Hello,
I wasn't sure which forum to post to, so if there is a more appropriate one
then please let me know.
Earlier this week I installed Build 55 on my old E250 to test out ZFS. I setup
a couple of zpools and used SVM to mirror my boot disks. Now I'm trying to run
a LiveUpgrade to Build 56 and the lucreate command has taken almost 45 minutes
just to get to the initial discovery phase:
bash-3.00# lucreate -c snv_55b -m /:/dev/dsk/c0t0d0s5:ufs -n snv_56b
Discovering physical storage devices
Discovering logical storage devices
<It has taken over 45 minutes just to get this far...>
After doing a ptree, I noticed that lucreate is calling on metadb:
8125 lucreate -c snv_55b -m /:/dev/dsk/c0t0d0s5:ufs -n snv_56b
8126 /etc/lib/lu/plugins/lupi_zones plugin
8127 /etc/lib/lu/plugins/lupi_svmio plugin
8162 metadb
8128 /etc/lib/lu/plugins/lupi_bebasic plugin
So I tried running metadb by itself in another shell and after 10 minutes I hit
Ctrl-C since the output never finished. I also tried running metastat and it
finally finished after 31 minutes:
bash-3.00# time metastat
....
....
real 31m9.738s
user 0m0.413s
sys 0m1.529s
I have the following output from mpstat and vmstat which seems to confirm that
nothing else is taking up the cpu.
bash-3.00# mpstat 5
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 5 0 122 1716 1612 293 4 6 5 0 367 1 12 0 87
1 6 0 39 155 152 380 5 6 4 0 470 1 10 0 89
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 1 0 100 401 301 64 0 4 0 0 34 0 2 0 98
1 2 0 3 102 101 51 0 3 1 0 131 0 2 0 97
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 101 402 302 53 0 3 1 0 93 0 3 0 97
1 0 0 0 103 102 53 0 3 0 0 34 0 1 0 99
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 101 401 301 58 0 3 1 0 112 0 4 0 96
1 1 0 2 108 107 58 0 3 1 0 57 0 2 0 98
CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 0 0 101 401 301 53 0 2 2 0 97 0 3 0 97
1 0 0 0 150 149 59 0 2 1 0 31 0 2 0 98
bash-3.00# vmstat 5
kthr memory page disk faults cpu
r b w swap free re mf pi po fr de sr m0 m3 m4 m5 in sy cs us sy id
0 0 0 2425768 586744 3 11 3 2 2 0 0 1 9 -0 1 1869 836 672 1 11 88
0 0 0 1979544 147216 0 4 0 0 0 0 0 0 0 0 0 503 148 109 0 4 96
0 0 0 1979544 147216 0 2 0 0 0 0 0 0 0 0 0 506 165 112 0 2 98
0 0 0 1979544 147216 0 1 0 0 0 0 0 0 0 0 0 502 221 155 0 2 98
0 0 0 1979544 147216 0 2 0 0 0 0 0 0 0 0 0 503 248 155 0 2 97
0 0 0 1979544 147216 0 1 0 0 0 0 0 0 0 0 0 503 140 107 0 2 98
The system in general seems responsive to other commands, just metadb and
metastat are super slow. Any thoughts on what might be causing this? Or other
commands I can try to use for debugging this?
Regards,
-Eric
This message posted from opensolaris.org
