LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Solaris / OpenSolaris (https://www.linuxquestions.org/questions/solaris-opensolaris-20/)
-   -   receiving error, and then machine rebooting (https://www.linuxquestions.org/questions/solaris-opensolaris-20/receiving-error-and-then-machine-rebooting-190879/)

evilchild 06-07-2004 07:40 PM

receiving error, and then machine rebooting
 
Hello all,
I am running a fresh install of Solaris 9 on a SPARCstation 20 with 128mb of ram, occasionally i received the following error "panic[cpu0]/thread=f64900a0: fatal asynchronous fault" then after that the machine reboots. I'm not sure exactly what I'm doing that causes this but it has occurred why browsing with Netscape, when port scanning from another machine, and when removing a directory from "/". What could this be? and how can i fix it.

Thanks Evilchild

jlliagre 06-08-2004 03:15 AM

If you haven't already done it, enable crash dump (kernel core files in /var/crash/<hostname>), cf. dumpadm -y.

Then wait for a new panic, and use one of the available tools to analyse the crash dump, iscda.sh or CAT:
http://wwws.sun.com/software/downloa.../3fce7df0.html

evilchild 06-08-2004 02:11 PM

I belive crash dump is already enabled, (because after a crash it writes files unix.17 and vmcore.17, and the numbers go up on every crash, in /var/crash/<my hostname>/) but being a new to Solaris, I'm not sure what tools to use when analysing the crash dumps, or which files are the dumps.

jlliagre 06-08-2004 03:58 PM

Just use those I recommend to you ... both are well documented.
You're right, the unix.* and vmcore.* are the crash dumps to examine.

evilchild 06-08-2004 09:25 PM

I understand now, but what is the path to these tools or do I need to download them, I would run "find" but i can't run it with out the machine crashing.

Thanks for your help

jlliagre 06-09-2004 02:06 PM

None of the tools I wrote about are installed by default, I gave you the link to download the second tool in my first reply, please follow it to get it.

If your host is so unstable than you can't issue a few commands without the system crashing again, try:
strings vmcore.17 |more
In the first lines, there may be a message giving hints on what happens.
You definitely need to try also:
/usr/platform/`uname -i`/sbin/prtdiag
and see if hardware problems are reported (likely, as the panic seems to happen at random times).


All times are GMT -5. The time now is 03:21 AM.