On Mon, 19 Sep 2016 13:49:16 -0400
Dan McDonald wrote:
>
> Something corrupted the anon cache entry, making it point to something in
> unused/unallocated/un-something hyperspace. The big question is what. This
> is why I asked about your hardware. It wasn't obvious, but
On Mon, 19 Sep 2016 13:33:02 -0400
Dan McDonald wrote:
>
> Generally speaking, this is memory corruption. What caused it I'm not sure,
> however.
>
A bit flip caused by radiation from Cosmos? ;-)
--
Hilsen/Regards
Michael Rasmussen
Get my public GnuPG keys:
michael
On Mon, 19 Sep 2016 13:33:02 -0400
Dan McDonald wrote:
> This process:
>
> R 15776 2671 10 10 0 0x4a004000 ff075dadd0a0 perl
> T 0xff072006ec60
>
>
Only perl running here:
# ps -ef |grep perl
root 2684 2674 3 Sep 17 ?
> On Sep 19, 2016, at 11:51 AM, Michael Rasmussen wrote:
>
> On Mon, 19 Sep 2016 11:35:33 -0400
> Dan McDonald wrote:
>
>>
>> I see you have a vmdump.2 present. Can you make that downloadable
>> somewhere? I'd like to take a closer look.
>>
>
On Mon, 19 Sep 2016 17:51:28 +0200
Michael Rasmussen wrote:
> ftp://ftp.datanom.net/pub/vmdump.2
>
Also via http:
http://ftp.datanom.net/vmdump.2
--
Hilsen/Regards
Michael Rasmussen
Get my public GnuPG keys:
michael rasmussen cc
On Mon, 19 Sep 2016 11:35:33 -0400
Dan McDonald wrote:
>
> I see you have a vmdump.2 present. Can you make that downloadable somewhere?
> I'd like to take a closer look.
>
ftp://ftp.datanom.net/pub/vmdump.2
> The stack trace looks... almost random. It could be a
> On Sep 17, 2016, at 6:08 PM, Michael Rasmussen wrote:
>
> And after reboot:
> Sep 17 21:51:07 nas savecore: [ID 833026 auth.error] Decompress the crash
> dump w
> ith
> Sep 17 21:51:07 nas 'savecore -vf /var/crash/unknown/vmdump.2'
> Sep 17 21:51:08 nas fmd: [ID 377184
On Mar 26, 2015, at 11:24 PM, wuffers m...@wuffers.net wrote:
So here's what I will attempt to test:
- Create thin vmdk @ 10TB with vSphere fat client: PASS
- Create lazy zeroed vmdk @ 10 TB with vSphere fat client: PASS
- Create eager zeroed vmdk @ 10 TB with vSphere web client: PASS!
On Mar 30, 2015, at 4:10 PM, Richard Elling
richard.ell...@richardelling.com wrote:
is compression enabled?
-- richard
Yes, LZ4. Dedupe off.
___
OmniOS-discuss mailing list
OmniOS-discuss@lists.omniti.com
On Mar 30, 2015, at 1:16 PM, wuffers m...@wuffers.net wrote:
On Mar 30, 2015, at 4:10 PM, Richard Elling
richard.ell...@richardelling.com wrote:
is compression enabled?
-- richard
Yes, LZ4. Dedupe off.
Ironically, WRITE_SAME is the perfect workload for dedup :-)
--
On Mar 26, 2015, at 11:47 AM, wuffers m...@wuffers.net wrote:
It looks like I'll have to make do with lazy zeroed or thin provisioned disks
of 10TB+ for my Veeam tests, if it doesn't cause another kernel panic. I'm
hesitant to create these now during business hours (and I shouldn't be..
On Thu, Mar 26, 2015 at 1:05 PM, Dan McDonald dan...@omniti.com wrote:
WRITE_SAME is one of the four VAAI primitives. Nexenta wrote this code
for NS, and upstreamed two of them:
WRITE_SAME is hardware assisted erase.
UNMAP is hardware assisted freeing.
Those are in upstream illumos.
Just remember that only WRITE_SAME and UNMAP are on stock illumos. If you want
the other two, you either get NexentaStor or you start an effort to upstream
them from illumos-nexenta.
Dan
___
OmniOS-discuss mailing list
On Thu, Mar 26, 2015 at 11:37 AM, Dan McDonald dan...@omniti.com wrote:
I mentioned earlier:
I know Nexenta's done a LOT of improvements on this in illumos-nexenta.
It might be time to upstream some of what they've done. I know it's a
moving target (COMSTAR is not a well-written
On Mar 25, 2015, at 11:51 AM, wuffers m...@wuffers.net wrote:
Going to do this as soon as I can.
Solaris docs say to put the following line in etc/system and reboot:
set kmem_flags=0xf
That's correct.
Can't I just set this dynamically like so (so I can potentially skip 2
reboots)?
On Tue, Mar 24, 2015 at 7:44 PM, Dan McDonald dan...@omniti.com wrote:
On Mar 24, 2015, at 7:44 PM, wuffers m...@wuffers.net wrote:
On r151012 since Nov. And yes, the LUs are exposed via COMSTAR.
If it helps to have some kmem_flags set, I can do that and try to
reproduce it in the
, 2015 7:45 PM
To: wuffers
Cc: omnios-discuss
Subject: Re: [OmniOS-discuss] kernel panic kernel heap corruption detected
when creating zero eager disks
On Mar 24, 2015, at 7:44 PM, wuffers m...@wuffers.net wrote:
On r151012 since Nov. And yes, the LUs are exposed via COMSTAR.
If it helps
On Wed, Mar 25, 2015 at 12:09 PM, Dan McDonald dan...@omniti.com wrote:
Can't I just set this dynamically like so (so I can potentially skip 2
reboots)?
echo kmem_flags/W0xf | mdb -kw
No, because those are read at kmem cache creation time at the system's
start.
Ahh, if I RTFM'd the
On Mar 25, 2015, at 2:17 PM, wuffers m...@wuffers.net wrote:
You reproduce this bug by configuring things a specific way, right? I ask
because you seem to have been running okay until you fell down this
particular panic rabbit hole with a particular set of things, correct?
The panic
On Mar 26, 2015, at 12:58 AM, wuffers m...@wuffers.net wrote:
| genunix:kmem_free+1c8 () |
stmf_sbd:sbd_handle_write_same_xfer_completion+14d () |
stmf_sbd:sbd_dbuf_xfer_done+b1 () | stmf:stmf_worker_task+376 () |
unix:thread_start+8 () |
Hmmph. The WRITE_SAME code, huh?
I know
On Wed, Mar 25, 2015 at 2:21 PM, Dan McDonald dan...@omniti.com wrote:
A with-kmem-flags coredump will be very useful.
Here we go. I reproduced this with no load on the SAN, just a DC and
vcenter server up, then created my 10TB disk in the vSphere fat client. As
expected, I got the kernel
Here's a good place to start. It may need to be kicked to the illumos
developer's list, but let's see what we can figure out first.
1.) What revision of OmniOS are you running?
2.) I notice a lot of STMF threads. COMSTAR (aka. STMF) is not the most stable
piece of software in illumos,
On r151012 since Nov. And yes, the LUs are exposed via COMSTAR.
If it helps to have some kmem_flags set, I can do that and try to reproduce
it in the same way, and have the dump accessible.
On Tue, Mar 24, 2015 at 6:41 PM, Dan McDonald dan...@omniti.com wrote:
Here's a good place to start. It
On Mar 24, 2015, at 7:44 PM, wuffers m...@wuffers.net wrote:
On r151012 since Nov. And yes, the LUs are exposed via COMSTAR.
If it helps to have some kmem_flags set, I can do that and try to reproduce
it in the same way, and have the dump accessible.
kmem_flags=0xf + the actual coredump
On May 21, 2014, at 5:53 AM, Svavar Örn Eysteinsson sva...@januar.is wrote:
SNIP!
panicstr = BAD TRAP: type=8 (#df Double fault)
rp=ff04e3069f10 addr=0
panicstack = unix:real_mode_stop_cpu_stage2_end+9de3 () |
unix:trap+ca5 () |
Any thoughts on this one? I can provide some more info if that helps.
The system is all desktop-grade hardware, with a core-i3 540 CPU and
8gigs of (non-ecc) ram. The pool in question is a 3-disk raidz built on
Toshiba DT01ACA3 3T SATA drives attached to the motherboard SATA ports.
The pool was
On Apr 16, 2014, at 12:39 PM, Kevin Swab kevin.s...@colostate.edu wrote:
SNIP!
Traversing all blocks to verify checksums ...
assertion failed for thread 0xfd7fff162a40, thread-id 1: c
SPA_MAXBLOCKSIZE SPA_MINBLOCKSHIFT, file
../../../uts/common/fs/zfs/zio.c, line 226
Abort (core
Doesn't matter where the panic is from -- it's caused by a corrupt block on
the disk.
A vmdump.N would be nice. You're running 008, I see, so I can use an 008 box
to examine the dump.
Dan
___
OmniOS-discuss mailing list
Although the creation of the zpool worked on OmniOS r151006y, the zfs
send I wanted to test on the pool failed with a kernel panic on r151006y
also.
zfs create test/testds
dd if=/dev/zero of=/test/testds/file1 bs=1M count=50
dd if=/dev/zero of=/test/testds/file2 bs=1M count=50
zfs snapshot
On 11/15/13, 5:39 AM, wuffers wrote:
So I'm adding VMware hosts (ESXi 5.5) to my OmniOS ZFS SAN, which are
already hosting some volumes for our Windows 2012 Hyper-V
infrastructure, running over SRP and Infiniband. In VMware, I had
uninstalled the default Mellanox 1.9.7 drivers and installed
When it pours, it rains. With r151006y, I had two kernel panics in quick
succession while trying to create some zero thick eager disks (4 at the
same time) in ESXi. They are now kernel heap corruption detected instead
of anon_decref.
Kernel panic 2 (dump info:
31 matches
Mail list logo