09-25-2002
Since you ask about system downtime, I would assume you did not add any of the patches in single-user mode?
Anytime you have to add a patch, you must check that it does or does not change the kernel. If it does not, then you can probably add it in multi-user mode while the system is quiet. Otherwise, you MUST bring the system to single-user (normally best to go all the way down and back up to single-user).
You may come back and say "But it worked fine on all the other systems." And at times, you can get away with it but it will always bite you at some point (usually when you can least afford it to happen). It's the Murphy factor.
8 More Discussions You Might Find Interesting
1. Windows & DOS: Issues & Discussions
Hi.
I'm trying to load red hat linux 7.2 on my machine so that I can learn how to use it. I'm having a very difficult time. I attempted to install it so that I can dual boot between linux and w2k. But now I can't boot up. I get a message saying that system32/hal.dll is either missing or... (4 Replies)
Discussion started by: paulSF
4 Replies
2. Solaris
Hi. Can anyone offer any help on a corrupt agen.exe.
When I try to run agen.exe from shell I get the message "unrecoverable error occurred"
I am using a Sun Solaris machine with Solaris 3.6.
Where does the agen.exe reside ?
If I copy it from another working machine will it work ??? (2 Replies)
Discussion started by: enzoli
2 Replies
3. UNIX for Advanced & Expert Users
I have a solaris 5.8 system, it wont boot because /usr is corrupt, which means i cannot bring up fsck to run a disk check. Any ideas? (3 Replies)
Discussion started by: frankkahle
3 Replies
4. High Performance Computing
Provides a description of how to set up a Solaris Cluster Express cluster in a VirtualBox on OpenSolaris.
More... (0 Replies)
Discussion started by: Linux Bot
0 Replies
5. Solaris
Hi,
Is it possible to have a Solaris cluster of 2 nodes at SITE-A using SVM and creating metaset using say 2 LUNs (on SAN). Then replicating these 2 LUNs to remote site SITE-B via storage based replication and then using these LUNs by importing them as a metaset on a server at SITE-B which is... (0 Replies)
Discussion started by: dn2011
0 Replies
6. Filesystems, Disks and Memory
I've seen several posts that are similar to what I'm trying to do but no dice. I have a server (SunFire V215) that went belly up. I've set up a new one. I'm getting a corrupt label: wrong magic number error. Everything I'm seeing says label the drives, but if I do I'm afraid I'll lose the data... (0 Replies)
Discussion started by: tnelson
0 Replies
7. Solaris
Problem occurs on one Solaris build. Every time we unzip the Jan CPU, there are several patches that error out (appears to be related to the GIF files). When we unzip the CPU on another Solaris build to a network storage area, we can execute without issue on the original machine.
Any ideas?
... (1 Reply)
Discussion started by: grahamr72
1 Replies
8. Solaris
Hi Gurus
I am not able to find the patching procedure for solaris 10 ( sol10 u11) to latest patchset with sun cluster having failover zones so that same I should follow.
Take an instance, there are sol1 and sol2 nodes and having two failover zones like sozone1-rg and sozone2-rg and currently... (1 Reply)
Discussion started by: nick101
1 Replies
REBOOT(8) System Manager's Manual REBOOT(8)
NAME
reboot - stopping and restarting the system
SYNOPSIS
/sbin/reboot [ -lqnhdarsfRD ]
/sbin/halt [ -lqndars ]
/sbin/fastboot [ -lqndarsRD ]
DESCRIPTION
2.11BSD is started by placing it in memory at location zero and transferring to its entry point. Since the system is not reentrant, it is
necessary to read it in from disk or tape each time it is to be boot strapped.
Rebooting a running system: When the system is running and a reboot is desired, shutdown(8) is normally used to stop time sharing and put
the system into single user mode. If there are no users then /sbin/reboot can be used without shutting the system down first.
Reboot normally causes the disks to be synced and allows the system to perform other shutdown activities such as resynchronizing hardware
time-of-day clocks. A multi-user reboot (as described below) is then initiated. This causes a system to be booted and an automatic disk
check to be performed. If all this succeeds without incident, the system is then brought up for multi-user operation.
Options to reboot are:
-l Don't try to tell syslogd(8) what's about to happen.
-q Reboot quickly and ungracefully, without shutting down running processes first.
-n Don't sync before rebooting. This can be used if a disk or the processor is on fire.
-h Don't reboot, simply halt the processor.
-d Dump memory onto the dump device, usually part of swap, before rebooting. The dump is done in the same way as after a panic.
-a Have the system booter ask for the name of the system to be booted, rather than immediately booting the default system (/unix).
-r Mount the root file system as read only when the system reboots. This is not supported by the kernel in 2.11BSD.
-s Don't enter multi-user mode after system has rebooted - stay in single user mode.
-f Fast reboot. Omit the automatic file system consistency check when the system reboots and goes multi-user. This is accomplished by
passing a fast reboot flag on to the rebooting kernel. This currently prevents the use of -f flag in conjunction with the -h (halt)
flag.
-D Set the autoconfig(8) debug flag. This is normally not used unless one is debugging the autoconfig program.
-R Tells the kernel to use the compiled in root device. Normally the system uses the device from which it was booted as the
root/swap/pipe/dump device.
Reboot normally places a shutdown record in the login accounting file /usr/adm/wtmp. This is inhibited if the -q or -n options are
present. Note that the -f (fast reboot) and -n (don't sync) options are contradictory; the request for a fast reboot is ignored in this
case.
Halt and fastboot are synonymous with ``reboot -h'' and ``reboot -f'', respectively.
Power fail and crash recovery: Normally, the system will reboot itself at power-up or after crashes if the contents of low memory are
intact. An automatic consistency check of the file systems will be performed, and unless this fails, the system will resume multi-user
operations.
SEE ALSO
autoconfig(8), sync(2), utmp(8), shutdown(8), syslogd(8)
3rd Berkeley Distribution May 24, 1996 REBOOT(8)