SNV101a:
Dec 28 01:00:10 s10services2 ^Mpanic[cpu0]/thread=ff0150456800:
Dec 28 01:00:10 s10services2 genunix: [ID 683410 kern.notice] BAD TRAP: type=e
(#pf Page fault) rp=ff000416ac10 addr=a
e1831b5
Dec 28 01:00:10 s10services2 unix: [ID 10 kern.notice]
Dec 28 01:00:10 s10
> Gino wrote:
> >
> > # mdb -k unix.0 vmcore.0
> > Loading modules: [ unix genunix specfs dtrace mac
> cpu.generic cpu_ms.AuthenticAMD.15 uppc pcplusmp
> scsi_vhci zfs sd mpt fcp fctl qlc sockfs ip hook neti
> sctp arp usba stmf idm md cpc random crypto smbsrv
>
sppp nsmb rdc ]
> $c
ip_tcp_input+0x6a(0, 0, ff04dc34d068, 0, ff052e7db408, 0)
ip_accept_tcp+0x7cf(ff04dc34d068, ff04e9519088, ff04e2cddc40,
ff04e80a3bc0, ff001fb94be8, ff001fb94be4)
squeue_polling_thread+0x13f(ff04e2cddc40)
thread_start+8()
>
gino
--
Just happened. Any ideas?
gino
Mar 11 12:48:14 cl1 unix: [ID 836849 kern.notice]
Mar 11 12:48:14 cl1 ^Mpanic[cpu1]/thread=ff001fb94c60:
Mar 11 12:48:14 cl1 genunix: [ID 335743 kern.notice] BAD TRAP: type=e (#pf Page
fault) rp=ff001fb94870 addr=0 occurred in module "ip" due
0 0.0 0 0.0 0 0.0
ohci#0 | 0 0.0 0 0.0 0 0.0 0 0.0
ohci#1 | 0 0.0 0 0.0 0 0.0 0 0.0
qlc#0 | 0 0.0 0 0.0 0 0.0 0 0.0
Any ideas?
tnx,
gino
This
---
tnx,
gino
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
0 30 714 653 756 12 237 2000 14745 91 0 5
2 962 0 48 546 124 807 22 253 1620 32557 4 0 88
3 119 0 48574 1295 10 295 1490 26893 92 0 5
any suggestion?
tnx,
Gino
This message posted from opensolaris.org
disp_lock_exit
24 0% 91% 0.00 3937 cpu[2] bcopy
---
tnx,
Gino
This message posted from opensolaris.org
Hi Sherry,
here you are:
Loading modules: [ unix genunix specfs dtrace cpu.AuthenticAMD.15 uppc pcplusmp
scsi_vhci ufs ip hook neti sctp arp usba qlc fctl nca lofs zfs random fcp sppp
md cpc fcip crypto logindmux ptm ipc nfs ]
> ::cpuinfo
ID ADDR FLG NRUN BSPL PRI RNRN KRNRN SWITCH
61 38 0 1
tnx,
gino
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
red pool.
It's not the first time I had processes not killable in S10.
If I remeber well someone already posted about this problem.
gino
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
Hi All,
S10u2
I was ready for Christmas lunch when our services monitoring system alerted me
that all the zones on a production server were unreachable ..
I logged in the server and found that a process on a zone spawned too much
eating all the ram (we need memory capping!!). Cpus were idling..
SIII cpu and I'm trying to see if company choice to move to Opterons
(based on SPEC numbers over $$ :) ) is really a good choice.
So I'm asking for other opinions/experiences, like yours :)
thanks,
gino
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
> > Also other sysadmin I know found the same when
> switched from Sparc to Opteron...
>
> Then you know the wrong sysadmins.
:|
> Simple test: Compile some files in parallel (dmake -j
> X) on a 2 cpu box:
>
> (1) dmake -j 2
> real 1:18.18
> user 1:37.05
> sys
>> T2000 slow, really slow for that app :(
>for the t2000 to be shown the best light you need to run 24-32 copies of the
>single >threaded app to shine.
I think the performances of my app are deeply depending on cache size.
For example a 4800 with 8 USIII 900Mhz is much faster the
cing them with Opterons.
I've reported some tests done on my old Blade1000 with USIII 750Mhz.
> > gino$ time ./doit
>
> ptime(1m) is supposed to be more reproducable, but
> does not time child processes (are there
> any?) of the command;
no, single process.
> >
aving (All systems wiht S10u1 or S10u2):
gino$ time ./doit
Opteron 848 (2200Mhz) 5m 21s
USIII 750Mhz18m 01s
Xeon 2.4 Ghz35m 12s
P4 2.0 Ghz more than on hour
T2000 slow, really slow fo
4340 74% 74% 0.00 6214 cpu[1] cpu_halt
973 17% 91% 0.00 8159 cpu[0] (usermode)
34 1% 92% 0.00 8373 cpu[0] mutex_enter
28 0% 92% 0.00 6343 cpu[1] fsflush_do_pages
28 0% 93% 0.00 8277 cpu[0]
> Is this behavior consistent everytime you run gcc,
> mysql, ..,? You need to isolate the problem to a
> single process that is triggering this behavior. Some
> of Brendan Gregg's DTrace scripts
> (http://users.tpg.com.au/bdgcvb/dtrace.html) might
> help you in diagnosing the issue. Besides DTrace
Hi All,
solaris 10 6/6, DL585 4x Opteron 880
We are facing high kernel cpu % during normale operation (gcc, mysql, ...)
For example:
CPU states: 45.6% idle, 11.8% user, 42.6% kernel, 0.0% iowait, 0.0% swap
Using DTrace during a simple compilation ( dtrace -n 'syscall:::entry {
@[probefunc]
> Dana H. Myers wrote:
> > Add the following line to /etc/system and reboot:
> >
> > set cpu\.AuthenticAMD\.15:ao_scrub_rate_dram = 0
> Let me add - if you've been trying all sorts of other
> mysterious
> workarounds, this would be a good time to undo them
> ;-)
DL585, undo all magics, up and
luster, then --power off-- the server.
We are running stable since a week in this way ...
gino
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
Oops ... now I've seen:
asy:cpqci_attach_state+30163445 ()
I'm removing cpqci driver ... let's see what's happens ..
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
> Hi Gino,
> the problm is in the "cpqci" module, which I assume
> is something
> from HP/Compaq which you're using to manage your SAN.
>
> This line is particularly worrying:
>
> > Aug 27 14:56:24 fb2 genunix: [ID 655072
> kern.notice] fe8
> >
> > Update: System hung again after 3h 56m, I've
> wasted
> > too much time on this problem already so I'm now
> > installing Solaris 10 01/06.
> >
> > J
> >
> > stejjh
>
> With those settings we have 2 DL585 up and running
> from 25h and 7h ...
Updates after 2 days of testing:
DL585, So
>
> Update: System hung again after 3h 56m, I've wasted
> too much time on this problem already so I'm now
> installing Solaris 10 01/06.
>
> J
>
> stejjh
With those settings we have 2 DL585 up and running from 25h and 7h ...
This message posted from opensolaris.org
Try putting
set pci_autoconfig:pci_boot_debug=1
in /etc/system, and redirecting the console to the serial port:
eeprom input-device=ttya output-device=ttya console=ttya
With those setting one of our DL585 has reached 18h of uptime and going on ...
:)
This message posted from opensolar
> > > Is it possible to enter kmdb when the machine
> hangs,
> > > by typing "F1-A"?
> >
> > unfortunately no. machine is totally frozen.
>
>
> Hmm, maybe it's possible to get a panic dump from the
> hanging machine
> using the "deadman" feature?
>
> See:
> http://blogs.sun.com/roller/page/esax
> > We are unable to investigate more ... hangs
> happens
> > after about 4 hours. Nothing on the logs, nothing
> on
> > the console. Hangs happens with server idle or
> busy.
>
> What kind of console device are you using? PS/2
> keyboard; USB; serial port?
We are using ssh access and also remote
we are experiencing the same here.
did you find a solution?
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
> Gino Ruopolo wrote:
>
> >>I've integrated the fix for CR 6451513, the HP
> >>DL585/DL380 etc. boot hang
> >>into Solaris Nevada b46; I'll backport it into an
> S10
> >>patch/update as soon
> >>as possible. In the meantime, it s
>
> I've integrated the fix for CR 6451513, the HP
> DL585/DL380 etc. boot hang
> into Solaris Nevada b46; I'll backport it into an S10
> patch/update as soon
> as possible. In the meantime, it should show up soon
> in Solaris Express.
>
> Thanks all for the help,
> Dana
>
> ___
> > > What exactly happens when you boot?
> > >
> > > Did you try to boot the kernel with flags "-kv"?
> >
> > cannot mount root path
>
> OK, so it's not hanging, it seems to have a problem
> with missing drivers.
Yes. The "hang problem" is still here with our DL585.
We hope there will be soon
> > Appling driver diskette does nothing. So I stated
> a
> > shell (open 6 on install menu) and executed HP
> > Smartarray drivers installation from diskette.
> > Lots of error but kernel module loaded!
> > So install goes on disks found ... reboot.
> >
> > After reboot the kernel on the dis
> What exactly happens when you boot?
>
> Did you try to boot the kernel with flags "-kv"?
cannot mount root path
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
> > > You probably must modify HP's ITU driver floppy
> > for
> > > use with Solaris Express /
> > > Solaris Nevada / OpenSolaris: rename the
> > DU/sol_210
> > > directory to DU/sol_211.
> >
> >
> > renaming directory doesn't work.
> > I applied drivers using a shell but after a reboot
> > the
> You probably must modify HP's ITU driver floppy for
> use with Solaris Express /
> Solaris Nevada / OpenSolaris: rename the DU/sol_210
> directory to DU/sol_211.
renaming directory doesn't work.
I applied drivers using a shell but after a reboot the solaris install hang
after grub because of
renaming directory doesn't work.
I applied drivers using a shell but after a reboot the solaris install hang
after grub because of missing raid drivers :(
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-disc
Same problems here. DL585 with last firmware.
We have 14 units to upgrade to U2 as we need ZFS.
Now testing linux ... :(
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
39 matches
Mail list logo