James R. Leu | 28 Mar 20:14 2014

AOE v83 - unknown device

Hello,

I'm using AOEv83 on my client machines and vbladev20 on my AOE Server.
Occasionally when I start a new vblade on the server, the client will
see the new aoe device, but attempts to mount it result in
mount returning 'unknown device'.  If I relaunch vblade with different
shelf/slot numbers it will mount fine.

After unloading/reloading the aoe module on the client I am able to
use the original shelf/slot numbers without a problem.

In the situation where it fails the minor number is really high.
ex 152.3584

when I use different shelf slot numbers, the minor number
looks like I would expect
ex 152.290

Has anyone else seen this?
Is there a limit to he minor number that aoe uses?

Thank you for your time.

--

-- 
James R. Leu | Systems and Development | INOC | Madison, WI, USA
Service. Not Software.®

------------------------------------------------------------------------------
Ed Cashin | 17 Jan 16:17 2014

Re: ggaoed and ro filesystem during heavy write

On Jan 16, 2014, at 10:42 AM, Lars Täuber <taeuber-cbj5AzTIRAc@public.gmane.org> wrote:

Hi Ed,


Thu, 16 Jan 2014 09:03:14 -0500
Ed Cashin <ecashin-kFqnmIA2sJXQT0dZR+AlfA@public.gmane.org> ==> <jleu-MY4a002iOHE@public.gmane.org> :
The AoE initiator (the side using the storage) called "aoe" does 
retransmit AoE write commands for aoe_deadsecs seconds.

what's the default value for this parameter?

180.

When an old request gets a response after the AoE command has been retransmitted already, it is an “unexpected response”.  You can watch for those by doing a “cat” on /dev/etherd/err.

— 
  Ed
------------------------------------------------------------------------------
CenturyLink Cloud: The Leader in Enterprise Cloud Services.
Learn Why More Businesses Are Choosing CenturyLink Cloud For
Critical Workloads, Development Environments & Everything In Between.
Get a Quote or Start a Free Trial Today. 
http://pubads.g.doubleclick.net/gampad/clk?id=119420431&iu=/4140/ostg.clktrk
_______________________________________________
Aoetools-discuss mailing list
Aoetools-discuss@...
https://lists.sourceforge.net/lists/listinfo/aoetools-discuss
Lars Täuber | 15 Jan 16:52 2014
X-Face
Picon

ggaoed and ro filesystem during heavy write

Hi,

I experience some problems with the latest ggaoed version and a fresh ubuntu 14.04 aoe client (from the
daily snapshots).

http://code.google.com/p/ggaoed/source/list

The kernel version on the client side is 3.13.0-3-generic

# modinfo aoe
filename:       /lib/modules/3.13.0-3-generic/kernel/drivers/block/aoe/aoe.ko
version:        85
description:    AoE block/char driver for 2.6.2 and newer 2.6 kernels
author:         Sam Hopkins <sah@...>
license:        GPL
srcversion:     5F0AC5D858A1164C5170585

The client is a testing box but the server is in productive state for years. So I can't change the server config.

I did a tcpdump and see that the server stops sending a response to the last write request of a series of write requests.
9 seconds after the client waited for responses without receiving any paket from the target it issues a
"Query Config Information Request" and marks the device as read only. This results in a read-only filesystem.
The responses to the "Query Config Information Requests" can be seen right after the requests.

I can "repair" this with an aoe-revalidate and remounting rw.
But this appears to happen right with the next longer write operation.

I'm stuck here.

It seems the client doesn't resend unresponded requests. Is this on purpose?

Thanks
Lars

------------------------------------------------------------------------------
CenturyLink Cloud: The Leader in Enterprise Cloud Services.
Learn Why More Businesses Are Choosing CenturyLink Cloud For
Critical Workloads, Development Environments & Everything In Between.
Get a Quote or Start a Free Trial Today. 
http://pubads.g.doubleclick.net/gampad/clk?id=119420431&iu=/4140/ostg.clktrk
Ed Cashin | 21 Oct 17:04 2013

Re: AOE sync problem

On Oct 21, 2013, at 10:53 AM, Jesse Becker wrote:

> On Mon, Oct 21, 2013 at 04:30:55PM +0200, devzero@... wrote:
>> yes, at least.
>> 
>> perhaps i would even think about adding a hint via printk`s in the driver code ("this is a block device
probably shared with other nodes, be warned not to use an ordinary filesystem with it"), but i´m not sure
if that would get accepted upstream.
> 
> If the mounts are read-only, I don't see why you *couldn't* mount it
> to multiple systems.  In this case, you could easily use a "normal"
> filesystem.  I can think of a few cases where this would be beneficial,
> even.  It's the SysAdmin's job to make sure things are handled properly
> so you don't trash you data.
> 
> That said, I completly agree with having a note in the Readme file, or
> even a "WARNING_README_FIRST" file.

I have been surprised in the past to observe writes being performed to the block device when the filesystem
had been mounted read only.  Because not many people do this, and those who do insist that it's safe, I
haven't needed to look into the matter more closely.

Regarding the idea for a printk warning, the aoe driver doesn't examine the contents of the data that the
Linux block layer wants read or written.  I think that a more natural place to generate a warning would be in
the individual file systems.  A file system like ext3 has the ability to store information to the block
device that specifies what host is using the block device exclusively.  It would mean patching all the
different file systems, though.

I'll make a note to update the README file.  Thanks, all, for the input.

--

-- 
  Ed Cashin
  ecashin@...

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk
Picon

AOE sync problem

Hi, we set up AOE on our systems, then we mounted the exported partitions but unfortunately the changes on mounted partitions are not reflected on the main exported partition. Also other systems can not see the changes. Everything seems alright; except synchronization.

Also i red in a blog post that we must use jumbo frames but i do not use jumbo frames. can this be the source of the problem??
I checked the network traffic using "tcpdump ether proto 0x88a2" in the result of this command i saw bunch of frames with destionation addresses like bc:5f:f4:6b:af:58 that does not exist in my local network!!! A part of output looks like the following:

09:23:42.000905 30:85:a9:92:ba:ff (oui Unknown) > 1c:6f:65:7f:42:14 (oui Unknown), ethertype Unknown (0x88a2), length 1060:
    0x0000:  1000 0000 0000 5a27 a1ed 4000 0224 0000  ......Z'.. <at> ..$..
    0x0010:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0020:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0030:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0040:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0050:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0060:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0070:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0080:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0090:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x00a0:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x00b0:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x00c0:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x00d0:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x00e0:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x00f0:  0000 0000 0000 0000 0000 0000 0000 0000  ................
    0x0100:  0000 0000 0000 0000 0000 0000 0000 0000  ................


As you can see most of the transferred frame is just zero!!! Any suggestions??
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk
_______________________________________________
Aoetools-discuss mailing list
Aoetools-discuss@...
https://lists.sourceforge.net/lists/listinfo/aoetools-discuss
Serge Leschinsky | 14 Oct 23:11 2013
Picon

build error ( gcc 4.8.1, glibc 2.18, kernel 3.10.14)

Hello,

I was constructing a new build environment (gcc 4.8.1 , glibc 2.18)  and found 
out that build of aoe6-84 / aoe6-85pre1 fails with the following error:

make[1]: Entering directory `/data/tmp/linux-3.10'
   CC [M]  /data/tmp/aoe6-84/linux/drivers/block/aoe/aoedev.o
In file included from /data/tmp/aoe6-84/linux/drivers/block/aoe/aoedev.c:15:0:
/data/tmp/aoe6-84/linux/drivers/block/aoe/aoe.h:16:0: warning: "DEFINE_SPINLOCK" 
redefined [enabled by default]
  #define DEFINE_SPINLOCK(x) spinlock_t x = SPIN_LOCK_UNLOCKED
  ^
In file included from include/linux/spinlock.h:81:0,
                  from include/linux/seqlock.h:29,
                  from include/linux/time.h:5,
                  from include/uapi/linux/timex.h:56,
                  from include/linux/timex.h:56,
                  from include/linux/sched.h:17,
                  from include/linux/blkdev.h:4,
                  from /data/tmp/aoe6-84/linux/drivers/block/aoe/aoedev.c:8:
include/linux/spinlock_types.h:84:0: note: this is the location of the previous 
definition
  #define DEFINE_SPINLOCK(x) spinlock_t x = __SPIN_LOCK_UNLOCKED(x)
  ^
In file included from /data/tmp/aoe6-84/linux/drivers/block/aoe/aoedev.c:15:0:
/data/tmp/aoe6-84/linux/drivers/block/aoe/aoe.h:16:43: error: 
'SPIN_LOCK_UNLOCKED' undeclared here (not in a function)
  #define DEFINE_SPINLOCK(x) spinlock_t x = SPIN_LOCK_UNLOCKED
                                            ^
/data/tmp/aoe6-84/linux/drivers/block/aoe/aoedev.c:26:8: note: in expansion of 
macro 'DEFINE_SPINLOCK'
  static DEFINE_SPINLOCK(devlist_lock);
         ^
make[2]: *** [/data/tmp/aoe6-84/linux/drivers/block/aoe/aoedev.o] Error 1
make[1]: *** [_module_/data/tmp/aoe6-84/linux/drivers/block/aoe] Error 2

Thanks,
Serge

------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60135031&iu=/4140/ostg.clktrk
David Leach | 3 Oct 20:36 2013
Picon

Question about identify device ATA command in vblade

Can someone explain why the ATA command Identify Device (0xEC) has the following test in the switch statement:

    case 0xec:        // identify device
        if (p->sectors != 1 || ndp < 512)
            return -1;


Given that the ATA spec says that the sector register is N/A (as are pretty much all the rest) when making this command I don't know why the blade server requires that submission of this command have AoE ATA command set sector to 1?

I have my suspicions but I would like to hear from others...

David
------------------------------------------------------------------------------
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
the latest Intel processors and coprocessors. See abstracts and register >
http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk
_______________________________________________
Aoetools-discuss mailing list
Aoetools-discuss@...
https://lists.sourceforge.net/lists/listinfo/aoetools-discuss
Derick Swanepoel | 6 Sep 15:16 2013
Picon

Poor performance on 10 Gbps SAN

Hi,

I'm in the process of building a 10 Gbps Ethernet SAN and would really like to use AoE, but I'm experiencing
poor performance (320 MB/s sequential reads and writes). The access pattern on this SAN will be almost
exclusively sequential so I am interested in maximizing throughput, not IOPS.

The status quo:
* RAID 60 (1 MB stripe size) on target capable of 900 MB/s write and 1400 MB/s read on ext4 (sequential).
* Jumbo frames (9000) configured and working on target, initiator and Dell PowerConnect 8132 switch (with
hardware flow control enabled).
* iperf benchmarks TCP performance at 9.87 Gb/s.
* aoetools-36, vblade-21 and kernel 3.10.7 with AoE v81

I'm using the following tuning parameters as suggested on this list:
net.core.rmem_default = 278528
net.core.rmem_max = 278528
net.core.wmem_default = 278528
net.core.wmem_max = 278528
net.core.netdev_max_backlog = 10000

I exported the RAID 60 as e0.0. Both read and write performance is disappointing, though:

# dd if=/dev/etherd/e0.0 of=/dev/null bs=1M count=4K
4096+0 records in
4096+0 records out
4294967296 bytes (4.3 GB) copied, 15.9536 s, 269 MB/s

# dd if=/dev/zero of=/dev/etherd/e0.0 bs=1M count=4K conv=fdatasync
4096+0 records in
4096+0 records out
4294967296 bytes (4.3 GB) copied, 13.6141 s, 315 MB/s

Read speed is even worse than write, which is odd. To rule out any doubts about RAID performance I exported a 4
GB ramdisk as e0.1, but the results are just as bad:

# dd if=/dev/etherd/e0.1 of=/dev/null bs=1M count=4K
4096+0 records in
4096+0 records out
4294967296 bytes (4.3 GB) copied, 15.1406 s, 284 MB/s

# dd if=/dev/zero of=/dev/etherd/e0.1 bs=1M count=4K conv=fdatasync
4096+0 records in
4096+0 records out
4294967296 bytes (4.3 GB) copied, 13.3615 s, 321 MB/s

dd, vblade and the aoe kernel threads are well below 100% CPU usage during these tests. What am I missing here?

Thanks in advance,
Derick
------------------------------------------------------------------------------
Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more!
Discover the easy way to master current and previous Microsoft technologies
and advance your career. Get an incredible 1,500+ hours of step-by-step
tutorial videos with LearnDevNow. Subscribe today and save!
http://pubads.g.doubleclick.net/gampad/clk?id=58041391&iu=/4140/ostg.clktrk
Erlend Hoel | 9 Jul 23:27 2013
Picon

[aoetools-discuss] Client being able to interact with local server

Hello, all!

First of all, please accept my humble apologies if this is the number
one question you don't like to get.  I am under the impression that
there have been others who have wished for this functionality, and
from what I gathered, they were told it didn't exist because it was
never a design goal.

I understand and accept that.  And for all I know, perhaps there are
some weighty and valid reasons why this should be so.  I am
nevertheless in a situation where I desire this functionality (let's
just say that experimentation rates higher on my list of priorities
than performance)

Would this be very hard to implement?  And, related to what I
mentioned above:  Is there a reason why one would _not_ wish to
include this functionality in the standard distribution?

Thank you for your time. :-)

Best regards,
Erlend

------------------------------------------------------------------------------
See everything from the browser to the database with AppDynamics
Get end-to-end visibility with application monitoring from AppDynamics
Isolate bottlenecks and diagnose root cause in seconds.
Start your free trial of AppDynamics Pro today!
http://pubads.g.doubleclick.net/gampad/clk?id=48808831&iu=/4140/ostg.clktrk
Fortunato Cuizon | 12 Jun 05:09 2013
Picon

Help with win32 port of vblade

I'm trying to port vblade to win32 using the winpcap library as a personal project, done it by just recoding something on linux.c...and compiled without problems on the latest orwell dev c++, however when the server is run, it can be seen on most aoe initiators, the problem is that when mounting the said "blade", it would not automount as an extra drive on windows while the original vblade server would have and no problems...this is just the problem i'm getting and if it would work, it's likely to work on mounting/booting an img file...any thoughts?

vblade-19, winpcap 4.12, orwell dev c++ (latest)

-changed getpkt, putpkt with wpcap related function
-changed putsec, getsec with ReadFile, WriteFile C functions

ported it okay, but with mounting problems, img booting (winxp) just okay on first part, booting to protected mode fails...(just there with the 'knight rider' status bar, nothing happening)

Regards

------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:

Build for Windows Store.

http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
Aoetools-discuss mailing list
Aoetools-discuss@...
https://lists.sourceforge.net/lists/listinfo/aoetools-discuss
Tim Dossett | 24 May 17:11 2013
Picon

Re: Read only image

I've never gotten the WinAoE driver to work with Windows 7, but others claim to have; perhaps someone on the list can help.

On Fri, May 24, 2013 at 7:38 AM, Angelica Delgado <angelicadel230-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

We tested the patch but the image gets stuck on aoe driver during boot.  It is Windows 7 64bit.

On May 20, 2013 9:59 AM, "Tim Dossett" <timothy.dossett-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
I don't know if AoE supports Windows 7 (last I checked WinAoE driver didn't), but it is possible to run several Windows XP PCs from 1 base image following this approach: http://etherboot.org/wiki/appnotes/cow


On Mon, May 13, 2013 at 8:51 PM, Angelica Delgado <angelicadel230-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

We want to run vblade in read only mode for a windows 7 image.  Do you all know if it is possible? We want this virtual drive to be shared by multiple computers?

Thanks.
Angelica


------------------------------------------------------------------------------
AlienVault Unified Security Management (USM) platform delivers complete
security visibility with the essential security capabilities. Easily and
efficiently configure, manage, and operate all of your security controls
from a single console and one unified framework. Download a free trial.
http://p.sf.net/sfu/alienvault_d2d
_______________________________________________
Aoetools-discuss mailing list
Aoetools-discuss-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/aoetools-discuss



------------------------------------------------------------------------------
Try New Relic Now & We'll Send You this Cool Shirt
New Relic is the only SaaS-based application performance monitoring service 
that delivers powerful full stack analytics. Optimize and monitor your
browser, app, & servers with just a few lines of code. Try New Relic
and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may
_______________________________________________
Aoetools-discuss mailing list
Aoetools-discuss@...
https://lists.sourceforge.net/lists/listinfo/aoetools-discuss

Gmane