** 2013-06-03 12:51:02.036 *** SESSION ID:(1609.1) 2013-06-03 12:51:02.036 *** CLIENT ID:() 2013-06-03 12:51:02.036 *** SERVICE NAME:() 2013-06-03 12:51:02.036 *** MODULE NAME:() 2013-06-03 12:51:02.036 *** ACTION NAME:() 2013-06-03 12:51:02.036 kstmmainvktm: succeeded in setting elevated priority highres_enabled *** 2013-06-03 12:51:02.036 VKTM running at (1)millisec precision with DBRM quantum (100)ms [Start] HighResTick = 4585953873256 kstmrmtickcnt = 0 : ksudbrmseccnt[0] = 1370256662 *** 2013-06-03 12:55:54.137 kstmchkdrift (kstmhighrestimecntkeeper:highres): Time jumped forward by (2984405)usec at (4586245994477) whereas (1000000) is allowed *** 2013-06-03 12:56:13.269 kstmchkdrift (kstmhighrestimecntkeeper:highres): Time jumped forward by (19133925)usec at (4586265128403) whereas (1000000) is allowed kstmchkdrift (kstmhighrestimecntkeeper:lowres): Time stalled at 1370256973 kstmchkdrift (kstmhighrestimecntkeeper:lowres): Stall, backward drift ended at 1370256974 drift: 1 kstmchkdrift (kstmhighrestimecntkeeper:highres): Time jumped forward by (7023737)usec at (4586963505968) whereas (1000000) is allowed
Mod. action: please, to keep tab formatting on forum side, use the tag
before and after your code output, thanks. lease, Edited by: N Gasparotto on Jun 3, 2013 4:16 PM

Thanks for your suggestion but as I have read the doc then the patch for Bug 9843304 is included in the 11.2.0.2 patchset and then also for PSUs 11.2.0.3.*
I think that the reason we see VKTM trace/drift is not high load as this is not the case, but more lock contention emulating high load. Question is just what lock and why and how to work around it...
Hoped that other SPARC T4 + 11R2 users might have seen similar issue and know what to tweak of either init ora/solaris setting or if bug is known.
Just to show a snap in time. miles varies off course and issue (as mentioned in Solaris 11 forum) is not seen constantly and not during below sampling...
#root:~# top -n 2
last pid: 6594; load avg: 8.07, 6.17, 7.32; up 6+21:28:53 18:04:14
249 processes: 239 sleeping, 5 zombie, 5 on cpu
CPU states: 95.5% idle, 1.0% user, 3.5% kernel, 0.0% iowait, 0.0% swap
Kernel: 67325 ctxsw, 624 trap, 53970 intr, 55554 syscall, 15 flt
Memory: 128G phys mem, 14G free mem, 64G total swap, 64G free swap
 PID USERNAME NLWP PRI NICE SIZE RES STATE TIME CPU COMMAND
20916 oracle 18 20 0 50G 50G cpu/86 146:50 0.77% oracle
6202 oracle 7 51 0 25G 25G cpu/90 0:25 0.18% oracle
prstat -m -U oracle -n 10
 PID USERNAME USR SYS TRP TFL DFL LCK SLP LAT VCX ICX SCL SIG PROCESS/NLWP
21426 oracle 28 0.0 0.0 0.0 0.0 0.0 72 0.0 2 11 39 0 oracle/1
20916 oracle 5.5 0.0 0.0 0.0 0.0 89 5.6 0.0 3 24 8K 0 oracle/18
6180 oracle 2.2 2.3 0.0 0.0 0.0 62 33 0.2 8K 287 20K 0 oracle/7
6212 oracle 1.9 1.9 0.0 0.0 0.0 67 29 0.2 8K 269 19K 0 oracle/8
6196 oracle 1.9 1.8 0.0 0.0 0.0 67 29 0.2 8K 279 20K 0 oracle/8
6209 oracle 1.8 1.8 0.0 0.0 0.0 67 30 0.2 8K 311 19K 0 oracle/8
6178 oracle 1.5 1.0 0.0 0.0 0.0 65 32 0.2 1K 18 1K 0 oracle/7
6202 oracle 1.4 1.0 0.0 0.0 0.0 67 30 0.1 974 18 1K 0 oracle/8
6156 oracle 1.9 0.4 0.0 0.0 0.0 87 11 0.1 7K 193 16K 0 oracle/17
6158 oracle 1.9 0.4 0.0 0.0 0.0 83 15 0.1 5K 217 11K 0 oracle/13
Total: 171 processes, 3808 lwps, load averages: 9.36, 7.14, 7.48
#root:~# plockstat -CH -n 10 -e 5 -p 20916
 0
Mutex hold
Count nsec Lock Caller
 2 22935 libc.so.1`_time_lock oracle`slgtd+0x70
 1 27510 libc.so.1`_time_lock oracle`slgtd+0x70
 2 13120 libc.so.1`libc_malloc_lock oracle`lwemfaa+0x60
 2 8720 libc.so.1`libc_malloc_lock oracle`sltstidinit+0x4
 2 7975 libc.so.1`__aio_mutex libc.so.1`_aio_close+0x44
 2 5150 libc.so.1`libc_malloc_lock oracle`sltsmxd+0x30
 2 4935 libc.so.1`libc_malloc_lock oracle`sltsmxi+0x4
 2 3990 libc.so.1`libc_malloc_lock oracle`sltstiddestroy+0x4
 2 3945 libc.so.1`libc_malloc_lock oracle`slwmmputmem+0x10
 1 5910 libc.so.1`_time_lock oracle`sLdiGetDate+0x6c

Solaris 10 shutdown

Similar Messages

Maybe you are looking for