I figured out the problem has to do with PCI cards. The nics work fine as long
as the SATA controller add-in card is not present. The SATA controller works
fine as long as I don't send data to or from the add-in NICS.
I even removed the nics and replaced them with one PCI-E card for testing (als
Yes I did the same steps, and can connect to iscsi and transfer data. With my
last configuration I mapped to dsk instead of rdsk. Still crashes though.
I use Citrix Xenserver for virtualization, this does not play a part in the
problem, as I have got the same results strictly using the Microsof
I did a 'pkg install storage-server'
and then followed http://www.cuddletech.com/blog/pivot/entry.php?id=968 and
the tests that i have done, copying files, running io meter for a few hours has
been fine. I havent tried running any vm's on there yet. will try it when i
have set up the server at
I did this, when the storage server crashed again, it created the
var/crash/ folders, but there are no contents inside?
Is this correct? If so , where is the dump and how do I look at it?
Thanks,
Amos
--
This message posted from opensolaris.org
___
op
Ok, I enable the crash dumping.
I don't understand the "mdb -k unix.n vmcore.n" where n is a number?
A random number?
Thanks.
Amos
> is there a kernwl dump in /var/crash/${hostname}
> if there is that. try running the following command
>
> mdb -k unix.n vmcore.n
>
> where n is a number
> a
On Tue, Dec 8, 2009 at 4:41 PM, Amos Deering wrote:
> Maybe I have done something wrong when I set it up.
>
> I will start from scratch.
>
> iSCSI is really the main thing, the Cifs was just a kicker.
>
> Any pointers on setting up COMSTAR iSCSI, I used the following link for my
> information.
>
Maybe I have done something wrong when I set it up.
I will start from scratch.
iSCSI is really the main thing, the Cifs was just a kicker.
Any pointers on setting up COMSTAR iSCSI, I used the following link for my
information.
http://wikis.sun.com/display/BluePrints/Provisioning+with+iSCSI+and
No, expectations of a stable file server while doing iscsi/cifs/nfs are
reasonable. I wish I had the skills to help you troubleshoot, but sadly I do
not. My only recommendation is to get Sun support if you can swing it, and work
with them towards a solution. My experience with Sun support was po
ok once it came back up, I tried to copy the VM while no VM that resides on the
iscsi storage was running, and it copied for a few seconds then restarted the
solaris server.
Are my expectations of OpenSolaris too much?
--
This message posted from opensolaris.org
OK scratch all I said, after copying one VM, then booting it, then copy another
one while the first is running it crashed both servers again.
I will retest with just the windows initiator again.
--
This message posted from opensolaris.org
___
opensolar
NFS still caused the opensolaris server to reboot once the VM started to copy.
There may just be a NFS configuration / permissions issue as the windows
machine I attempted to test on first was not able to write to it.
COMSTAR is working stable thus far, and at this point it is all going through
ok upgraded to svn_128, I was a little hesatent as I attempted to go to a
previous dev build, and the system would not boot, then I tried again and then
my CLI would not function properly.
This time though it all seems to work. As of now I am able to comunicate with
my Microsoft initiator, the
I had the exact same problem with 2009.06. I worked with Sun support for weeks,
and it came down to a bug that was fixed in svn_124. I upgraded at that time to
124, and the box has been solid since then.
-Scott
--
This message posted from opensolaris.org
___
Just a data point - I have a similar setup here, using snv-127 respin of
OpenSolaris. Both the COMSTAR iSCSI target and the NFS volume are fully
accessible by my Xen 5.5 server without any problems (well, right now I can't
test this with the latest upgrade, as the Xen server has been retired due
If you haven't subscriptions from Sun then you haven't updates for apps and
security updates too. So switch to dev version and test it again.
--
This message posted from opensolaris.org
___
opensolaris-discuss mailing list
opensolaris-discuss@opensolari
There is no /var/crash directory.
a...@oss:~# cd /var/crash
bash: cd: /var/crash: No such file or directory
a...@oss:~# cd /var/crash[K[K[K[K[K[K
a...@oss:/var# dir
admcron games ldlpnfs pkg saf spoolyp
audit dbidmap ldap mail nis preserve samba
oh, one more thing, try upgrading to /dev before filling any bugs, i'm
sure a great many things were fixed since that opensolaris was
released.
To switch to the /dev repository
pfexec pkg set-publisher -O http://pkg.opensolaris.org/dev
opensolaris.org && pfexec pkg refresh --full && pfexec pkg inst
is there a kernwl dump in /var/crash/${hostname}
if there is that. try running the following command
mdb -k unix.n vmcore.n
where n is a number
after that run, in the prompt of mdb, run
$C
That should tell you where the kernel crashed and will give someone
more skilled than i am more informatio
I have recently become aware of the capabilities of OpenSolaris, and am
attempting to setup a Storage Box to give a combination of CIFS and iSCSI
and/or NFS.
I have seemingly succsessfully been able to configure a zfs iscsi share via
iscsitgt, yet the performance is just not up to par, so I dec
19 matches
Mail list logo
