That's a very big jump to cause and it's unlikely to be your issue. You need to do some troubleshooting and data collection to understand what's happening on the system when the problem occurs. If you want to monitor interrupts, have a look at intrstat(1M) which will show statistics for the interrupts. The following will show you more detail about how Solaris has configured interrupts:
$ echo "::interrupts" | mdb -k The question you need to ask is "What is the system doing or having done to it when the problem is seen". I'd start by writing a small script to collect the following to individual output files, note you'll also need to capture date & timestamps in each file so you know what the system was doing at the time. It'll make reviewing the data much easier. * prstat -c 5 * vmstat 5 * iostat -xntcz * intrstat 5 * netstat -I <nic> -i 5 That should cover most of the basics. You can add more flags and tune the interval however you like. If you find anything interesting you can drill down using more appropriate flags to the commands or other Solaris commands and/or DTrace. Look for CPU's jumping to 100% for sustained period of time or see if the system is paging heavily. The way Solaris handles it's memory is completely different to Linux. If you have any ZFS filesystems on your system you'll also have the ZILs and ARCs consuming memory so paging could be a likely cause if you're doing I/O when the 'hangs' occur. If you'd prefer a GUI approach to get a quick assessment the "System Monitor" under "Applications -> System Tools -> System Monitor" would show you what's going on at a very high level. Hope it helps. -- This message posted from opensolaris.org _______________________________________________ opensolaris-help mailing list opensolaris-help@opensolaris.org
