Andreas, vmstat should be running constantly, e.g. vmstat 5 > /tmp/vmstat.log then, send the vmstat.log file please.
On Thu, Jan 1, 2009 at 19:08, Andreas Höschler <[email protected]> wrote: > Hi Peter, > >> Please do run vmstat, start 1st vm, then 2nd and send vmstat output. >> Peter > > I can't stop the first vm right now (being used), but here is the > vmstat output with only the first vm running. > > -bash-3.00# vmstat > kthr memory page disk faults > cpu > r b w swap free re mf pi po fr de sr s0 s1 -- -- in sy cs > us sy id > 0 0 5 3506236 94148 38 137 8 20 82 0 17087 14 -0 0 0 809 3668 2720 > 3 12 85 > > Now I am starting the second vm ... > > I see > > 5289 root 168M 154M run 40 0 0:00:58 89% VirtualBox/15 > > with prstat. System is still responsive ... vmstat shows the following! > > -bash-bash-3.00# vmstat > kthr memory page disk faults > cpu > r b w swap free re mf pi po fr de sr s0 s1 -- -- in sy cs > us sy id > 0 0 5 3506536 94332 38 138 8 20 82 0 17061 14 -0 0 0 809 3689 2721 > 3 12 85 > > It takes an eternity for the solaris instance to boot, but host remains > responsive ... Ok, now things get bad. System really slows down. > > -bash-3.00# vmstat > kthr memory page disk faults > cpu > r b w swap free re mf pi po fr de sr s0 s1 -- -- in sy cs > us sy id > 0 0 5 3506480 94324 38 138 8 20 82 0 17054 14 -0 0 0 809 3690 2729 > 3 12 85 > > > -bash-3.00# vmstat > kthr memory page disk faults > cpu > r b w swap free re mf pi po fr de sr s0 s1 -- -- in sy cs > us sy id > 0 0 5 3506452 94316 38 138 8 20 83 0 17074 15 -0 0 0 809 3690 2732 > 3 12 85 > > Solaris instance is still not booted yet. It gives me time out error > messages while starting some services. The host is unresponsive. With > some patience I neverteless can execute another vmstat. > > -bash-3.00# vmstat > kthr memory page disk faults > cpu > r b w swap free re mf pi po fr de sr s0 s1 -- -- in sy cs > us sy id > 0 0 5 3506368 94292 38 138 8 20 83 0 17133 15 -0 0 0 809 3689 2737 > 3 12 85 > > Ok, he machine is not doing much anymore. Solaris instance ist still > noit booted. Entering command son the host terminal is close to > impossible > > -bash-3.00# vmstat > kill 5289 kthr memory page disk > faults cpu > r b w swap free re mf pi po fr de sr s0 s1 -- -- in sy cs > us sy id > 0 0 5 3506164 94244 39 139 8 21 88 0 17911 15 -0 0 0 809 3687 2750 > 3 12 85 > > I kill the second solaris instance with "kill 5289"... After a couple > of seconds the host system is responsive again! > > Any ideas? > > Thanks, > > Andreas > >>>>>> Ray sessions dies and a bit later the complete machine (no ssh >>>>>> possible anymore). I checked the machine with >>>>>> >>>>>> prstat >>>>>> swap -s >>>>>> >>>>>> shortly before its death but found nothing unusal. Any idea what >>>>>> might cause this problem? The Ultra 20 with 4 GB should be able to >>>>>> host these two virtual machines that are doing close to nothing. >>>>>> Anyway both test machines froze after a while. Can anybody confirm >>>>>> this? Ideas to overcome this are very welcome! >>>>>> >>>>> How much memory have you assigned to the VM's? How much remains for >>>>> the host? It was explained here yesterday that each page a guest >>>>> touches is made non-swappable in the host. After a while, there may >>>>> be too little memory left for the host to operate, and since part >>>>> of each VM runs in host process space, they may freeze along with >>>>> other processes in the host. >>>> >>>> I have assigned 800 MB to each guest. With 4 GB RAM this leaves 2.4 >>>> GB >>>> to the OS. Should be sufficient!? >>> >>> Here is the result of my latest tests. I have a Ultra 20 (1 x 250 GB >>> SATA, 4 GB RAM) with a ZFS / Solaris 10 u6 installation and am running >>> VirtualBox 2.1.0. One Solaris 10 image/guest is running fine. When I >>> start the second Solaris 10 image / guest, the machine gets >>> unresponsive (unusable slow). >>> >>> prstat >>> ============================== >>> PID USERNAME SIZE RSS STATE PRI NICE TIME CPU >>> PROCESS/NLWP >>> 2499 root 963M 886M sleep 60 0 3:35:52 0.6% >>> VirtualBox/15 >>> 1288 root 91M 1084K sleep 59 -10 0:01:24 0.2% java/22 >>> 1089 noaccess 163M 680K sleep 59 0 0:01:28 0.2% java/19 >>> 4705 root 490M 419M sleep 59 0 0:03:51 0.2% >>> VirtualBox/15 >>> 1834 utwww 274M 820K sleep 59 0 0:01:04 0.2% java/63 >>> 2082 root 50M 768K sleep 60 0 0:05:33 0.1% Xnewt/1 >>> 4702 root 3768K 196K cpu0 59 0 0:00:00 0.1% prstat/1 >>> 1862 root 24M 4012K sleep 59 0 0:00:34 0.1% Xorg/1 >>> >>> >>> df -h >>> ============================== >>> Filesystem size used avail capacity Mounted on >>> rpool/ROOT/s10x_u6wos_07b >>> 228G 52G 170G 24% / >>> /devices 0K 0K 0K 0% /devices >>> ctfs 0K 0K 0K 0% /system/contract >>> proc 0K 0K 0K 0% /proc >>> mnttab 0K 0K 0K 0% /etc/mnttab >>> swap 3.1G 416K 3.1G 1% /etc/svc/volatile >>> objfs 0K 0K 0K 0% /system/object >>> sharefs 0K 0K 0K 0% /etc/dfs/sharetab >>> /usr/lib/libc/libc_hwcap2.so.1 >>> 222G 52G 170G 24% /lib/libc.so.1 >>> fd 0K 0K 0K 0% /dev/fd >>> swap 3.1G 260K 3.1G 1% /tmp >>> swap 3.1G 32K 3.1G 1% /var/run >>> rpool/export 228G 19K 170G 1% /export >>> rpool/export/home 228G 18K 170G 1% /export/home >>> rpool 228G 35K 170G 1% /rpool >>> 192.168.1.1:/home 34G 16G 18G 48% /home >>> >>> iostat -D >>> ============================== >>> sd0 sd1 nfs1 nfs2 >>> rps wps util rps wps util rps wps util rps wps util >>> 7 7 4.7 0 0 0.0 0 0 0.0 0 0 0.0 >>> >>> iostat -x 4 >>> ============================== >>> device r/s w/s kr/s kw/s wait actv svc_t %w %b >>> sd0 114.7 36.0 3634.8 2318.9 11.9 0.9 84.8 99 90 >>> sd1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 >>> nfs1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 >>> nfs2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0 0 >>> >>> Both machines are configured to use 500 - 600 MB of RAM. What can I >>> do? > > > _______________________________________________ > vbox-users mailing list > [email protected] > http://vbox.innotek.de/mailman/listinfo/vbox-users > _______________________________________________ vbox-users mailing list [email protected] http://vbox.innotek.de/mailman/listinfo/vbox-users
