DSM-G600, DNS-3xx and NSA-220 Hack Forum

jcard · 2011-03-26 06:33:16

In the previous 12 months Alt-F has seen 6 releases, which gives an average of
a new release every 2 months.

During the previous year, thanks to many user reports and suggestions,
and to many bug-fixes and improvements, Alt-F has evolved in a way that
means that it is now mature and ready to leave the Beta status, what
will hopefully happens in the next release.

Online documentation, Wiki entries, packages updates and bug fixes will be the priority
for the next release.

Alt-F is intended to be a serious and sustainable alternative to the vendor's firmware,
ready to be used by the average user through its administering web pages.

Enjoy

-What's new:
-kernel 2.6.35.11, as kernel 2.6.36.x had issues with A1 boards
-IPv6 support in kernel and packages that support it
-more robust reloading method, thanks to putterboy findings;
   also, when reloading fails, the box will most often be rebooted instead of freezing.

-filesystem check at boot now does not stops the boot process,
   as fsck is parallelized/serialized and runs in the background
-disk spindown now enforced be sysctrl
-new cleanup service to clean (truncate) log files

-Debian Installer page, to chroot or kexec Debian Squeeze
-Save/Load setting to/from desktop computer
-View/filter/download all existing log files
-Pages that show temperatures show them in Celsius and Fahrenheit
-Status page now uses bar graphs and an AJAX-like refresh mechanism,
   based on original idea and code contributed by Dwight Hubbard
-many improvements and bug fixes (about 280 changesets affecting 424 files since B6)

-New Alt-F packages:
   automatic-0.6.4, forked-daapd-0.13,
   alsa-lib-1.0.18, dbus-1.2.26, libantlr-3.2, libavl-0.3.5, libconfuse-2.6,
   libevent-2.0.10, libgcrypt-1.2.4, libgpg-error-1.6, libunistring-0.9.3, mxml-2.6,
   libcurl-7.19.6, ncurses-5.6, python-2.45, uemacs-4.0.15

-Updated Alt-F packages:
   transmission-2.22, minidlna-1.0.19, mediatomb-0.12.1, ffmpeg-0.6.1,
   avahi-0.6.28, netatalk-2.1.5, cups-1.4.6, hplip-3.11.1,
   openssh-0.9.8.r, vsftpd-2.3.4, rsync-3.0.7, wget-1.11.4, dropbear-0.53, dnsmasq-2.57,
   e2fsprogs-1.41.14, smartmontools-5.40, dosfstools-3.0.11, ntfs-3g-2010.10.2,
   ethtool-2.6.36, Berkeley-db-4.8.30, libiconv-1.13.1, libfuse-2.8.5, libdaemon-0.14,
   tiff-3.9.4, jpeg-8c, taglib-1.6.3, zlib-1.2.5

User who have an already Alt-F flashed box should use the TryIt button
in the Firmware Upgrader web page before flashing this new version.
Users that are using a reloaded Alt-F can proceed as above, in order
to try the new version without messing up with the fun_plug
installation.
The unstable package feed was updated with new and recompiled
packages. Most if not all new packages will only work on 0.1B7.
The old feed is still available under the name unstable-old
The experimental feed was deleted.

mike7 · 2011-03-26 11:03:03

jcard, thank you for updating.

But this version can't recognize my usb disks.

alt-f.log (48 hits)
Line 57: usbcore: registered new interface driver usbfs
Line 57: usbcore: registered new interface driver usbfs
Line 58: usbcore: registered new interface driver hub
Line 59: usbcore: registered new device driver usb
Line 59: usbcore: registered new device driver usb
Line 102: ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
Line 104: orion-ehci orion-ehci.0: new USB bus registered, assigned bus number 1
Line 106: orion-ehci orion-ehci.0: USB 2.0 started, EHCI 1.00
Line 107: hub 1-0:1.0: USB hub found
Line 109: Initializing USB Mass Storage driver...
Line 110: usbcore: registered new interface driver usb-storage
Line 110: usbcore: registered new interface driver usb-storage
Line 111: USB Mass Storage support registered.
Line 120: usbcore: registered new interface driver usbhid
Line 120: usbcore: registered new interface driver usbhid
Line 121: usbhid: USB HID core driver
Line 121: usbhid: USB HID core driver
Line 125: usb 1-1: new high speed USB device using orion-ehci and address 2
Line 125: usb 1-1: new high speed USB device using orion-ehci and address 2
Line 188: usbcore: registered new interface driver usblp
Line 188: usbcore: registered new interface driver usblp
Line 189: usb 1-1: device descriptor read/64, error -110
Line 240: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usbfs
Line 240: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usbfs
Line 241: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver hub
Line 242: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new device driver usb
Line 242: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new device driver usb
Line 285: Mar 26 11:38:48 Dl007 user.info kernel: ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver
Line 287: Mar 26 11:38:48 Dl007 user.info kernel: orion-ehci orion-ehci.0: new USB bus registered, assigned bus number 1
Line 290: Mar 26 11:38:48 Dl007 user.info kernel: orion-ehci orion-ehci.0: USB 2.0 started, EHCI 1.00
Line 291: Mar 26 11:38:48 Dl007 user.info kernel: hub 1-0:1.0: USB hub found
Line 293: Mar 26 11:38:48 Dl007 user.info kernel: Initializing USB Mass Storage driver...
Line 294: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usb-storage
Line 294: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usb-storage
Line 295: Mar 26 11:38:48 Dl007 user.info kernel: USB Mass Storage support registered.
Line 304: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usbhid
Line 304: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usbhid
Line 305: Mar 26 11:38:48 Dl007 user.info kernel: usbhid: USB HID core driver
Line 305: Mar 26 11:38:48 Dl007 user.info kernel: usbhid: USB HID core driver
Line 309: Mar 26 11:38:48 Dl007 user.info kernel: usb 1-1: new high speed USB device using orion-ehci and address 2
Line 309: Mar 26 11:38:48 Dl007 user.info kernel: usb 1-1: new high speed USB device using orion-ehci and address 2
Line 393: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usblp
Line 393: Mar 26 11:38:48 Dl007 user.info kernel: usbcore: registered new interface driver usblp
Line 394: Mar 26 11:38:51 Dl007 user.err kernel: usb 1-1: device descriptor read/64, error -110
Line 519: usbfs /proc/bus/usb usbfs rw,relatime,devgid=9,devmode=664 0 0
Line 519: usbfs /proc/bus/usb usbfs rw,relatime,devgid=9,devmode=664 0 0
Line 519: usbfs /proc/bus/usb usbfs rw,relatime,devgid=9,devmode=664 0 0
Line 569: 31 2 root SW 0 0% 0% [usbhid_resumer]

jcard · 2011-03-26 17:44:50

mike7 wrote:
jcard, thank you for updating.

But this version can't recognize my usb disks.

alt-f.log (48 hits)
Line 57: usbcore: registered new interface driver usbfs
Line 57: usbcore: registered new interface driver usbfs

You should attach the whole alt-f.log file, or give more details

-What revision level is your box board? A1, B1, C1?
-Are you reloading Alt-F or have you flashed it?
-Do you have an external usb drive attached? Try to plug it only after Alt-F starts. Same problem?
-Did the usb drive worked with the previous Alt-F version you used? What version was it?
-Does a flash usb pen works?
-Do you have an usb hub?
-Does the internal drives work?

Joao "Sherlock" Cardoso

PS-I must say that the same happened to me during the release tests, when I was reloading Alt-F from dlink 1.0.9 firmware. I had a 4 port usb hub and a printer attached to it on a rev-B1 box. Everything else worked fine. The problem disappeared as soon as I flashed Alt-F.

mike7 · 2011-03-26 21:59:51

jcard wrote:
PS-I must say that the same happened to me during the release tests, when I was reloading Alt-F from dlink 1.0.9 firmware. I had a 4 port usb hub and a printer attached to it on a rev-B1 box. Everything else worked fine. The problem disappeared as soon as I flashed Alt-F.

Exactly the same, except I don't have a printer. But for the moment I'm not ready to flash.

PS. It worked in Alt-f 0.1B6 with the same configuration.

Files attached...

Attachments:

alt-f.zip, Size: 10,586 bytes, Downloads: 572

SpaceAceX · 2011-03-27 21:43:21

Does this custom firmware work with WD20EARS + Transmission and Automatic?

thebeck · 2011-03-28 17:55:09

jcard wrote:
-New Alt-F packages:
   automatic-0.6.4, forked-daapd-0.13,
   alsa-lib-1.0.18, dbus-1.2.26, libantlr-3.2, libavl-0.3.5, libconfuse-2.6,
   libevent-2.0.10, libgcrypt-1.2.4, libgpg-error-1.6, libunistring-0.9.3, mxml-2.6,
   libcurl-7.19.6, ncurses-5.6, python-2.45, uemacs-4.0.15

-Updated Alt-F packages:
   transmission-2.22, minidlna-1.0.19, mediatomb-0.12.1, ffmpeg-0.6.1,
   avahi-0.6.28, netatalk-2.1.5, cups-1.4.6, hplip-3.11.1,
   openssh-0.9.8.r, vsftpd-2.3.4, rsync-3.0.7, wget-1.11.4, dropbear-0.53, dnsmasq-2.57,
   e2fsprogs-1.41.14, smartmontools-5.40, dosfstools-3.0.11, ntfs-3g-2010.10.2,
   ethtool-2.6.36, Berkeley-db-4.8.30, libiconv-1.13.1, libfuse-2.8.5, libdaemon-0.14,
   tiff-3.9.4, jpeg-8c, taglib-1.6.3, zlib-1.2.5

Would it be possible to rev/include the following in the final Alt-F package list?
rsync: 3.0.8
avahi: 0.6.29
samba: 3.5.8 <-- probably requires glibc (http://www.gnu.org/software/libc/)
mysql: 5.5.10
openssl: 1.0.0.d
openssh: 5.8p1
wget: 1.12

Also, I think when you say "openssh-0.9.8.r", I think you mean to say openssl-0.9.8.r. Am I correct?

Note: netatalk 2.2-beta3 has a serious performance issue (http://sourceforge.net/tracker/?func=de … tid=108642), so I would recommend sticking with 2.1.5 for now and not including netatalk 2.2.x until this issue has been resolved. Once it has been resolved, though, I would love to move to netalk 2.2.x because it has AFP 3.3 protocol support.

Last edited by thebeck (2011-03-28 17:57:52)

jcard · 2011-03-29 20:10:32

SpaceAceX wrote:
Does this custom firmware work with WD20EARS

Although I don't have one myself, reports say that 2TB disks work with Alt-F.
Partitioning is always done on 4KB sectors boundaries/length

+ Transmission and Automatic?

Read the announcement more carefully ;-)

jcard · 2011-03-29 20:28:18

thebeck wrote:
Would it be possible to rev/include the following in the final Alt-F package list?
rsync: 3.0.8
avahi: 0.6.29
samba: 3.5.8 <-- probably requires glibc (http://www.gnu.org/software/libc/)
mysql: 5.5.10
openssl: 1.0.0.d
openssh: 5.8p1
wget: 1.12

Yes, as time goes by... and there is no such thing as a "final Alt-F"

Updating patch or minor releases is generally easy, but updating major releases is not.

Cross-compiling and uClibc means that ./configure/make often fails, and patches have to be developed and tested, and the final app has also to be tested. This means that I have to understand what/how the package does/works, and I don't know all about everything (only a small bit of everything)

Samba is a good example. It took me ages to fit it in the available flash-memory available space.
I started upgrading samba a couple of times, but never finished it. Perhaps for the next release...

Alt-F is not like Debian or other disk-based distribution. Alt-F, if flashed, boots and works without a disk.

Also, I think when you say "openssh-0.9.8.r", I think you mean to say openssl-0.9.8.r. Am I correct?

Right.

Note: netatalk 2.2-beta3 has a serious performance issue (http://sourceforge.net/tracker/?func=de … tid=108642), so I would recommend sticking with 2.1.5 for now and not including netatalk 2.2.x until this issue has been resolved. Once it has been resolved, though, I would love to move to netalk 2.2.x because it has AFP 3.3 protocol support.

Normally I don't use beta or the first major release of a package.

thebeck · 2011-03-31 17:01:58

jcard wrote:
Alt-F is not like Debian or other disk-based distribution. Alt-F, if flashed, boots and works without a disk.

I totally understand that the flash-memory space is quite small. And totally understand that if there were no disks, Alt-F would just work. Considering this, do you think it makes sense for Alt-F to have some sort of packaging scheme? For example, if there are no disks, then what good is samba and netatalk? If some offerings came as a package, then there may be room in the flash-memory for other more core items.

Anyway, I am very much looking forward to getting rid of my d-link firmware. I am going to experiment with the Alt-F reloaded version for a while before I flash. Keep up the GREAT work!!!

fonz · 2011-03-31 17:08:14

thebeck wrote:
If some offerings came as a package, then there may be room in the flash-memory for other more core items.

As far as I understand, you can add ffp or optware "as usual". If Alt-F is using ARM-EABI, the EABI packages from
http://www.inreto.de/testing/ffp/0.6/arm/ might fit pretty well.

thebeck · 2011-03-31 17:48:42

Very true.

I was thinking more core things like nut and a newer busybox (Alt-F come with 1.15, I think).

kncgcg · 2011-03-31 18:33:08

Good job, will always support your fw
Just updated 0.1B7 from 0.1B6, installed ipkg, but Available Packages showing 'None'.
Besides, tried to install ffp, but was 'Download failed'.
i owned a B1 and flashed (not reloading).

jcard · 2011-03-31 21:10:45

fonz wrote:
thebeck wrote:
If some offerings came as a package, then there may be room in the flash-memory for other more core items.
As far as I understand, you can add ffp or optware "as usual". If Alt-F is using ARM-EABI, the EABI packages from
http://www.inreto.de/testing/ffp/0.6/arm/ might fit pretty well.

Alt-F is built with EABI and OABI support, and already has a web page to install ffp-0.5 packages, see a screenshot:
http://sites.google.com/site/altfirmwar … edirects=0

[added] I hope you don't change the directory/pages structure, or it will stop working

Optware packages don't work, as there are incompatibilities with the uClibc libraries; as yours ffp packages uses a hardcoded rpath, they run flawlessly under Alt-F.

The list of current disk-instalable Alt-F packages can be seen here:
http://sites.google.com/site/altfirmware/home/packages

Last edited by jcard (2011-03-31 21:18:15)

jcard · 2011-03-31 21:15:36

kncgcg wrote:
Good job, will always support your fw
Just updated 0.1B7 from 0.1B6, installed ipkg, but Available Packages showing 'None'.
Besides, tried to install ffp, but was 'Download failed'.
i owned a B1 and flashed (not reloading).

It must be a transient network problem, as packages are downloaded from the same site as ipkg itself, and you succeeded installing it.

You can try to hit the DefaultFeed button, and then the UpdateList button.

jcard · 2011-03-31 22:08:36

I just posted the following message in the Alt-F Google group:

Hi,
Is anybody using 0.1B7 with rev-A1 boards?
If yes, can you please see the kernel log (System->Utilities->View
logs) and post it if something looks suspicious?
Even if nothing is suspicious, please answer with a "I'm on a A1, no
problem found"
If I don't get any answer I'm forced to discontinue rev-A1 board
support.
I need at least two persons that offers themselves to test Alt-F pre-
releases on A1 and C1 boards.
Sorry for the bad mood,
João

Please take it in consideration.

konke · 2011-03-31 23:34:42

The new release works like a charm. Thanks
Two questions:
1. Is there any wiki on samba shares in the alt-f, i cant make it work with users and file rights, i want one user to login with samba share and ftp login, password protected. It works now with that all the people on the network (workgroup) can acces all the files on the Dns. I dont think its wrong in alt-f its just me that dont know how to setup samba share. Even tried with SWAT. Verry advanced.
2. I made my two disks as RAID1 (Mirroring) If one of the disks brake down, can a replace that with another disk and the raid keeps working?

I flashed my C1 DNS323 with the B6 release, yesterday i flashed with the new release B7. Dont have any things to complaine about so far. exept for the samba settings

Excuse me for my bad english. Hope you understand.

Last edited by konke (2011-03-31 23:41:38)

jcard · 2011-04-01 05:43:30

konke wrote:
1. Is there any wiki on samba shares in the alt-f, i cant make it work with users and file rights, i want one user to login with samba share and ftp login, password protected. It works now with that all the people on the network (workgroup) can acces all the files on the Dns. I dont think its wrong in alt-f its just me that dont know how to setup samba share. Even tried with SWAT. Verry advanced.

You have to create users using the web interface.

The first time a user is created, you have to select a filesystem where a set of directories are created.

-A directory named Users, where each user has its own directory, only accessible to him using its password. The user directory will have the same name as the users name.
-A "Public Read Only" directory is also created, to put, say, family photos, that all users can read but not write (only 'root' user can write on it)
-A "Public Read Write" directory is also created, so all users can exchange data.

Each of these directories is available though samba and ftp.
You can customize directory permissions using Setup->Directories, but the default permissions for directories and shares should suffice.

Notice that if the user name is the same as the name he uses in MS-windows, say "Joao Cardoso", and the password is also the same, he can access its share from MS-windows without being prompted for a password. This feature (some might call it a bug) is a concession to MSW users.
Note that I'm assuming that each MSW user has a MSW account and user name, which is not always true, blame MS.
For ftp, the nick name has to be used.

Linux user names can't have spaces on them, so the nick name, say "jcard", is used to authenticate it instead.

If you want to setup more directories/shares you have to use the very simple samba web configuration page.
Go to Services->Network->smb->configure and add the directory you want to share in the "Directories to export to other hosts" section.
You have a very limited set of options to choose from, but it might be the base for a more complex setup, either by editing the /etc/samba/smb.conf file (look how other shares are setup), or by using swat.

2. I made my two disks as RAID1 (Mirroring) If one of the disks brake down, can a replace that with another disk and the raid keeps working?

Yes. But it is not an automatic procedure.

While dlink firmware disk-partitioning scheme is a rigid one, and it is possible to automatically know what the system is currently using, Alt-F does not impose any partitioning scheme, it accepts and works with any disk the user plugs in.
As such it is almost impossible to know what the user wants to do when a disk is plugged.
The Disk Wizard is only intended for first-time disk setup.

The procedure to fix a degraded RAID1 disk is simple and is outlined in one the Alt-F Group topics. Soon I will add a wiki entry with those steps, which are all web based.

Of course the RAID will continue serving files while the repair is being done. I even swap disks with the box powered up (don't take this as an endorsement). RAID means availability with no downtime, not "Redundant Array...."

Last edited by jcard (2011-04-01 05:47:02)

thebeck · 2011-04-01 08:11:42

jcard wrote:
I just posted the following message in the Alt-F Google group:

Hi,
Is anybody using 0.1B7 with rev-A1 boards?
If yes, can you please see the kernel log (System->Utilities->View
logs) and post it if something looks suspicious?
Even if nothing is suspicious, please answer with a "I'm on a A1, no
problem found"
If I don't get any answer I'm forced to discontinue rev-A1 board
support.
I need at least two persons that offers themselves to test Alt-F pre-
releases on A1 and C1 boards.
Sorry for the bad mood,
João
Please take it in consideration.

I have an A1 board and a B1 board. Everything seems to work fine reloaded on the A1 board. I have not flashed, though. Attached is the kernel.log from the A1 board. The B1 board is purposed for something else so I have not had a chance to play around with Alt-F on the B1 just yet. Please let me know if you would like me to do anything else on the A1 board.

Attachments:

KernelLog.log, Size: 8,628 bytes, Downloads: 557

SpaceAceX · 2011-04-01 19:04:39

How do you install ALT-F from D-Link stock firmware 1.08?

jcard · 2011-04-01 22:55:10

thebeck wrote:
jcard wrote:
I just posted the following message in the Alt-F Google group:

Hi,
Is anybody using 0.1B7 with rev-A1 boards?
If yes, can you please see the kernel log (System->Utilities->View
logs) and post it if something looks suspicious?
Even if nothing is suspicious, please answer with a "I'm on a A1, no
problem found"
If I don't get any answer I'm forced to discontinue rev-A1 board
support.
I need at least two persons that offers themselves to test Alt-F pre-
releases on A1 and C1 boards.
Sorry for the bad mood,
João
Please take it in consideration.
I have an A1 board and a B1 board. Everything seems to work fine reloaded on the A1 board. I have not flashed, though. Attached is the kernel.log from the A1 board. The B1 board is purposed for something else so I have not had a chance to play around with Alt-F on the B1 just yet. Please let me know if you would like me to do anything else on the A1 board.

So 0.1B7 is working on A1 boards; flashed or not flashed Alt-F is the same, just dressed differently.
I have a B1 board myself, so I don't need any feedback about it.

Thanks a lot.
João

jcard · 2011-04-01 22:56:28

SpaceAceX wrote:
How do you install ALT-F from D-Link stock firmware 1.08?

Take a look on my signature and read the instructions on the site?

skydreamer · 2011-04-03 13:23:07

I can confirm another successful install on A1 board.
One thing that caught my attention with Alt-F-01.B7, when DHCP is used to configure the network card the MTU size is greyed out and defaults to 1500?
MTU can be set higher with the static IP.

Another small comment is in relation to the fsck on the status page, file system maintenance box:
To Alt-F uninitiated like me that have a lot of valuable data on the disks the choice of word "cleaning...step 1: 41%" evokes heart stopping action of wiping out the disks.
I actually panicked the very first time and only calmed down after checking the processes running to see what is going on.
I think it might be better to stick with boring "checking...step 1: 41%" or "verifying...step 1: 41%" but it is not a big deal.

--------------------------

I was testing the performance of Alt-F-01.B7 and Samba transfers.

Copying file 1GB in size was actually faster on Alt-F than stock firmware although only marginaly- 18.6 as opposed to 18.3 MB/s
However reading files was much slower on Alt-F: only at around 17MB/s as compared to 31MB/s on stock firmware.
Is this in line with expectations?
My DNS-323 has 2x2TB ST32000542AS in RAID 1.

Last edited by skydreamer (2011-04-03 20:06:45)

jcard · 2011-04-05 00:56:43

skydreamer wrote:
I can confirm another successful install on A1 board.
One thing that caught my attention with Alt-F-01.B7, when DHCP is used to configure the network card the MTU size is greyed out and defaults to 1500?
MTU can be set higher with the static IP.

I have to try it. The docs I used didn't mention a mtu option for the dhcp stanza.

Another small comment is in relation to the fsck on the status page, file system maintenance box:
To Alt-F uninitiated like me that have a lot of valuable data on the disks the choice of word "cleaning...step 1: 41%" evokes heart stopping action of wiping out the disks.
I actually panicked the very first time and only calmed down after checking the processes running to see what is going on.
I think it might be better to stick with boring "checking...step 1: 41%" or "verifying...step 1: 41%" but it is not a big deal.

OK, you have a point

I have already changed 'clean' to 'check'. But it is also ambiguous, as it also tries to automatically fix inconsistencies -- thus the former use of 'clean': just remove the dirt, but kip the shirt

I was testing the performance of Alt-F-01.B7 and Samba transfers.

Copying file 1GB in size was actually faster on Alt-F than stock firmware although only marginaly- 18.6 as opposed to 18.3 MB/s
However reading files was much slower on Alt-F: only at around 17MB/s as compared to 31MB/s on stock firmware.
Is this in line with expectations?
My DNS-323 has 2x2TB ST32000542AS in RAID 1.

No.

You are using a Gbit network, can you tell us what MTU and filesystem are you using?
And are you talking of MB or MiB? There is a 5% difference between the two, and some apps lie about that.
Are you reading back the same 1GB file you started writing to the nas? Are you sure that no PC caching is involved? With PCs now having 4/8 GB of memory, writing and then immediately reading a file can produce strange figures.

In theory, a raid1 array can read at raid0 speeds, but that depends on the implementation.
The figures you posted for the vendor's firmware shows that it might be the case.

I can't get those read figures for samba (nor NFS) neither for Alt-F nor Debian Squeeze. Anyone has?
The maximum reading I got was 40MB/s for ftp when using a 7000 MTU. (22MBs for MTU=1500)

I have observed that while reading from the box the full network bandwidth was not being used by the box.

A subject to investigate. I have not done any kernel or apps tuning whatsoever, they are using their default settings.

skydreamer · 2011-04-05 04:40:50

Thanks jcard,

MTU setting should be fine with DHCP client- the stock firmware allows for that.

Perhaps "repairing" might be also a good choice?

I am using the highest MTU of 9000, FS ext2.

PC Caching issue is redundant I believe since it would affect both the Alt-F and Stock firmware transfer results and yes, it is the same file.

I did not bother to measure the transfer speed with great accuracy since I wanted to see only the diferrence between the stock FW and Alt-F.

The values that I indicated are actually based on average transfer speed of what Windows 7 was reporting - again because I used the same system for both tests it should not matter.

Also the effect of caching should be dimished when reading from the DNS-323.

I ran the tests again and with the same results but this time I took a snapshot of the running system. First is stock firmware PC->DNS, then DNS->PC followed by Alt-F PC->DNS and DNS->PC.

Could the slower speed be caused by swapping? From the Alt-F interface Swap: 3.9/1031 MB, I have only samba daemon enabled.

Mem: 60324K used, 1488K free, 0K shrd, 10816K buff, 30080K cached
CPU: 9% usr 72% sys 0% nice 0% idle 0% io 0% irq 18% softirq
Load average: 1.07 0.30 0.10
PID PPID USER STAT VSZ %MEM %CPU COMMAND
30890 1350 nobody R 5128 8% 82% smbd -D
49 5 root SW 0 0% 18% [pdflush]
1661 1 root S 6648 11% 0% /web/webs
2015 1 root S 5576 9% 0% /ffp/sbin/unfsd -e /ffp/etc/exports
30859 2061 root S 5024 8% 0% sshd: root@pts/0
1350 1 root S 4828 8% 0% smbd -D
1360 1350 root S 4828 8% 0% smbd -D
1352 1 root S 2968 5% 0% nmbd -D
2061 1 root S 2332 4% 0% /ffp/sbin/sshd
1903 1 root S 1600 3% 0% crond
1 0 root S 1596 3% 0% init
1702 1 root S 1596 3% 0% -sh
30869 30859 root S 1348 2% 0% -sh
30917 30869 root R 1336 2% 0% top
2032 1 root S 1292 2% 0% /ffp/bin/ntpd -x -g -f /ffp/etc/ntp.drift
1397 1 root S 1180 2% 0% xmldb -n config
1693 1 root S 1120 2% 0% op_server 3 3 3
1824 1 root S 1044 2% 0% lpd Waiting
1632 1 root S 920 1% 0% chkbutton
1726 1 root S 780 1% 0% check_hotplug
1857 1 root S 728 1% 0% apkg
1687 1 root S 700 1% 0% fancontrol 0
1856 1 root S 700 1% 0% mserver
2005 1 root S 548 1% 0% /ffp/sbin/rpc.portmap
1806 1 root S 496 1% 0% atd
204 1 root SW 0 0% 0% [mtdblockd]
51 1 root DW 0 0% 0% [kswapd0]
230 1 root SW< 0 0% 0% [loop0]
50 5 root SW 0 0% 0% [pdflush]
4 1 root SW< 0 0% 0% [khelper]
1440 1 root SW 0 0% 0% [md0_raid1]
5 1 root SW< 0 0% 0% [kthread]
11 5 root SW< 0 0% 0% [kblockd/0]
218 5 root SW< 0 0% 0% [kcryptd/0]
219 5 root SW< 0 0% 0% [kmirrord/0]
52 5 root SW< 0 0% 0% [aio/0]
14 1 root SW 0 0% 0% [khubd]
190 1 root SW 0 0% 0% [scsi_eh_0]
191 1 root SW 0 0% 0% [scsi_eh_1]
192 1 root SW 0 0% 0% [scsi_eh_2]
193 1 root SW 0 0% 0% [scsi_eh_3]
2 1 root SWN 0 0% 0% [ksoftirqd/0]

Mem: 60744K used, 1068K free, 0K shrd, 10772K buff, 30776K cached
CPU: 18% usr 43% sys 0% nice 0% idle 0% io 0% irq 36% softirq
Load average: 1.17 0.42 0.15
PID PPID USER STAT VSZ %MEM %CPU COMMAND
30890 1350 nobody R 5128 8% 96% smbd -D
1632 1 root S 920 1% 1% chkbutton
30859 2061 root S 5024 8% 0% sshd: root@pts/0
30934 30869 root R 1340 2% 0% top
1661 1 root S 6648 11% 0% /web/webs
2015 1 root S 5576 9% 0% /ffp/sbin/unfsd -e /ffp/etc/exports
1350 1 root S 4828 8% 0% smbd -D
1360 1350 root S 4828 8% 0% smbd -D
1352 1 root S 2968 5% 0% nmbd -D
2061 1 root S 2332 4% 0% /ffp/sbin/sshd
1903 1 root S 1600 3% 0% crond
1 0 root S 1596 3% 0% init
1702 1 root S 1596 3% 0% -sh
30869 30859 root S 1348 2% 0% -sh
2032 1 root S 1292 2% 0% /ffp/bin/ntpd -x -g -f /ffp/etc/ntp.drift
1397 1 root S 1180 2% 0% xmldb -n config
1693 1 root S 1120 2% 0% op_server 3 3 3
1824 1 root S 1044 2% 0% lpd Waiting
1726 1 root S 780 1% 0% check_hotplug
1857 1 root S 728 1% 0% apkg
1687 1 root S 700 1% 0% fancontrol 0
1856 1 root S 700 1% 0% mserver
2005 1 root S 548 1% 0% /ffp/sbin/rpc.portmap
1806 1 root S 496 1% 0% atd
49 5 root SW 0 0% 0% [pdflush]
204 1 root SW 0 0% 0% [mtdblockd]
51 1 root DW 0 0% 0% [kswapd0]
230 1 root SW< 0 0% 0% [loop0]
50 5 root SW 0 0% 0% [pdflush]
4 1 root SW< 0 0% 0% [khelper]
1440 1 root SW 0 0% 0% [md0_raid1]
5 1 root SW< 0 0% 0% [kthread]
11 5 root SW< 0 0% 0% [kblockd/0]
218 5 root SW< 0 0% 0% [kcryptd/0]
219 5 root SW< 0 0% 0% [kmirrord/0]
52 5 root SW< 0 0% 0% [aio/0]
14 1 root SW 0 0% 0% [khubd]
190 1 root SW 0 0% 0% [scsi_eh_0]
191 1 root SW 0 0% 0% [scsi_eh_1]
192 1 root SW 0 0% 0% [scsi_eh_2]
193 1 root SW 0 0% 0% [scsi_eh_3]
2 1 root SWN 0 0% 0% [ksoftirqd/0]
3 1 root SW< 0 0% 0% [events/0]

Mem: 61156K used, 1020K free, 0K shrd, 2200K buff, 47808K cached
CPU: 11% usr 71% sys 0% nic 0% idle 1% io 0% irq 15% sirq
Load average: 1.22 1.03 0.59 2/40 2813
PID PPID USER STAT VSZ %MEM %CPU COMMAND
2811 2260 nobody R 7920 13% 86% smbd -D
322 2 root SW 0 0% 4% [md0_raid1]
1713 2 root SW 0 0% 3% [flush-9:0]
13 2 root DW 0 0% 2% [kswapd0]
2813 1658 root R 1180 2% 2% top
447 1 root S 580 1% 1% sysctrl
9 2 root SW 0 0% 1% [kblockd/0]
2260 1 root S 7912 13% 0% smbd -D
2345 2260 root S 7912 13% 0% smbd -D
2258 1 root S 6232 10% 0% nmbd -D
434 1 root S 1204 2% 0% syslogd -C -m 0 -D
1658 1657 root S 1192 2% 0% -sh
1657 511 root S 1188 2% 0% dropbear -i
781 1 root S 1188 2% 0% /bin/sh --
1 0 root S 1184 2% 0% init
511 1 root S 1184 2% 0% inetd
436 1 root S 1172 2% 0% klogd
8 2 root SW 0 0% 0% [bdi-default]
4 2 root SW 0 0% 0% [events/0]
53 2 root SW< 0 0% 0% [loop0]
3 2 root SW 0 0% 0% [ksoftirqd/0]
15 2 root SW 0 0% 0% [aufsd_pre/0]
7 2 root SW 0 0% 0% [sync_supers]
11 2 root SW 0 0% 0% [ata_sff/0]
14 2 root SW 0 0% 0% [aufsd/0]
28 2 root SW 0 0% 0% [mtdblock0]
30 2 root SW 0 0% 0% [mtdblock2]
16 2 root SW 0 0% 0% [scsi_eh_0]
17 2 root SW 0 0% 0% [scsi_eh_1]
29 2 root SW 0 0% 0% [mtdblock1]
6 2 root SW 0 0% 0% [async/mgr]
32 2 root SW 0 0% 0% [mtdblock4]
31 2 root SW 0 0% 0% [mtdblock3]
2 0 root SW 0 0% 0% [kthreadd]
5 2 root SW 0 0% 0% [khelper]
19 2 root SW 0 0% 0% [scsi_eh_3]
12 2 root SW 0 0% 0% [khubd]
35 2 root SW 0 0% 0% [usbhid_resumer]
18 2 root SW 0 0% 0% [scsi_eh_2]
10 2 root SW 0 0% 0% [ata_aux]

Mem: 61128K used, 1048K free, 0K shrd, 2348K buff, 47952K cached
CPU: 9% usr 54% sys 0% nic 0% idle 17% io 0% irq 19% sirq
Load average: 1.18 1.00 0.63 2/40 2815
PID PPID USER STAT VSZ %MEM %CPU COMMAND
2811 2260 nobody R 7932 13% 75% smbd -D
9 2 root SW 0 0% 5% [kblockd/0]
322 2 root SW 0 0% 1% [md0_raid1]
13 2 root DW 0 0% 1% [kswapd0]
2815 1658 root R 1180 2% 0% top
1657 511 root S 1188 2% 0% dropbear -i
3 2 root SW 0 0% 0% [ksoftirqd/0]
2260 1 root S 7912 13% 0% smbd -D
2345 2260 root S 7912 13% 0% smbd -D
2258 1 root S 6232 10% 0% nmbd -D
434 1 root S 1204 2% 0% syslogd -C -m 0 -D
1658 1657 root S 1192 2% 0% -sh
781 1 root S 1188 2% 0% /bin/sh --
1 0 root S 1184 2% 0% init
511 1 root S 1184 2% 0% inetd
436 1 root S 1172 2% 0% klogd
447 1 root S 580 1% 0% sysctrl
8 2 root SW 0 0% 0% [bdi-default]
4 2 root SW 0 0% 0% [events/0]
53 2 root SW< 0 0% 0% [loop0]
1713 2 root SW 0 0% 0% [flush-9:0]
15 2 root SW 0 0% 0% [aufsd_pre/0]
7 2 root SW 0 0% 0% [sync_supers]
11 2 root SW 0 0% 0% [ata_sff/0]
14 2 root SW 0 0% 0% [aufsd/0]
28 2 root SW 0 0% 0% [mtdblock0]
30 2 root SW 0 0% 0% [mtdblock2]
16 2 root SW 0 0% 0% [scsi_eh_0]
17 2 root SW 0 0% 0% [scsi_eh_1]
29 2 root SW 0 0% 0% [mtdblock1]
6 2 root SW 0 0% 0% [async/mgr]
32 2 root SW 0 0% 0% [mtdblock4]
31 2 root SW 0 0% 0% [mtdblock3]
2 0 root SW 0 0% 0% [kthreadd]
5 2 root SW 0 0% 0% [khelper]
19 2 root SW 0 0% 0% [scsi_eh_3]
12 2 root SW 0 0% 0% [khubd]
35 2 root SW 0 0% 0% [usbhid_resumer]
18 2 root SW 0 0% 0% [scsi_eh_2]
10 2 root SW 0 0% 0% [ata_aux]

zero13th · 2011-04-05 11:09:20

hi Jcard, thanks for B7.

i want to report that some of config always revert back to default after restarting device

/etc/hosts
/etc/httpd.conf
/etc/samba/smb.conf
/etc/resolv.conf

is it normal? i must manually re-config httpd.conf and smb.conf if i want to access device from another ip segment
btw i'm using mikrotik rb750 router

and if harddisks on standby mode, is it possible to make fan stay off?
syslog attach.

thanks

Last edited by zero13th (2011-04-05 11:10:15)

Attachments:

syslog, Size: 32,234 bytes, Downloads: 563

DSM-G600, DNS-3xx and NSA-220 Hack Forum

Announcement

#1 2011-03-26 06:33:16

Alt-F-0.1B7 is released on Alt-F first anniversary

#2 2011-03-26 11:03:03

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#3 2011-03-26 17:44:50

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

mike7 wrote:

#4 2011-03-26 21:59:51

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

jcard wrote:

#5 2011-03-27 21:43:21

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#6 2011-03-28 17:55:09

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

jcard wrote:

#7 2011-03-29 20:10:32

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

SpaceAceX wrote:

#8 2011-03-29 20:28:18

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

thebeck wrote:

#9 2011-03-31 17:01:58

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

jcard wrote:

#10 2011-03-31 17:08:14

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

thebeck wrote:

#11 2011-03-31 17:48:42

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#12 2011-03-31 18:33:08

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#13 2011-03-31 21:10:45

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

fonz wrote:

thebeck wrote:

#14 2011-03-31 21:15:36

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

kncgcg wrote:

#15 2011-03-31 22:08:36

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#16 2011-03-31 23:34:42

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#17 2011-04-01 05:43:30

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

konke wrote:

#18 2011-04-01 08:11:42

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

jcard wrote:

#19 2011-04-01 19:04:39

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#20 2011-04-01 22:55:10

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

thebeck wrote:

jcard wrote:

#21 2011-04-01 22:56:28

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

SpaceAceX wrote:

#22 2011-04-03 13:23:07

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#23 2011-04-05 00:56:43

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

skydreamer wrote:

#24 2011-04-05 04:40:50

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

#25 2011-04-05 11:09:20

Re: Alt-F-0.1B7 is released on Alt-F first anniversary

Board footer