Alt-F does not load

[vijay]

Hello,

I must have tried at least 10 times without any success. Once it
looked like there was some fsck-ing going on for a while (because I
could not shutdown from within Dlink fw). But I so far have not seen
the ALT-F welcome screen. I keep going back to the D-link scrn.

1. First I tried copying the two files to my "Volume_1" and rename to
Alt-f.sh and run (since I have ffp working). This is the one that
seemed to do the fsck (or so I think). But, upon restart, back to
dlink scrn.

2. Second, since I have ffp, I renamed the 'ffp' dir to 'xffp" and
renamed the fun_plug to xfun_plug. Copied the Alt-F back to the root,
restart, and back to Dlink scrn!

I have tried the above two things multiple times.

Basically, one of my drives went bad, and Dlink will simply not
rebuild even tho I put a new drive. I do not want to reformat
everything, and have to reestablish all my data (750GB!).

Btw, I am a total Linux newbie. I must be doing something basically
wrong.

Can someone please help? I will really appreciate it. Thank you.

[".:: Júlio Leal ::."]

Hi! There!!

Well I have one DNS-323 Rev C1, and I lost a couple of times, to put the Alt-f up and running.

Of course I don't know if this your case, but just when disabled uPNP, the "iTunes Server" (I don't remenber the name of the service), and I also delete one add-one, that I installed befone (NFS Service). I also had the ffp working, but I deleted before try the alt-f.

So, I copy the Alt-f*B7.tar and fun_plug to Volume_1 directory, and through the GUI, I reboot the NAS, and after that the Alt-F come up, in my case I have two disks with 1,5 TB in Raid 1 and after the first boot the alt-f expend some time doing a fsck. But after that, my md0 was mounted, and I could
see the data.

Cheers,

J�lio C�sar Leal

vijay escreveu o seguinte no dia 09/04/11 23:09:

[vijay]

Julio,

That DID it! The only service I had on was the UPnP server. I shut it
down. And, right away on a reboot, I could see how quickly Alt F took
control of the box unlike Dlink's "Flash, Flash, Flash, Flash, ..."
and finally come online.

Now, I need some guidance on how I can force ALT F to build (transfer)
data from my old RAID 1 disk to the new disk. I did not see a ready
option to do that.

Also, I do not know which disk is md0 and md1. But, only one has the
"raid 1" red tag on it, presumably the old disk. Also, Dlink must have
made two partitions - sdb1 (998GB), sdb3 (2GB). So, do I have to
create a new md2 RAID disk, and then ask ALT F to duplicate data.
Anyway, I need to read some documentation.

I will also read up on the other thread about "RAID 1 Degraded".

Thank you again,
Vijay

[Dwight Hubbard]

Name the alt_f funplug script fun_plug and reboot the nas or use funplug_manager if you want to multi-boot funplugs.  

Reloading generally works most reliably if it happens before any services or fun-plugs get started (I've only gotten it to work from the command line very rarely)

[Joao Cardoso]

Now, I need some guidance on how I can force ALT F to build (transfer)

If you need further help you have to describe your current situation in more detail.

 

data from my old RAID 1 disk to the new disk. I did not see a ready
option to do that.

Also, I do not know which disk is md0 and md1.

Raid arrays, like md0, md1, etc,  reside on both disks, not on a single disk.

In the RAID page you can see the raid type and components of each raid array, like sda2, sdb2... if a raid1 array is in degraded mode, then either one of its two components has failed, and it should be shown in red, or is missing (only one component is shown)

E.g., a normal situation:

md0 18.6 GB raid1 sda3 sdb3

md1 35.4 GB raid0 sda4 sdb4

sda3/sdb4 are the components of raid1 md0. sda3 is the third partition of disk 'sda' -- you can see which disk 'sda' is in the status page.

1-Now, If I deliberately fail sda3, selecting partition sda3 under "Component Operations"->partition and then, also under "Component Operations"->operation "Fail", it looks like this

md0 18.6 GB raid1 sda3 sdb3

md1 35.4 GB raid0 sda4 sdb4

(raid1 and sda3 are in red, raid1 degraded, sda3 failed)

2-If I now repeat the above but "Remove" the component, it appears like this (raid1 in red)

md0 18.6 GB raid1 sdb3

md1 35.4 GB raid0 sda4 sdb4

3-If I repeat the above but now "Add" the component again everything is back to normal (I had previously added an intent-bitmap to the array, recovering is much faster, "Raid Operation"->"Create Bitmap")

Inserting a new disk and partitioning it with Disk->Partitioner, with one of the partitions being of type RAID, and with the same size of the other still available raid *component* (sdb3 in this case),  allows you to do step 3. Don't forget to always create a partition of type swap on new disks, 500MB is fine; USB attached disks don't benefice from swap partitions.

As I don't have an extra disk to show you, i did the following: I repeated 1 and 2, then I went to to Disk->Partitioner, unchecked the sda3 partition "Keep" checkbox, made sure it is of type RAID, then hit the "Partition" button (only the unchecked partitions will be affected by the partitioning) and afterwards in the RAID page I selected "Add" sda3, and the following appears (sda3 is now in green, a spare component until recovering finished, raid1 is still in red, degraded)

md0 18.6 GB raid1 sda3 sdb3 recovering

md1 35.4 GB raid0 sda4 sdb4 

But, only one has the
"raid 1" red tag on it, presumably the old disk. Also, Dlink must have
made two partitions - sdb1 (998GB), sdb3 (2GB). So, do I have to
create a new md2 RAID disk, and then ask ALT F to duplicate data.
Anyway, I need to read some documentation.

I will also read up on the other thread about "RAID 1 Degraded".

 

Read also the "Windows cannot see my dns-323 with Alt-F firmware" topic; from post 9 it address RAID issues.

[cdv...@gmail.com]

Greetings Joao,

Thank you for your offer to help.

I just saw your response. It is a tad late for me in the night to
respond plus I need to reconfigure AltF since DNS is back to Dlink
f/w.

I scanned your response -- quite detailed (thanks). I need to
understand this. One thing jumped out. I do *not* have an "sda" device
at all. It should be because it is the new disk on the right hand side
(when looking at the front panel)!

But, I had formatted it in Windows (for ext 2). I guess I need to
repartition and format in Alt F and then add it.

But allow me to post some screenshots of this in the next day or two.

Regards and thanks again,
Vijay

> --
> You received this message because you are subscribed to the Google Groups
> "Alt-F" group.
> To post to this group, send email to al...@googlegroups.com.
> To unsubscribe from this group, send email to
> alt-f+un...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/alt-f?hl=en.
>

[cdv...@gmail.com]

Hello Joao,

Finally, here is some update....Summary: It seems to have worked but
now cannot boot into Alt-F.

Following your accurate guidelines, I was able to add the second disk
(actually sda, on the right). Then, under Disk Partition, I copied my
sdb's partition table (which already included a "swap") to sda. I
unchecked all "Keep" in sda, and hit Partition. Then, I "Add"ed the
sda2 (998gb) to md0, and got the "Recovering" msg. I guess this means
"Rebuilding"? It worked overnight. In the morning, I checked the sda
disk and found all data was copied over to sda from sdb. That is
super! Here are a few details.

Partition right disk, 1000.2 GB, Hitachi HDS721010CLA332
Keep Dev Start sector Lenght Size (GB) Type
sda1
sda2
sda3
sda4
Free: 0.003

Filesystems
Dev. Size FS Label Mount Options FS Operations New FS Operations
md0 929.0GB ext2
sda3 1.5GB none
sda4 0.5GB none
sdb4 486.2MB ext2

RAID Maintenance
Dev. Capacity Level Components Array RAID Operations Component Operations
md0 929.0 GB raid1 sda2 sdb2
md1 sdb3

1. BUT, I ran into a few problems. After I saw all the data, I turned
it off. And it rebooted into Stock f/w. Now, I want to get back, and
it just won't boot back up with the Alt F f/w. When I try to delete
the folder (Alt-F) on my Volume_1, I get this msg:

some "image file is in use..."

I figured that my older ffp (which I had enabled in the AM after the
"recovering") might be stopping this, I have since removed that folder
on both drives outside of DNS - directly on the drives.

2. The funny thing is, since morning, my stock F/W that boots up all
the time cannot shutdown the system!

3. Assuming it is something wrong I am doing, once I get it back up,
how do I get my old ffp to load up (since I put in a lot of time
getting Mediatomb to work, and not so perfectly)?

4. I don't understand -- in Windows, when I check Properties of each
volume), I am shown 1.19 TB. But my disk is only 1TB (and partition
size is 998 GB).

Thanks,
Vijay

[Joao Cardoso]

The sda3 and sda4 partitions might be causing problems with the reloading.
You can see in the raid session that another raid exists. Dlink
creates a jbod raid with free space left from disks created with
raid1. Now that array is not starting as it misses the sda3 component.
Also sda4 has no filesystem on it, and dlink uses to store some info in there.

> RAID Maintenance
> Dev.    Capacity        Level   Components              Array   RAID Operations         Component Operations
> md0     929.0 GB        raid1   sda2 sdb2
> md1                     sdb3
>
> 1. BUT, I ran into a few problems. After I saw all the data, I turned
> it off. And it rebooted into Stock f/w. Now, I want to get back,

Do you have some alt-f log file? Reloaded, fail?
Does the fw offers to format the disks?
Do you have the ONCE commented on fun_plug?
fun_plug exec permissions? After all the raid/disks manipulation I
would double check them all.

You can try to boot only with only the 'original' disk...

> it just won't boot back up with the Alt F f/w. When I try to delete
> the folder (Alt-F) on my Volume_1, I get this msg
> some "image file is in use..."

Is the fs mounted read-only? Can you create and them delete some other file?

[cdv...@gmail.com]

Hello,

You were right -- with only one "original" disk, I could boot into
Alt-F, and I see this:

Filesystems
Dev. Size FS Label Mount Options FS Operations New FS Operations
md0 929.0GB ext2

sda4 486.2MB ext2

RAID Maintenance
Dev. Capacity Level Components Array RAID Operations Component Operations
md0 929.0 GB raid1 sda2

md1 sda3

Plus I see in the attached ALT-F HomePage-1diskonly.png in Disk >
Partition (note this sda used to be sdb in a 2-disk setting).

Plus I could create and delete a file on the disk - so FS is R/W.
Also, Dlink f/w offers to set the RAID type and reformat the TWO
drives when I log into it.

Now, when I inserted the second disk (hot-plug)...I see the img shown
in Alt-F 2disks.png and the 2disks-FS page.

And, I got a bunch of these in the system log.
Apr 14 18:00:29 DNS user.notice hot: md0 waiting to be fscked
Apr 14 18:00:39 DNS user.notice hot: md0 waiting to be fscked
Apr 14 18:00:49 DNS user.notice hot: md0 waiting to be fscked

Running Processes:
Mem: 21848K used, 40328K free, 0K shrd, 584K buff, 15096K cached
CPU: 0% usr 9% sys 0% nic 90% idle 0% io 0% irq 0% sirq
Load average: 0.00 0.04 0.05 1/35 3630

I am not sure what I need to do next here...seems like I have to
repartition the new drv and "Add" again? But, it might develop the
same problem.

Thanks,
Vijay

[Joao Cardoso]

Pictures are low res, icouldnt understand them, better copy/paste as text.
if you can, you have to add sdb3 to md1. This depends what kind md1
is. If it does not appear as a raid component you have to partition it
(and only it) as type raid. If you have no data on md1 you can just
destroy md1 array. Dont know what dl fw will do.
To solve the waiting to be fsck message, search in the forum, i posted
two messages regarding the subject. Avoid reboots.

The general idea is to try to left the box as dl fw expects to find
it: md0 as raid1 with all its components, md1 as jbod also with all
its components, sda1 sdb1 as swap, sda4/sdb4 as ext2/3 fs.

The half-dozen dl fw closed source program makes us to guess what it
will do and expect. That was another motivation for Alt-F.

Cant do serious work on a smartphone. Or I'm too old...

>> creates a jbod raid with free space left from disks created wi> --

[cdv...@gmail.com]

Hello Joao,

Your first idea worked. I destroyed md1 because reformatting sdb3 and adding to md1 with an existing sda3 did not work.

Then, I destroyed md0 too, but recreated it. Then it "recovered" and is now fine.

1. Question: I flashed the settings and rebooted. But, it went back to Dlink f/w. I did not get an error either when I saved to Flash. So, anyway, I recopied Alt F files, and booted back up in to it. I get it! I will still need to boot Alt F up the old way, but the settings (password, raid etc.) are loaded from Flash. Yes?

2. Question: I already had "FFP"; so I could see all the pkgs loaded. But, I could not get mediatomb to start even thru telnet cli (log shows "egiga0 not found"). The pkg is reported to be installed. Any ideas how I can make that work as before? Because, after getting RAID1 to work, flashing my settings to Alt F, the thing I will need will be the mediatomb.

Here are a few reports fyi:

Status page (something new I see in RED: -23 mounts..." ext2 RO..":

Swap: 93.0/942 MB Uptime: 0 day(s) 0 hour(s) Date: Mon Apr 18 12:26:02 NONE 2011

Network Speed: 1000Mb/s MTU: 1500 TX: 17.6 KiB Rx: 98.4 KiB MAC: 00:22:B0:64:01:BA

Disks
Bay Dev. Model Capacity Power Status Temp Health
right sda Hitachi HDS721010CLA332 1000.2 GB active or idle 40°C/104°F passed
left sdb Hitachi HDT721010SLA360 1000.2 GB active or idle 45°C/113°F passed

RAID
Dev. Capacity Level State Status Action Done ETA
md0 929.0 GB raid1 clean OK idle

Mounted Filesystems
Dev. Label Capacity Available FS Mode Dirty Automatic FSCK in
sdb3 3.0GB 3.0GB ext2 RO * -23 mounts or 175 days
sda3 1.5GB 1.4GB ext2 RW * 26 mounts or 179 days
sdb4 486.2MB 472.7MB ext2 RW * 29 mounts or 179 days
sda4 485.8MB 458.4MB ext2 RW * 33 mounts or 179 days

Filesystem Maintenance
Part. Label Operation
md0 cleaning...step 1: 10%

3. Question: Just last night, Alt F went thru a "cleaning" operation. Now, again, same thing...because of reboot?

Also, I am positive I had set the size of the partition of sdb3 to 1.5GB. But, Alt F seems to have doubled it!

Another report:

Filesystems
Dev. Size FS Label Mount Options FS Operations New FS Operations

md0 929.0GB cleaning...step 1: 22%
sda3 1.5GB ext2
sda4 485.8MB ext2
sdb3 3.0GB ext2 ro
sdb4 486.2MB ext2

RAID Maintenance
Dev. Capacity Level Components Array RAID Operations Component Operations

md0 929.0 GB raid1 sda2 sdb2

I appreciate the patience and the assistance you have been providing me. I will offer my assistance to you by writing in an organized way for your wiki about my experience.

Thank you,
Vijay

> For more options, visit this group at http://groups.google.com/group/alt-f?hl=en.
>
>

[Joao Cardoso]

On Monday, April 18, 2011, <cdv...@gmail.com> wrote:
> Hello Joao,
>

> Your first idea worked. I destroyed md1 because reformatting sdb3 and adding to md1 with an existing sda3 did not work.
>
> Then, I destroyed md0 too, but recreated it. Then it "recovered" and is now fine.
>
> 1. Question: I flashed the settings and rebooted. But, it went back to Dlink f/w.

How is ONCE set in fun_plug?

> I did not get an error either when I saved to Flash. So, anyway, I recopied Alt F files, and booted back up in to it. I get it! I will still need to boot Alt F up the old way, but the settings (password, raid etc.) are loaded from Flash. Yes?

Yes. But some settings they might be out of sync if disc contents has changed.

>
> 2. Question: I already had "FFP"; so I could see all the pkgs loaded. But, I could not get mediatomb to start even thru telnet cli (log shows "egiga0 not found"). The pkg is reported to be installed. Any ideas how I can make that work as before? Because, after getting RAID1 to work, flashing my settings to Alt F, the thing I will need will be the mediatomb.

Or you solve the egiga0 problem (its called eth0 in Alt-F) or you
uninstall that ffp package and install Alt-F mediatomb package.

> Here are a few reports fyi:
>
> Status page (something new I see in RED: -23 mounts..." ext2 RO..":

Filesystem are mounted RO (read only) when they can't be fsck
automatically and manual intervention is needed. Take a look at the
System Log.

> Swap: 93.0/942 MB Uptime: 0 day(s) 0 hour(s) Date: Mon Apr 18 12:26:02 NONE 2011
>
> Network Speed: 1000Mb/s MTU: 1500 TX: 17.6 KiB Rx: 98.4 KiB MAC: 00:22:B0:64:01:BA
>
> Disks
> Bay Dev. Model Capacity Power Status Temp Health
> right sda Hitachi HDS721010CLA332 1000.2 GB active or idle 40°C/104°F passed
> left sdb Hitachi HDT721010SLA360 1000.2 GB active or idle 45°C/113°F passed
>
> RAID
> Dev. Capacity Level State Status Action Done ETA
> md0 929.0 GB raid1 clean OK idle
>
> Mounted Filesystems
> Dev. Label Capacity Available FS Mode Dirty Automatic FSCK in
> sdb3 3.0GB 3.0GB ext2 RO * -23 mounts or 175 days
> sda3 1.5GB 1.4GB ext2 RW * 26 mounts or 179 days
> sdb4 486.2MB 472.7MB ext2 RW * 29 mounts or 179 days
> sda4 485.8MB 458.4MB ext2 RW * 33 mounts or 179 days
>
> Filesystem Maintenance
> Part. Label Operation
> md0 cleaning...step 1: 10%
>
> 3. Question: Just last night, Alt F went thru a "cleaning" operation. Now, again, same thing...because of reboot?

Unclean reboot, yes. Othetwise cleaning would be a matter of seconds.

> Also, I am positive I had set the size of the partition of sdb3 to 1.5GB. But, Alt F seems to have doubled it!

I would need the disks partition table to diagnose it. But not now.
If your disks were partitioned with dlink, then a md1 JBOD should
exists, with sda3/sdb3 as its components. Jbod is evil!

>
> Another report:
>
> Filesystems
> Dev. Size FS Label Mount Options FS Operations New FS Operations
> md0 929.0GB cleaning...step 1: 22%
> sda3 1.5GB ext2
> sda4 485.8MB ext2
> sdb3 3.0GB ext2 ro
> sdb4 486.2MB ext2

I would "upgrade" the ext2 to ext3. Dlink also recognizes ext3, that
could be the reason for the checking after reboot ( if no unclean
shutdiwn was made). Start with less data important filesystems, just
to see how it works.

>
> RAID Maintenance
> Dev. Capacity Level Components Array RAID Operations Component Operations
> md0 929.0 GB raid1 sda2 sdb2
>
> I appreciate the patience and the assistance you have been providing me. I will offer my assistance to you by writing in an organized way for your wiki about my experience.

Thanks. You can also help others replying to their posts (if you are
sure of your answers:)

> Thank you,
> Vijay
>
>
>
> On Fri, Apr 15, 2011 at 6:27 AM, Joao Cardoso <joao.fs...@gmail.com> wrote:
>> Pictures are low res, icouldnt understand them, better copy/paste as text.
>> if you can, you have to add sdb3 to md1. This depends what kind md1
>> is. If it does not appear as a raid component you have to partition it
>> (and only it) as type raid. If you have no data on md1 you can just
>> destroy md1 array. Dont know what dl fw will do.
>> To solve the waiting to be fsck message, search in the forum, i posted
>> two messages regarding the s
>

[cdv...@gmail.com]

Hello,

I got this problem now that after another 3 hours of "recovering", I cannot see the directories on my RAID (that is under md0). Last few times, I was able to.

find: /mnt/sda3/.systemfile: Input/output error

/mnt/md0	Owner	Group	Permissions
md0	root	root	rwxrwxrwx
sda3	root	root	rwxrwxrwx
sda4	root	root	rwxrwxrwx
sdb3	root	root	rwxrwxrwx
sdb4	root	root	rwxrwxrwx

Status Page:

System

Temperature 43.0°C/109.4°F	Fan speed 2908	Load 0.00	CPU 17%
Swap: 0.0/1035 MB	Uptime: 0 day(s) 3 hour(s)	Date: Tue Apr 19 15:30:19 NONE 2011

Network

Speed: 1000Mb/s MTU: 1500 TX: 2.4 MiB Rx: 7.8 MiB MAC: 00:22:B0:64:01:BA

Disks

Bay	Dev.	Model	Capacity	Power Status	Temp	Health

left

sda

Hitachi HDT721010SLA360

1000.2 GB

active or idle

42°C/107.6°F	passed
right	sdb

Hitachi HDS721010CLA332

1000.2 GB

active or idle

37°C/98.6°F

passed

Mounted Filesystems

sda3

3.0GB

3.0GB

ext2

RO

*

-30 mounts or 174 days

sda4

486.2MB

472.7MB

ext2

RW

*

29 mounts or 179 days

sdb3

1.5GB

1.4GB

ext2

RW

*

26 mounts or 179 days

sdb4

485.8MB

458.4MB

ext2

RW

*

33 mounts or 179 days

Remotely Mounted Filesystems

Host	Dir/Share	FS
cdv-visionman	IPC$	cifs
cdv-visionman	md0	cifs

But, I cannot see md0 or its dirs under Networks in Win XP.

Any ideas? Thanks.

[Joao Cardoso]

> find: /mnt/sda3/.systemfile: Input/output error

The "input/output error" indicates a serious problem in the sda disk.
I would recomend doing a smart short test to begin with (Disk
Utilities->Health->Start short test). After the test ends post the
result (Show Status).

What does the filesystems page says? Does md0 appears there?

What is the System Log contents?

Looks like raid "recovering" does not succeed.