Bug #2967
oi_151a5 seems unusually sluggish, with increased swapping
Status:
Closed
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
Start date:
2012-07-05
Due date:
% Done:
0%
Estimated time:
Difficulty:
Medium
Tags:
needs-triage
Description
as mentioned here (news://news.gmane.org:119/jt49so$f1k$1@dough.gmane.org)
things seem terribly sluggish all of a sudden...
attached is nvidia-shuttleX-bug-report.log.gz which give quite a bit of data
Files
Related issues
Updated by Richard PALO over 8 years ago
Had to revert, too painful to stay on a5...
here's a comparison on a4 with a5 with prstat -ac invocation (using the second display):
on a4:
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
1917 richard 285M 111M sleep 49 0 0:00:38 16% firefox/20
1699 richard 179M 96M sleep 59 0 0:00:22 5,9% Xorg/3
1996 richard 139M 20M sleep 59 0 0:00:00 1,6% gnome-terminal/2
1919 richard 308M 128M sleep 49 0 0:00:35 1,2% thunderbird-bin/20
1911 richard 23M 14M sleep 58 0 0:00:03 1,1% osstat/1
2004 richard 5248K 3204K cpu0 59 0 0:00:00 0,5% prstat/1
1867 richard 72M 59M sleep 59 0 0:00:06 0,5% java/20
1760 root 29M 7688K sleep 59 0 0:00:01 0,3% pkg.depotd/64
1339 root 32M 7764K sleep 49 0 0:00:01 0,3% pkg.depotd/64
6 root 0K 0K sleep 99 -20 0:00:07 0,3% zpool-rpool/136
1869 richard 140M 19M sleep 59 0 0:00:00 0,2% wnck-applet/1
149 root 0K 0K sleep 99 -20 0:00:00 0,1% zpool-apool/136
1857 richard 23M 17M sleep 59 0 0:00:00 0,1% metacity/1
1858 richard 148M 33M sleep 59 0 0:00:02 0,1% gnome-panel/1
1153 zfssnap 15M 5104K sleep 59 0 0:00:00 0,1% time-sliderd/2
NPROC USERNAME SWAP RSS MEMORY TIME CPU
40 richard 483M 742M 36% 0:01:49 27%
73 root 140M 189M 9,2% 0:00:34 1,0%
1 zfssnap 6200K 14M 0,7% 0:00:00 0,1%
7 postgres 51M 54M 2,6% 0:00:00 0,1%
2 netcfg 1908K 5944K 0,3% 0:00:00 0,0%
Total: 138 processes, 862 lwps, load averages: 0,94, 0,88, 0,48
on a5:
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
9992 richard 5256K 3156K cpu0 59 0 0:00:01 5,2% prstat/1
1957 richard 451M 128M sleep 49 0 0:16:58 1,8% thunderbird-bin/23
1929 richard 44M 25M sleep 59 0 0:06:29 1,1% osstat/1
9754 richard 277M 112M sleep 59 0 0:01:07 0,7% firefox/17
1708 richard 2710M 1199M sleep 59 0 0:11:51 0,7% Xorg/3
1911 richard 73M 48M sleep 59 0 0:02:28 0,4% java/20
9840 richard 140M 19M sleep 59 0 0:00:01 0,3% gnome-terminal/2
1743 root 29M 3072K sleep 59 0 0:01:28 0,2% pkg.depotd/64
1521 root 32M 3572K sleep 59 0 0:01:24 0,2% pkg.depotd/64
147 root 0K 0K sleep 99 -20 0:00:50 0,2% zpool-apool/136
6 root 0K 0K sleep 99 -20 0:02:00 0,2% zpool-rpool/136
1160 zfssnap 15M 2716K sleep 59 0 0:00:30 0,1% time-sliderd/2
1023 postgres 13M 2196K sleep 59 0 0:00:13 0,0% postgres/1
1021 postgres 53M 39M sleep 59 0 0:00:11 0,0% postgres/1
556 root 2656K 1544K sleep 59 0 0:00:07 0,0% in.routed/1
NPROC USERNAME SWAP RSS MEMORY TIME CPU
42 richard 3092M 1766M 86% 0:39:52 10%
73 root 136M 56M 2,7% 0:06:34 0,8%
7 postgres 51M 48M 2,3% 0:00:33 0,1%
1 zfssnap 6200K 7908K 0,4% 0:00:30 0,1%
2 netcfg 1908K 2164K 0,1% 0:00:00 0,0%
Total: 140 processes, 876 lwps, load averages: 0,96, 0,79, 0,64
notice SWAP and in particular Xorg? in.routed seems a bit high in comparison as well
here's in.routed on a4
richard@shuttleX:~$ ps auxw | grep routed root 555 0.0 0.1 2664 1636 ? S 07:14:55 0:00 /usr/sbin/in.routed root 1592 0.0 0.1 2524 1584 ? S 07:16:00 0:00 /usr/sbin/in.routed
Updated by Richard PALO over 8 years ago
Quite possibly only a problem on 32bit.
I'll try for awhile on an acer X3200 which is also nvidia...
here is the prstat -ac which is quite different for Xorg et co:
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
2134 richard 248M 136M sleep 49 0 0:00:10 0,7% thunderbird-bin/17
1218 richard 92M 61M sleep 59 0 0:00:06 0,4% Xorg/3
2170 richard 18M 11M sleep 59 0 0:00:05 0,3% multiload-apple/1
2144 richard 233M 111M sleep 49 0 0:00:12 0,2% firefox/21
2254 richard 6112K 3780K cpu1 59 0 0:00:00 0,1% prstat/1
2136 richard 79M 18M sleep 59 0 0:00:00 0,1% gnome-terminal/2
2157 richard 74M 60M sleep 59 0 0:00:04 0,1% java/20
6 root 0K 0K sleep 99 -20 0:00:07 0,1% zpool-rpool/142
2035 root 35M 12M sleep 59 0 0:00:01 0,1% pkg.depotd/64
2143 richard 80M 17M sleep 59 0 0:00:00 0,0% wnck-applet/1
2164 richard 91M 25M sleep 59 0 0:00:00 0,0% isapython2.6/1
2127 richard 22M 15M sleep 59 0 0:00:00 0,0% metacity/1
2128 richard 93M 31M sleep 59 0 0:00:00 0,0% gnome-panel/1
964 zfssnap 14M 5424K sleep 59 0 0:00:00 0,0% time-sliderd/2
2161 richard 7100K 5124K sleep 59 0 0:00:00 0,0% xscreensaver/1
NPROC USERNAME SWAP RSS MEMORY TIME CPU
40 richard 500M 627M 7,9% 0:00:38 2,1%
69 root 121M 156M 2,0% 0:00:23 0,1%
1 zfssnap 5904K 13M 0,2% 0:00:00 0,0%
7 postgres 56M 58M 0,7% 0:00:00 0,0%
2 netcfg 1908K 5968K 0,1% 0:00:00 0,0%
Total: 129 processes, 669 lwps, load averages: 0,11, 0,20, 0,14
richard@x3200:~$
Updated by Richard PALO over 8 years ago
still scratching my head on the 32bit problem, so I thought I would try some live differences, so I rebooted into a5, updated my devzone to a5 for good measure, and noticed that I wasn't swapping near as bad. Took a new nvidia-shuttleX-bug-report.log.gz and am quite surprised to notice the differences (using meld).
Notice in the earlier report the following 'driver not attached', which are present now in the :
sysdef:
*
* Hostid
*
00c2173f
...
fw, instance #0
cpu, instance #0 (driver not attached)
sb, instance #1
used-resources (driver not attached)
iscsi, instance #0
agpgart, instance #0 (driver not attached)
xsvc, instance #0
options, instance #0
pseudo, instance #0
pppt, instance #0 (driver not attached)
zconsnex, instance #1
zcons, instance #0
fct, instance #0 (driver not attached)
stmf, instance #0 (driver not attached)
stmf_sbd, instance #0
nvidia, instance #255
fcp, instance #0 (driver not attached)
srn, instance #0
fm, instance #0
dtrace, instance #0
fasttrap, instance #0
fbt, instance #0
lockstat, instance #0
profile, instance #0
sdt, instance #0
systrace, instance #0
fcsm, instance #0 (driver not attached)
md, instance #0 (driver not attached)
devinfo, instance #0
zfs, instance #0
fcip, instance #0 (driver not attached)
acpippm, instance #0
ppm, instance #0
audio, instance #0
acpi_drv, instance #0
aggr, instance #0 (driver not attached)
arp, instance #0
bl, instance #0 (driver not attached)
bridge, instance #0 (driver not attached)
clone, instance #0
cn, instance #0
conskbd, instance #0
consms, instance #0
cpuid, instance #0
crypto, instance #0
cryptoadm, instance #0
dld, instance #0
dlpistub, instance #0
icmp, instance #0
icmp6, instance #0
ip, instance #0
ip6, instance #0
ipnet, instance #0
ippctl, instance #0 (driver not attached)
ipsecah, instance #0
ipsecesp, instance #0
iptun, instance #0 (driver not attached)
iwscn, instance #0
keysock, instance #0
kmdb, instance #0 (driver not attached)
kssl, instance #0
llc1, instance #0 (driver not attached)
lofi, instance #0
log, instance #0
mm, instance #0
openeepr, instance #0
physmem, instance #0 (driver not attached)
poll, instance #0
power, instance #0
ptc, instance #0
ptsl, instance #0
ramdisk, instance #1024 (driver not attached)
random, instance #0
rts, instance #0
sad, instance #0
simnet, instance #0 (driver not attached)
smbios, instance #0
softmac, instance #0
spdsock, instance #0
sy, instance #0
sysevent, instance #0
sysmsg, instance #0
tcp, instance #0
tcp6, instance #0
tl, instance #0
tzmon, instance #0
ucode, instance #0 (driver not attached)
udp, instance #0
udp6, instance #0
vnic, instance #0
wc, instance #0
vboxdrv, instance #0
vboxnet, instance #0
vboxflt, instance #0
vboxusbmon, instance #0 (driver not attached)
pool, instance #0
sppp, instance #0
sppptun, instance #0
winlock, instance #0 (driver not attached)
nsmb, instance #0
pm, instance #0
ipf, instance #0
smbsrv, instance #0 (driver not attached)
dump, instance #0
fssnap, instance #0 (driver not attached)
kstat, instance #0
ksyms, instance #0
logindmux, instance #0
ptm, instance #0
pts, instance #0
bpf, instance #0
Also, here is a recent prstat -ac
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
1110 richard 367M 180M sleep 59 0 0:05:20 4,0% firefox/19
908 richard 180M 103M sleep 59 0 0:03:30 2,0% Xorg/3
1084 richard 46M 37M sleep 59 0 0:04:54 1,1% osstat/1
8203 richard 5256K 3172K cpu0 59 0 0:00:00 0,7% prstat/1
1052 richard 166M 62M sleep 49 0 0:00:16 0,5% nautilus/1
1060 richard 72M 59M sleep 59 0 0:01:15 0,4% java/20
7471 root 29M 7764K run 59 0 0:00:04 0,2% pkg.depotd/64
1062 richard 139M 19M sleep 59 0 0:00:02 0,2% wnck-applet/1
1138 richard 140M 21M sleep 59 0 0:00:07 0,2% gnome-terminal/2
845 zfssnap 15M 5120K run 59 0 0:00:18 0,1% time-sliderd/2
1048 richard 138M 18M sleep 59 0 0:00:02 0,1% metacity/1
1071 richard 8072K 5636K sleep 59 0 0:00:01 0,1% xscreensaver/1
6 root 0K 0K sleep 99 -20 0:00:39 0,1% zpool-rpool/136
842 postgres 13M 2800K sleep 49 0 0:00:07 0,0% postgres/1
147 root 0K 0K sleep 99 -20 0:00:07 0,0% zpool-apool/136
NPROC USERNAME SWAP RSS MEMORY TIME CPU
42 richard 635M 882M 43% 0:23:58 9,4%
72 root 114M 160M 7,8% 0:01:35 0,4%
7 postgres 51M 54M 2,6% 0:00:18 0,1%
1 zfssnap 6200K 14M 0,7% 0:00:18 0,1%
6 daemon 7228K 8828K 0,4% 0:00:00 0,0%
Total: 139 processes, 817 lwps, load averages: 0,45, 0,47, 0,41
Updated by Richard PALO over 8 years ago
now a5 crashdumps... toxic
Jul 16 19:05:59 shuttleX genunix: [ID 603404 kern.notice] NOTICE: core_log: firefox[1882] core dumped: /var/cores/core.firefox.uid101.1882@1342458298 Jul 16 19:05:59 shuttleX unix: [ID 836849 kern.notice] Jul 16 19:05:59 shuttleX ^Mpanic[cpu0]/thread=e0d48040: Jul 16 19:05:59 shuttleX genunix: [ID 110603 kern.notice] hment_remove() missing in hash table pp=fbf550d0, ht=d4fa1190,entry=0x356 hash index=0xa8d Jul 16 19:05:59 shuttleX unix: [ID 100000 kern.notice] Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074c18 unix:hment_remove+ab (fbf550d0, d4fa1190,) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074c68 unix:hat_pte_unmap+191 (d4fa1190, 356, 10, ) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074d38 unix:hat_unload_callback+165 (dfd24f08, c0300000,) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074da8 genunix:segvn_unmap+19b (e0b2c500, c0300000,) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074df8 genunix:as_free+119 (eac3b278, db695b88,) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074e28 genunix:relvm+c7 (d7f9e900, 400, 0, d) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074e78 genunix:proc_exit+50d (3, b, db695b88, d85) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074e98 genunix:exit+11 (3, b, 0, fe9f21fa) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074ef8 genunix:psig+5bd (34, d8074f44, 0, a3) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074f78 genunix:post_syscall+413 (0, e0d48040, fe800c) Jul 16 19:05:59 shuttleX genunix: [ID 353471 kern.notice] d8074f98 genunix:syscall_exit+47 (e0d48040, 0, e0d480) Jul 16 19:05:59 shuttleX unix: [ID 100000 kern.notice] Jul 16 19:05:59 shuttleX genunix: [ID 672855 kern.notice] syncing file systems... Jul 16 19:05:59 shuttleX genunix: [ID 904073 kern.notice] done Jul 16 19:06:00 shuttleX genunix: [ID 111219 kern.notice] dumping to /dev/zvol/dsk/rpool/dump, offset 65536, content: kernel Jul 16 19:06:09 shuttleX genunix: [ID 100000 kern.notice] Jul 16 19:06:09 shuttleX genunix: [ID 665016 kern.notice] ^M100% done: 92028 pages dumped, Jul 16 19:06:09 shuttleX genunix: [ID 851671 kern.notice] dump succeeded
richard@shuttleX:~$ fmdump -Vp -u d7af342f-d0ed-efaa-ada5-fb1004bda7aa
TIME UUID SUNW-MSG-ID
juil. 16 2012 19:08:16.482002000 d7af342f-d0ed-efaa-ada5-fb1004bda7aa SUNOS-8000-KL
TIME CLASS ENA
juil. 16 19:07:54.4220 ireport.os.sunos.panic.dump_pending_on_device 0x0000000000000000
nvlist version: 0
version = 0x0
class = list.suspect
uuid = d7af342f-d0ed-efaa-ada5-fb1004bda7aa
code = SUNOS-8000-KL
diag-time = 1342458495 224877
de = fmd:///module/software-diagnosis
fault-list-sz = 0x1
fault-list = (array of embedded nvlists)
(start fault-list[0])
nvlist version: 0
version = 0x0
class = defect.sunos.kernel.panic
certainty = 0x64
asru = sw:///:path=/var/crash/shuttleX/.d7af342f-d0ed-efaa-ada5-fb1004bda7aa
resource = sw:///:path=/var/crash/shuttleX/.d7af342f-d0ed-efaa-ada5-fb1004bda7aa
savecore-succcess = 0
os-instance-uuid = d7af342f-d0ed-efaa-ada5-fb1004bda7aa
panicstr = hment_remove() missing in hash table pp=fbf550d0, ht=d4fa1190,entry=0x356 hash index=0xa8d
panicstack = unix:hment_remove+ab (fbf550d0, d4fa1190,) | unix:hat_pte_unmap+191 (d4fa1190, 356, 10, ) | unix:hat_unload_callback+165 (dfd24f08, c0300000,) | genunix:segvn_unmap+19b (e0b2c500, c0300000,) | genunix:as_free+119 (eac3b278, db695b88,) | genunix:relvm+c7 (d7f9e900, 400, 0, d) | genunix:proc_exit+50d (3, b, db695b88, d85) | genunix:exit+11 (3, b, 0, fe9f21fa) | genunix:psig+5bd (34, d8074f44, 0, a3) | genunix:post_syscall+413 (0, e0d48040, fe800c) | genunix:syscall_exit+47 (e0d48040, 0, e0d480) |
crashtime = 1342458360
panic-time = 16 juillet 2012 19:06:00 CEST CEST
(end fault-list[0])
fault-status = 0x1
severity = Major
__ttl = 0x1
__tod = 0x50044a80 0x1cbac450
I'll add another bug report to some nautilus strangeness..
Updated by Ken Mays about 8 years ago
- Status changed from New to Closed
Closing ticket, since this is an old ticket against an older interim release. Richard, seems some issue with hardware with specific interim releases. Best to stay with a reliable install or update to oi_151a7.
Updated by Marcel Telka 5 months ago
- Related to Bug #1034: panic with b148 hment_remove() missing in hash table added