K,
another node would be awesome! we’ll setup remote access to it (SSH) so that when we finalize the tests on the virtual machine and decide which client to go with, we’ll implement the solution on Adam and your node and see how things go practically.
There might be some missing info in the pages, so if you’re gonna do the implementation and find something missing, let us know.
1. sudo apt-get install no-ip
2. sudo apt-get install hdparm
3. Creating Dynamic DNS at no-ip could use a bit more guidance for newbies.
4. More info about adding Samba users (where the password goes .. using sudo)
If this could be automated into a script it might be easier for people to setup.
3) write the info you need added here so we could slap it in
4) I can whip a PHP script in a web interface to do this — at a later time when the box ready for cloning
I am going to redo the install and write everything down this time .. keeping it as simple/straightforward as possible for someone without any previous linux experience to go through it.
When we were going through the implementation, we were at a stage were we justed wanted the bloody thing to work! (motherboard manual sucked & we could’ve fried everything, old case sucked, Samba’s documentation is as simple as space shuttle design, …etc. — you get the picture).
As I said before, we just want to get one node (Adam) fully completed, then we’ll be able to re-create another, and then I’ll start writing the web-based config stuff & in the end bundle it all as a distro.
MBh
I set it up using one harddrive; didn’t have a chance to evaluate the RAID part. The majority of it was fairly straightforward .. I am getting a *new* system this weekend, and then I will really be able to put it through the tests.
Bojacob:
Good luck with that .. my upload is ridiculously throttled (4mb/640k) to the extent that I shifted my torrenting overseas to maintain ratios.
This is great. I’ve already set up a web server on my home pc with virtual access to the media on my NAS sadly, my connection hasn’t been upgraded that much and also the issue with the upload comes up again. Anyway, I’m willing to help as much as I can here.
I’m still trying to figure out if it’s possible to use another antenna. Another may fit, but the question is: can the internal circuitry handle the extra power or provide it?
I’ll look for the specs & design of that router & have a friend (electrical engineer) take a look at it.
I doubt someone is going to sue us (Kuwait and all) and especially not over colors (Unless its T-Mobile Magenta). However, if you guys have any suggestions on the color scheme .. I am all ears.
The Open-Mesh system looks great .. you would probably need some beefier network equipment (with external antennas). The system is built on sharing one connection .. we would need to configure the routers to connect to the underlying Robin network (to share content).
Regarding the most recent post about the CLI hub daemon. Could http://opendchub.sourceforge.net/ be what you’re looking for? however it doesn’t look like the project is very active.
Thanks N, we’re currently testing verlihub. It’s one of the more recent/active projects and it happens to be included in the ubuntu repositories. More on that later. (I’m sick of waiting for this network to get online!)
I’ll work on it as well as soon as I manage to get my network up (just moved..) could you also make an guide (or link to one) and make it publicly available through here? That would help.
That’s a great deal, since online, one disk costs 54 KD, excluding the shipping costs, since each is about 1kg in weight, it would’ve almost reached 200 KD anyway.
Plus the risk of having the disk damaged due to bad handling, or just because of our bad luck, a dead one!
Since we were so happy about the deal, something had to screw us & laugh at our confused faces when we hooked the disks to the new SATA controllers we bought (2x SATA controllers each providing 4 SATA ports — 8KD each)
Turns out, the built in RAID controller on the SATA card controller was buggy and doesn’t seem to allow the system to boot (loads before the BIOS!!!) due to hooking 1TB disks to it.
Its internal BIOS program dates to 2004. So we looked up it online and found some updates, that will hopefully solve our problem. Otherwise, we’re screwed, and will have to look for different cards…
Tomorrow we’ll do the SATA card BIOS update and see how things go.
P.S.: Adding the 1TB disks has prompted us to change the existing RAID scheme to not lose more than 1 disk in the whole system. Will post details on this once we have things running.
Alright. We spent over 13 hours in total just trying to make these bloody cards work with the 1TB disks, and I kid you not, those were the most frustrating 13 hours ever!!!
We started around 17:30 on Thursday, by looking up online the required files to update the SATA controller’s BIOS. Thankfully, Silicon Image’s website was a complete mess, and their versioning convention, is unconventional.
Took around an hour to find the update utilities and the binary files of the required BIOS.
There were 2 types of update utilities: Windows tools that run under the virtual DOS environment (command line), and the pure DOS tools. The first was released in 2007 & the latter in 2008!!
Anyway, naturally, we chose to run the windows tools, so we read the document that came with the tool and fired the command line prompt and launched the update tool. Fantastic. It didn’t work. Couldn’t detect any controllers, it said, even though windows was able to detect the controller & install its legacy drivers. Piece of junk.
There was an option to manually point the utility to the PCI bus of the SATA controller, but their great document mentioned no way of how to do that or how to even find the PCI bus number of the bloody device.
So we move to plan B: Pure DOS. Which means, we need a floppy!
None of the computers in the dewaniya had a floppy drive, let alone a floppy disk! So I went to Khaldiya’s co-op to scavenge one, but there was no shop to sell them, so I went to the computer booth in the family center in Abdullah Al-Salem and got one for 100 fils (5 irani breads!)
As soon as my foot left the family center, 3baid called saying that he tried using an old Pentium 3 machine which had a floppy and win98 (Arabic — double damage), but it asked for the win98 CD (to create a bootable floppy)…..
After cursing & swearing at our luck, I headed back to the dewaniya and 3baid took off for some family quality time.
I reached the dewaniya and tried to see which files were missing from the crappy win98 box, and when it asked to insert the CD, it showed which file it was looking for, so I searched for it on the machine and found it, then pointed the floppy tool to it, and it took! Then it prompted for another file… Again, I searched for it, and found it, and it took it, then it prompted for yet another freakin file. I searched for it, and of course, it wasn’t found.
A brief session of cursing.
So, creating a win98 bootable floppy from this machine is useless. I searched online for bootable floppy images, and I did find them. Some exe files that run on Windows, and format the floppy to make it bootable with the required files.
I downloaded the files to my laptop, and then hop to the P3 box, to find that it has no network card. No worries! It has USB, and I have my external hard disk with me -> WIN!
I plugin the disk to the USB ports, it detects the new hardware, then asks to insert the drivers CD……
Brief cursing session.
So, now we have a machine that does have a floppy drive but there’re no means of copying the files to it, except using a CD; There were no blank CDs in the dewaniya, only DVDs (which don’t work on the P3’s drive), and it was around 21:30 or 22:00, so all shops closed.
I thought of removing the floppy drive from the old box, and I grabbed the screw driver and punched the screws to undo them, but the screw driver rotated in its location … turns out the screws were not your ordinary type and needed a different type of screw drivers, which of course we didn’t have.
Yet another brief cursing session.
One of the guys came over with his PC, which had a floppy drive!!! So I hijacked it for a bit and slapped the disk into it and took the exe file from my laptop (which doesn’t have a floppy drive) and put it on the machine; double-clicked the exe and it started formatting!!! FINALLY!
At 65% I noticed that the hard disk LED of the machine isn’t stopping; I move the mouse, but the cursor doesn’t move. The machine froze!
I rebooted and tried again, but this time the machine froze at 5%.
The PC’s owner started whining about the crashes giving his disks bad sectors (???!!?!?) and I asked him to take the floppy drive, but he refused.
Your usual brief cursing session.
3baid comes along, and finds me in a state of utter despair. I told him the story, and he said that there’s an old computer with a floppy drive, and he went to some dark corner and pulled it out!!!!
(If you’re thinking of why I didn’t call 3baid before, well, because his mobile was out of battery and left it home to recharge)
So, we found a floppy drive, and from yet another (third) ancient PC we found an IDE cable for the floppy. We hooked it to the XP machine in which the SATA card was installed and booted. 10 minutes later windows started and inserted the floppy disk to run the exe tool on it. Failure: Disk I/O errors.
Curses.
Going to the format menu of the floppy, there was an option to make it bootable, so we tried it and it went smooth without errors. Fantastic!
We restarted with the floppy inside, and a message showed in DOS: DISK I/O ERROR.
Curses.
We went back to teh format menu of the floppy, and formatted the drive to be blank. It worked without errors. Then I copied the exe file that creates the bootable image to the floppy. Copy successful. I put the drive into the old P3 machine and open the floppy then copy the exe to the desktop; Failure: can’t read from disk.
One would think the disk is damaged, but it’s not, because it reads & writes fine on the XP machine — after a couple of time of formatting it.
We run the exe file on the XP machine one more time and it formats the floppy successfully. Then I accessed the floppy and deleted some un-needed programs like: deltree, edit, help, xcopy, …etc. to free some space, then copied the pure DOS update utility and the BIOS binary image file.
We rebooted and the floppy booted into DOS! YES!! We run the update utility and the BIOS is flashed successfully. At last!!
We danced and sang in joy and took a screenshot of the output of the DOS update utility.
We restarted teh machine without the floppy, and it refused to boot. After a couple of tries, still no use. The SATA card has precedence to the BIOS, which means you won’t be able to even see the BIOS menu.
The box booted fine without the card. After some investigation, we found out that we used the BIOS image which was supposed to be used by the OEM manufacturers, not end-users, and that the end-user BIOS image was there, but I copied the wrong file…
Not so brief cursing session.
We had another card, but we didn’t want to use it and opted to revive this dead piece of junk. We booted into the floppy and tried to run the utility again, but this time it didn’t detect the controller.
One of the guys suggested a brilliant idea, to boot into windows, then hibernate and then insert the SATA controller and see if windows detects it.
Hibernate option wasn’t available as part of the shutdown, even though it was enabled in the power-management menu. We got around this, by changing the action of pressing the power button! Plugged the card, power on; No boot.
A hibernation turns off the computer and when resumed, it boots into the BIOS then when it reaches the OS boot, it detects that it should resume an old session. So, since the card sits before the BIOS, we couldn’t boot.
We tried setting the machine on stand by, and then plugged the card, but that didn’t work either.
We booted into the floppy, then plugged the card while the PC was on. No use. Tried the same with windows. No use.
The SATA controller had 2 pins that looked like jumpers, named JP4. I searched online to see if we can use them to reset the BIOS.
I downloaded schematic design of the board. I downloaded the documentation. I got PDF files of clock timings, hex codes for PCI protocols and many other details, but nothing about the bloody pins.
Later on, I found a PDF of a motherboard that has 3 pins named JP4 and says when the BIOS gets corrupted due to bad flashing, short the pins to reset. So I grabbed a jumper from an old CD and short the 2 pins on the controller. Machine didn’t boot. I then removed the jumper, since it may have been resetted, but machine didn’t boot either.
Curses.
I got fed up & moved away for some rest, and village upgrades on Travian, while 3baid went through the crappy site of Silicon Image, page by page, line by line.
Eventually, 3baid found a legacy BIOS for our card, and then concluded that even if we had used the end-user BIOS image, which came with the OEM image, we would’ve gotten the same results, because we were using the wrong BIOS version all together.
3baid downloads the latest legacy BIOS image, and put it on the floppy. We put the 2nd card (which is working but not accepting the 1TB disks), then we boot into DOS, and update the BIOS of the controller. Successful operation.
We remove the floppy disk and reboot, and the machine boots and the controller’s BIOS version gpt updated from 2004 to 2006.
We put the now updated card on Adam, and hook the 3x 1TB disks to it and boot, and it detects them perfectly and all goes well.
The time is now somewhere about Friday 03:00.
We created 2 RAID arrays out of the disks, in the process to create another logical system, to eventually migrate the data from the old array to the new one, then destroy the old array, and join the 500GB disks to the new array.
I just created a 2nd volume group with the new RAID arrays of the new disks and noticed some missing information in the “Server Software” section.
- Should mention to type: “sudo su” then continue to type the commands, as they all require root privilege.
- Retarded way of creating a volume group and requires info which isn’t mentioned of how to obtain it
- No mention of how to create a filesystem (XFS) (Important as it affects performance)
I’ll update those today or tomorrow.
On another note, I forgot to mention that yesterday after getting the arrays up, we toyed a bit with Verlihub and we liked it, so far.
It has a good base of commands, and allows to define our own and even allow users to execute certain commands. What’s also good, is that if the user failed to connect to the server due to lack of shares or any reason, the server would suggest connecting to a list of other servers!
Though this should be on the client side, in case the main hub went down. The server-side redirection can be used to do load balancing at extreme cases.
Next step would be to install a client on Adam and connect it to itself. This would be done on every nod, as every node is both a hub & a client, though the typical usage would be a client.
Yeah, thanks a lot for the image 3baid, to remind us of the painful memories of that dreadful day…
And speaking of floppies, I cloned that floppy to an ISO file, just in case the floppy died. I’ll see if it’s possible to use the same ISO and burn it on a CD-RW — this will make our lives (and other people’s?) easier.
Maybe if I get some free time, I’ll try to even make it boot from a USB memory stick. Maybe…
The floppy contains the basic programs needed to boot plus the programs needed to upgrade the Silicon Image 3114 SATA Controller BIOS.
I’ll upload it to Adam at a later time and host it there, in case someone out there needed it one day.
Here’s on overview of what we did to the RAID arrays:
3baid suggested that we make one array in total only, so that we don’t lose 2 disks for parity (one disk per array), totaling in 2TB, rather than 1TB.
This means that the data on our current 1.4TB needs to be migrated to the new 1.9TB array, then we dismantle the old array and join the disks to the new one.
There’s a slight risk with the scheme: The first array has 4 disks connected to the motherboard, and the 2nd array has 3 disks (currently) that are connected to a SATA-to-PCI controller. If the PCI card/bus failed for some reason, we lose all data, since the total number of lost disks is greater than 1.
I don’t think the loss will be immediate, as the data itself is retained safely, as long as it’s being read from only. It can be recovered by simply using another PCI bus (if the BUS was dead) or replacing the card.
However, in the unfortunate event of the card dying while copying, there will be data loss. Whether all or some, major or minor, that I don’t know.
Such an event is highly unlikely, but judging from our luck, I can only hope & pray…
In comment #40, I mentioned that I created a 2nd Volume Group; the reason behind this is that, even though it’s possible to migrate current physical extents from the old disks/array to the new one(s), the dewaniya is known for its unstable power due to abuse, so a safer approach was to create a new VG and then copy the data.
If you have a UPS that can hold for hours (depending on your data size), then it’s possible to migrate your data from old disks to new ones smoothly.
Proceeding to plan, after I copied some data from the old array to the new one (which now resides on a new Volume Group and a new Logical Volume), 3baid continued with the rest of data; he had some problems with the data copying as it wasn’t properly copied (mention details 3baid). He ran rsync (smart move ;)) to copy the differences in files only, rather than over-writing everything all over again.
After that, 3baid proceeded with dismantling the old array & VG (details?), then added the old 500GB disks to the new array. The disks were added as /dev/sde not /dev/sde1, which is not correct.
Luckily, but the time I woke up, the arrays have finished reshaping and the added disks were assigned as spares not integrated fully into the array (due to the mistake).
Our 1TB disks are partitioned into 2 partitions of size 500101 MB, using cfdisk. Each partition belongs to an array. This allows us to add the 500GB disks to the new array (one 500GB disk per array).
The disks were removed from the arrays and the partitions’ size was double checked to make sure it’s the same as the ones on the 1TB disks in the new array. All is correct. Proceeding to adding the partition on the 500Gb disk to the array gives a horrifying error:
root@Adam:/# mdadm –add /dev/md2 /dev/sdf1
mdadm: /dev/sdf1 not large enough to join array
Goolgizing gives 5 or so results, which date to March 2008, but none with a solution!!!! I ended up doing what any Linux addict would do: RTFM!
There was mention of meta-data, bitmap & super-block info being written to the disks to keep information about the disk & array, a max of 128kB. Nothing mentions of how or what might cause the problem above, so I proceed to reading the source code of mdadm, where the error messages are printed, to figure out what causes this.
There 2 checks which printed that error in case of failure and both were related to super-block persistence check.
I’m quite sure yet, but from what I did, it seems like the superblock data was written to the beginning of the disk, and was not over-written even after deleting the existing partition and creating a new one.
The problem was resolved after performing the following:
dd if=/dev/zero of=/dev/sde bs=256k count=1
Then I repartitioned the disk and added the partition to the array, and it worked! I was glad, since that error freaked me out and took me 2 hours to resolve it…
If you’re someone who are facing the same problem, maybe running this would solve your issue:
mdadm –zero-superblock –force /dev/sdx
The above might be better to use, since the superblock data is saved on either the beginning of the disk, or at its end, depending on the version of format of the metadata.
The arrays now are rebuilding, and hopefully we won’t have any power issues to hinder the process. Unfortunately, the rebuilding is being performed at a speed of 9MB/s, compared to 30MB/s when the new array was being rebuilt. The difference? at 30MB/s, the arrays rebuilt in 4 hours each. At 9MB/s, it will take each array 15 stinky hours!!!
You know, with my array at home, I never had such bad luck, so I’m suspecting 3baid….
I forgot to mention that the array’s reshaping/rebuilding speed has increased to 20MB/s, and the first array will finish in 59 minutes.
I still have no clue why it was limited to 9MB/s .. I hope it wasn’t because some disks are on a PCI bus & the rest are on the motherboard’s ports, because if so, I can imagine that we’ll have one crappy performance when reading & writing to the volume…
I tried the ISO file I made out of the floppy’s image file and it didn’t work. There’s a PC with Nero, which I’ll use to create a bootable CD with the required applications on it.
After that, I’ll upload it to Adam and make it publicly available.
Seems like “pvresize” took a long time and got stuck due to an error somewhere.
I rebooted the machine after over 24 hours passed and nothing happened. The 2nd array /dev/md3 started rebuilding for some reason, and this time, at 34MB/s !!! (compared to 9MB/s)
I waited for the rebuild to finish then proceeded and everything went smooth.
Now we have 3.7TB of total storage which is 38% used only.
It sure looks better than the previous grendaizar picture heh. So how can we access the current files.. if possible? Also how’s verlihub coming? Anything we can idle on atm? I got my buntu machine just sitting there..
I got verlihub running, but figured since you’re working on the dc daemon you’d probably only need one server and I can probably find a way to set it up to idle on the server with the shares of stuff on my NAS.
I got a hard time trying to convert the floppy image into a bootable ISO CD image.
I tried all possible ways using mkisofs, yet failed. Weird enough, the resulting images boot fine using qemu, but not on vmware nor a normal PC.
My last solution is to use freeDOS; I’ll modify the base ISO file’s bootloader to boot into safe-mode by default and run the BIOS flashing tool automatically.
The original size of the base freeDOS ISO is 7MB, since it contains packages to install DOS. I’ll delete these to reduce the total size to 1MB.
I hope the safe-mode boot image allows the execution of EXE files…
N, we haven’t finished experimenting with verlihub, but you should be able to connect on port 4111. If you can, great! If you can’t, wait until we get it running properly or create a hub of your own so that it runs as a fallback/backup.
Sniffing the local traffic when only the Direct Connect client was trying to establish a connection, resulted in connections over port 4111 (TCP) only.
So I guess the problems of not being able to connect from outside is caused by the dumb SpeedTouch Wireless router.
Okay, I’ve literally spent HOURS trying to get the port forwarding to work with the new router (Belkin N1) and nothing! First of all, the WLAN IP doesn’t seem to be correct to begin with, (one time it was a myspace.com server?!) the hostname ‘Adam’ is not recognized by the router for some odd reason and there’s no way to assign static IPs from the router that I’m aware of. There are no firmware updates btw.
I’m going to email Belkin about this and if they don’t offer a solution, I think the best thing to do is make the server become a DHCP server as well :/
See if the thing has a firmware update available. Usually such crap comes with the default firmware, but they fix it later on.
Let’s hope the router will be able to work as DHCP relay client so that those who connect wirelessly can obtain the IP from Adam (if we had to go that way).
By the way, it’s been about 4 days that the Internet traffic at home is very slow, when I have Azureus running (downloading at 10kB/s, uploading at 17 kB/s) — and my connection is 512kbps.
I hope QualityNet didn’t start throttling P2P traffic … that would be very bad for them. They wouldn’t be living up to their name…
Where’d the join in post go? I’m using ShakesPeer on the Mac it is working wonderfully. To connect though, have to enter the host as HOSTNAME:PORT for it to work.
I’m working to get a client running on my pc at home to share off the NAS.
When I got up this morning, I tried to browse your files, Verlihub kicked me off the server I think and sent me to a public hub lol. I couldn’t get back on, I had to close the client and reconnect. I was using (DC++) linuxdcpp. It worked fine after that.
Here’s a question.. Does DC++ count as P2P traffic?
So there wasn’t a problem with transferring files? Did you get the file list in one piece?
April 8, 2008 at 9:17 am
Picking up from where we left from the comments here: http://3baidsblog.wordpress.com/2008/04/05/dewaniya-net/#comments
K,
another node would be awesome! we’ll setup remote access to it (SSH) so that when we finalize the tests on the virtual machine and decide which client to go with, we’ll implement the solution on Adam and your node and see how things go practically.
There might be some missing info in the pages, so if you’re gonna do the implementation and find something missing, let us know.
April 8, 2008 at 2:56 pm
I am going to follow the guides on this page from an idiots perspective .. this site might benefit as a wiki-page.
April 8, 2008 at 4:18 pm
MBH, Openfiler updates have been delayed. Good thing we didn’t wait :P
April 8, 2008 at 8:36 pm
Just finished setting up dn2 .. initial points:
1. sudo apt-get install no-ip
2. sudo apt-get install hdparm
3. Creating Dynamic DNS at no-ip could use a bit more guidance for newbies.
4. More info about adding Samba users (where the password goes .. using sudo)
If this could be automated into a script it might be easier for people to setup.
April 8, 2008 at 9:36 pm
Thanks for the feedback :)
April 8, 2008 at 9:39 pm
3) write the info you need added here so we could slap it in
4) I can whip a PHP script in a web interface to do this — at a later time when the box ready for cloning
April 9, 2008 at 12:10 am
I am going to redo the install and write everything down this time .. keeping it as simple/straightforward as possible for someone without any previous linux experience to go through it.
April 9, 2008 at 7:50 pm
Cool — I appreciate your effort!
When we were going through the implementation, we were at a stage were we justed wanted the bloody thing to work! (motherboard manual sucked & we could’ve fried everything, old case sucked, Samba’s documentation is as simple as space shuttle design, …etc. — you get the picture).
As I said before, we just want to get one node (Adam) fully completed, then we’ll be able to re-create another, and then I’ll start writing the web-based config stuff & in the end bundle it all as a distro.
April 9, 2008 at 7:54 pm
Holy hell, how did I miss all this!
I admire the initiative, and I sure as heck will bug you guys with my unwanted help!
Issue with the upload speed is the friggin asymmentic nature of ADSL and our crappy MOC lines.
Anyway, I have my own personal NAS server running. I’ll see if I can bug Qnet into upping my upload from 512k.
Gotta push for co-location. Home lines aren’t going to cut it once this grows. I’ll see what I can do from my side.
April 10, 2008 at 11:03 am
Bojacob, yeah the upload thing is still being looked into. I think KEMS provides symmetrical DSL. More ISP profiling here :)
April 10, 2008 at 1:20 pm
MBh
I set it up using one harddrive; didn’t have a chance to evaluate the RAID part. The majority of it was fairly straightforward .. I am getting a *new* system this weekend, and then I will really be able to put it through the tests.
Bojacob:
Good luck with that .. my upload is ridiculously throttled (4mb/640k) to the extent that I shifted my torrenting overseas to maintain ratios.
April 10, 2008 at 2:39 pm
Could this be a possibility for sharing files:
http://sourceforge.net/projects/mutella/
April 10, 2008 at 3:05 pm
Whats any project without a logo ..
Here is an idea, its supposed to look like two ds (diwaniya) connected:
http://farm4.static.flickr.com/3013/2402455965_ed155851b2.jpg?v=0
April 10, 2008 at 3:08 pm
Mutella is based on the gnutella network, which connects to globlal peers. Btw, if you don’t mind, could you add your 4MB data to the ISP blog?
April 10, 2008 at 3:16 pm
K, I was hoping the logo would contain an N somewhere :/
“Dewaniya” is very generic, don’t you think?
April 10, 2008 at 3:42 pm
This is great. I’ve already set up a web server on my home pc with virtual access to the media on my NAS sadly, my connection hasn’t been upgraded that much and also the issue with the upload comes up again. Anyway, I’m willing to help as much as I can here.
April 11, 2008 at 2:32 am
Thanks N, we’ll let you know of any updates.
April 11, 2008 at 9:58 am
Now with N:
http://farm3.static.flickr.com/2007/2404379669_7f2f17e0e5_o.png
Just an idea at the moment .. Going to add some colors and play around with the lines.
April 11, 2008 at 12:23 pm
An idea just popped in my head for the logo .. estekanat chay Oo 6al3a menha wayrat network!
3aba6…
April 11, 2008 at 1:35 pm
“D2D: Dewaniya-to-Dewaniya protocol” XD
April 11, 2008 at 1:40 pm
Hmmm.. how about the “D” becomes the handle of an estikana?
Just an image I found for more ideas
April 11, 2008 at 2:21 pm
Actually the head of the d in my logo is from a 2: d2d.
Another look with colors (ubuntu colors)
http://farm4.static.flickr.com/3249/2405481404_fd25b67f76_o.png
April 11, 2008 at 5:36 pm
K, wouldn’t that be a violation of copyright?
I prefer to have our own theme, you know? And would love to make it something stupid/3bee6 :D (though we don’t have to)
April 12, 2008 at 2:14 am
I like the new logo better. MBH you can’t get enough 3abu6 can you? :P
April 12, 2008 at 11:41 am
3baid, Yes.. I’m afraid I can’t get enough :p
Guys, you MUST look at this: http://www.open-mesh.com/store/products.php?product=Accton-Mini%252dRouter
April 12, 2008 at 11:50 am
Oh and have a sample overview of their management & overview:
http://open-mesh.com/view.php
Put the SSID as “test” (without the double quotes). Wait for the page to load & see the amazing mash up show details!
There’s also a free extension to this existing one: http://code.google.com/p/meshnet/
I’m still trying to figure out if it’s possible to use another antenna. Another may fit, but the question is: can the internal circuitry handle the extra power or provide it?
I’ll look for the specs & design of that router & have a friend (electrical engineer) take a look at it.
April 12, 2008 at 1:13 pm
I don’t know MBH, I doubt we’ll get the enough range. You’d have to have this in every home from one area to another!
April 12, 2008 at 9:02 pm
3baid,
Dude, that uses 9-12 volts only .. just stick it in light posts :p~
And this is not meant for hooking long-distance nodes. This is for creating a mesh network of near-by nodes. Useful for neighbors.
April 13, 2008 at 2:28 pm
I doubt someone is going to sue us (Kuwait and all) and especially not over colors (Unless its T-Mobile Magenta). However, if you guys have any suggestions on the color scheme .. I am all ears.
The Open-Mesh system looks great .. you would probably need some beefier network equipment (with external antennas). The system is built on sharing one connection .. we would need to configure the routers to connect to the underlying Robin network (to share content).
April 13, 2008 at 2:35 pm
I finally picked up my new Dewaniya test system .. Its a dual P3 Compaq ML370 with 6 18.2gb scsi drives. :)
I hope this thing turns on ..
April 13, 2008 at 7:39 pm
Oh this baby will run!!!
Dual P3 ?!?! Dear God!
Can you take pictures of the processor & mobo? (silly but yeah .. )
April 26, 2008 at 8:37 pm
Regarding the most recent post about the CLI hub daemon. Could http://opendchub.sourceforge.net/ be what you’re looking for? however it doesn’t look like the project is very active.
I also found http://www.verlihub-project.org/doku.php which seems to be a little more active.
p.s. also found http://www.dc.ds.pg.gda.pl/ but also, doesn’t seem to be as active as verlihub.
April 26, 2008 at 11:49 pm
Thanks N, we’re currently testing verlihub. It’s one of the more recent/active projects and it happens to be included in the ubuntu repositories. More on that later. (I’m sick of waiting for this network to get online!)
April 27, 2008 at 8:20 am
I’ll work on it as well as soon as I manage to get my network up (just moved..) could you also make an guide (or link to one) and make it publicly available through here? That would help.
April 27, 2008 at 7:07 pm
Ofcourse :)
April 30, 2008 at 10:15 pm
w00t! We got 3x 1TB disks for 200 KD!!
That’s a great deal, since online, one disk costs 54 KD, excluding the shipping costs, since each is about 1kg in weight, it would’ve almost reached 200 KD anyway.
Plus the risk of having the disk damaged due to bad handling, or just because of our bad luck, a dead one!
Since we were so happy about the deal, something had to screw us & laugh at our confused faces when we hooked the disks to the new SATA controllers we bought (2x SATA controllers each providing 4 SATA ports — 8KD each)
Turns out, the built in RAID controller on the SATA card controller was buggy and doesn’t seem to allow the system to boot (loads before the BIOS!!!) due to hooking 1TB disks to it.
Its internal BIOS program dates to 2004. So we looked up it online and found some updates, that will hopefully solve our problem. Otherwise, we’re screwed, and will have to look for different cards…
Tomorrow we’ll do the SATA card BIOS update and see how things go.
P.S.: Adding the 1TB disks has prompted us to change the existing RAID scheme to not lose more than 1 disk in the whole system. Will post details on this once we have things running.
May 2, 2008 at 2:29 pm
Alright. We spent over 13 hours in total just trying to make these bloody cards work with the 1TB disks, and I kid you not, those were the most frustrating 13 hours ever!!!
We started around 17:30 on Thursday, by looking up online the required files to update the SATA controller’s BIOS. Thankfully, Silicon Image’s website was a complete mess, and their versioning convention, is unconventional.
Took around an hour to find the update utilities and the binary files of the required BIOS.
There were 2 types of update utilities: Windows tools that run under the virtual DOS environment (command line), and the pure DOS tools. The first was released in 2007 & the latter in 2008!!
Anyway, naturally, we chose to run the windows tools, so we read the document that came with the tool and fired the command line prompt and launched the update tool. Fantastic. It didn’t work. Couldn’t detect any controllers, it said, even though windows was able to detect the controller & install its legacy drivers. Piece of junk.
There was an option to manually point the utility to the PCI bus of the SATA controller, but their great document mentioned no way of how to do that or how to even find the PCI bus number of the bloody device.
So we move to plan B: Pure DOS. Which means, we need a floppy!
None of the computers in the dewaniya had a floppy drive, let alone a floppy disk! So I went to Khaldiya’s co-op to scavenge one, but there was no shop to sell them, so I went to the computer booth in the family center in Abdullah Al-Salem and got one for 100 fils (5 irani breads!)
As soon as my foot left the family center, 3baid called saying that he tried using an old Pentium 3 machine which had a floppy and win98 (Arabic — double damage), but it asked for the win98 CD (to create a bootable floppy)…..
After cursing & swearing at our luck, I headed back to the dewaniya and 3baid took off for some family quality time.
I reached the dewaniya and tried to see which files were missing from the crappy win98 box, and when it asked to insert the CD, it showed which file it was looking for, so I searched for it on the machine and found it, then pointed the floppy tool to it, and it took! Then it prompted for another file… Again, I searched for it, and found it, and it took it, then it prompted for yet another freakin file. I searched for it, and of course, it wasn’t found.
A brief session of cursing.
So, creating a win98 bootable floppy from this machine is useless. I searched online for bootable floppy images, and I did find them. Some exe files that run on Windows, and format the floppy to make it bootable with the required files.
I downloaded the files to my laptop, and then hop to the P3 box, to find that it has no network card. No worries! It has USB, and I have my external hard disk with me -> WIN!
I plugin the disk to the USB ports, it detects the new hardware, then asks to insert the drivers CD……
Brief cursing session.
So, now we have a machine that does have a floppy drive but there’re no means of copying the files to it, except using a CD; There were no blank CDs in the dewaniya, only DVDs (which don’t work on the P3’s drive), and it was around 21:30 or 22:00, so all shops closed.
I thought of removing the floppy drive from the old box, and I grabbed the screw driver and punched the screws to undo them, but the screw driver rotated in its location … turns out the screws were not your ordinary type and needed a different type of screw drivers, which of course we didn’t have.
Yet another brief cursing session.
One of the guys came over with his PC, which had a floppy drive!!! So I hijacked it for a bit and slapped the disk into it and took the exe file from my laptop (which doesn’t have a floppy drive) and put it on the machine; double-clicked the exe and it started formatting!!! FINALLY!
At 65% I noticed that the hard disk LED of the machine isn’t stopping; I move the mouse, but the cursor doesn’t move. The machine froze!
I rebooted and tried again, but this time the machine froze at 5%.
The PC’s owner started whining about the crashes giving his disks bad sectors (???!!?!?) and I asked him to take the floppy drive, but he refused.
Your usual brief cursing session.
3baid comes along, and finds me in a state of utter despair. I told him the story, and he said that there’s an old computer with a floppy drive, and he went to some dark corner and pulled it out!!!!
(If you’re thinking of why I didn’t call 3baid before, well, because his mobile was out of battery and left it home to recharge)
So, we found a floppy drive, and from yet another (third) ancient PC we found an IDE cable for the floppy. We hooked it to the XP machine in which the SATA card was installed and booted. 10 minutes later windows started and inserted the floppy disk to run the exe tool on it. Failure: Disk I/O errors.
Curses.
Going to the format menu of the floppy, there was an option to make it bootable, so we tried it and it went smooth without errors. Fantastic!
We restarted with the floppy inside, and a message showed in DOS: DISK I/O ERROR.
Curses.
We went back to teh format menu of the floppy, and formatted the drive to be blank. It worked without errors. Then I copied the exe file that creates the bootable image to the floppy. Copy successful. I put the drive into the old P3 machine and open the floppy then copy the exe to the desktop; Failure: can’t read from disk.
One would think the disk is damaged, but it’s not, because it reads & writes fine on the XP machine — after a couple of time of formatting it.
We run the exe file on the XP machine one more time and it formats the floppy successfully. Then I accessed the floppy and deleted some un-needed programs like: deltree, edit, help, xcopy, …etc. to free some space, then copied the pure DOS update utility and the BIOS binary image file.
We rebooted and the floppy booted into DOS! YES!! We run the update utility and the BIOS is flashed successfully. At last!!
We danced and sang in joy and took a screenshot of the output of the DOS update utility.
We restarted teh machine without the floppy, and it refused to boot. After a couple of tries, still no use. The SATA card has precedence to the BIOS, which means you won’t be able to even see the BIOS menu.
The box booted fine without the card. After some investigation, we found out that we used the BIOS image which was supposed to be used by the OEM manufacturers, not end-users, and that the end-user BIOS image was there, but I copied the wrong file…
Not so brief cursing session.
We had another card, but we didn’t want to use it and opted to revive this dead piece of junk. We booted into the floppy and tried to run the utility again, but this time it didn’t detect the controller.
One of the guys suggested a brilliant idea, to boot into windows, then hibernate and then insert the SATA controller and see if windows detects it.
Hibernate option wasn’t available as part of the shutdown, even though it was enabled in the power-management menu. We got around this, by changing the action of pressing the power button! Plugged the card, power on; No boot.
A hibernation turns off the computer and when resumed, it boots into the BIOS then when it reaches the OS boot, it detects that it should resume an old session. So, since the card sits before the BIOS, we couldn’t boot.
We tried setting the machine on stand by, and then plugged the card, but that didn’t work either.
We booted into the floppy, then plugged the card while the PC was on. No use. Tried the same with windows. No use.
The SATA controller had 2 pins that looked like jumpers, named JP4. I searched online to see if we can use them to reset the BIOS.
I downloaded schematic design of the board. I downloaded the documentation. I got PDF files of clock timings, hex codes for PCI protocols and many other details, but nothing about the bloody pins.
Later on, I found a PDF of a motherboard that has 3 pins named JP4 and says when the BIOS gets corrupted due to bad flashing, short the pins to reset. So I grabbed a jumper from an old CD and short the 2 pins on the controller. Machine didn’t boot. I then removed the jumper, since it may have been resetted, but machine didn’t boot either.
Curses.
I got fed up & moved away for some rest, and village upgrades on Travian, while 3baid went through the crappy site of Silicon Image, page by page, line by line.
Eventually, 3baid found a legacy BIOS for our card, and then concluded that even if we had used the end-user BIOS image, which came with the OEM image, we would’ve gotten the same results, because we were using the wrong BIOS version all together.
3baid downloads the latest legacy BIOS image, and put it on the floppy. We put the 2nd card (which is working but not accepting the 1TB disks), then we boot into DOS, and update the BIOS of the controller. Successful operation.
We remove the floppy disk and reboot, and the machine boots and the controller’s BIOS version gpt updated from 2004 to 2006.
We put the now updated card on Adam, and hook the 3x 1TB disks to it and boot, and it detects them perfectly and all goes well.
The time is now somewhere about Friday 03:00.
We created 2 RAID arrays out of the disks, in the process to create another logical system, to eventually migrate the data from the old array to the new one, then destroy the old array, and join the 500GB disks to the new array.
May 2, 2008 at 7:46 pm
I just created a 2nd volume group with the new RAID arrays of the new disks and noticed some missing information in the “Server Software” section.
- Should mention to type: “sudo su” then continue to type the commands, as they all require root privilege.
- Retarded way of creating a volume group and requires info which isn’t mentioned of how to obtain it
- No mention of how to create a filesystem (XFS) (Important as it affects performance)
I’ll update those today or tomorrow.
On another note, I forgot to mention that yesterday after getting the arrays up, we toyed a bit with Verlihub and we liked it, so far.
It has a good base of commands, and allows to define our own and even allow users to execute certain commands. What’s also good, is that if the user failed to connect to the server due to lack of shares or any reason, the server would suggest connecting to a list of other servers!
Though this should be on the client side, in case the main hub went down. The server-side redirection can be used to do load balancing at extreme cases.
Next step would be to install a client on Adam and connect it to itself. This would be done on every nod, as every node is both a hub & a client, though the typical usage would be a client.
May 3, 2008 at 5:40 pm
Floppy.. heh :/
http://dewaniyanet.wordpress.com/2008/05/03/flashing-the-silicon-image-bios/
May 4, 2008 at 4:32 pm
Yeah, thanks a lot for the image 3baid, to remind us of the painful memories of that dreadful day…
And speaking of floppies, I cloned that floppy to an ISO file, just in case the floppy died. I’ll see if it’s possible to use the same ISO and burn it on a CD-RW — this will make our lives (and other people’s?) easier.
Maybe if I get some free time, I’ll try to even make it boot from a USB memory stick. Maybe…
The floppy contains the basic programs needed to boot plus the programs needed to upgrade the Silicon Image 3114 SATA Controller BIOS.
I’ll upload it to Adam at a later time and host it there, in case someone out there needed it one day.
May 4, 2008 at 5:36 pm
Here’s on overview of what we did to the RAID arrays:
3baid suggested that we make one array in total only, so that we don’t lose 2 disks for parity (one disk per array), totaling in 2TB, rather than 1TB.
This means that the data on our current 1.4TB needs to be migrated to the new 1.9TB array, then we dismantle the old array and join the disks to the new one.
There’s a slight risk with the scheme: The first array has 4 disks connected to the motherboard, and the 2nd array has 3 disks (currently) that are connected to a SATA-to-PCI controller. If the PCI card/bus failed for some reason, we lose all data, since the total number of lost disks is greater than 1.
I don’t think the loss will be immediate, as the data itself is retained safely, as long as it’s being read from only. It can be recovered by simply using another PCI bus (if the BUS was dead) or replacing the card.
However, in the unfortunate event of the card dying while copying, there will be data loss. Whether all or some, major or minor, that I don’t know.
Such an event is highly unlikely, but judging from our luck, I can only hope & pray…
In comment #40, I mentioned that I created a 2nd Volume Group; the reason behind this is that, even though it’s possible to migrate current physical extents from the old disks/array to the new one(s), the dewaniya is known for its unstable power due to abuse, so a safer approach was to create a new VG and then copy the data.
If you have a UPS that can hold for hours (depending on your data size), then it’s possible to migrate your data from old disks to new ones smoothly.
Proceeding to plan, after I copied some data from the old array to the new one (which now resides on a new Volume Group and a new Logical Volume), 3baid continued with the rest of data; he had some problems with the data copying as it wasn’t properly copied (mention details 3baid). He ran rsync (smart move ;)) to copy the differences in files only, rather than over-writing everything all over again.
After that, 3baid proceeded with dismantling the old array & VG (details?), then added the old 500GB disks to the new array. The disks were added as /dev/sde not /dev/sde1, which is not correct.
Luckily, but the time I woke up, the arrays have finished reshaping and the added disks were assigned as spares not integrated fully into the array (due to the mistake).
Our 1TB disks are partitioned into 2 partitions of size 500101 MB, using cfdisk. Each partition belongs to an array. This allows us to add the 500GB disks to the new array (one 500GB disk per array).
The disks were removed from the arrays and the partitions’ size was double checked to make sure it’s the same as the ones on the 1TB disks in the new array. All is correct. Proceeding to adding the partition on the 500Gb disk to the array gives a horrifying error:
root@Adam:/# mdadm –add /dev/md2 /dev/sdf1
mdadm: /dev/sdf1 not large enough to join array
Goolgizing gives 5 or so results, which date to March 2008, but none with a solution!!!! I ended up doing what any Linux addict would do: RTFM!
There was mention of meta-data, bitmap & super-block info being written to the disks to keep information about the disk & array, a max of 128kB. Nothing mentions of how or what might cause the problem above, so I proceed to reading the source code of mdadm, where the error messages are printed, to figure out what causes this.
There 2 checks which printed that error in case of failure and both were related to super-block persistence check.
I’m quite sure yet, but from what I did, it seems like the superblock data was written to the beginning of the disk, and was not over-written even after deleting the existing partition and creating a new one.
The problem was resolved after performing the following:
dd if=/dev/zero of=/dev/sde bs=256k count=1
Then I repartitioned the disk and added the partition to the array, and it worked! I was glad, since that error freaked me out and took me 2 hours to resolve it…
If you’re someone who are facing the same problem, maybe running this would solve your issue:
mdadm –zero-superblock –force /dev/sdx
The above might be better to use, since the superblock data is saved on either the beginning of the disk, or at its end, depending on the version of format of the metadata.
The arrays now are rebuilding, and hopefully we won’t have any power issues to hinder the process. Unfortunately, the rebuilding is being performed at a speed of 9MB/s, compared to 30MB/s when the new array was being rebuilt. The difference? at 30MB/s, the arrays rebuilt in 4 hours each. At 9MB/s, it will take each array 15 stinky hours!!!
You know, with my array at home, I never had such bad luck, so I’m suspecting 3baid….
May 4, 2008 at 9:23 pm
“Such an event is highly unlikely, but judging from our luck, I can only hope & pray…”
@_@ Hey!
May 4, 2008 at 10:31 pm
Oh yeah, sorry.
Correction: Judging by 3baid’s luck, I can only hope & pray…
May 4, 2008 at 10:33 pm
I forgot to mention that the array’s reshaping/rebuilding speed has increased to 20MB/s, and the first array will finish in 59 minutes.
I still have no clue why it was limited to 9MB/s .. I hope it wasn’t because some disks are on a PCI bus & the rest are on the motherboard’s ports, because if so, I can imagine that we’ll have one crappy performance when reading & writing to the volume…
We’ll see tomorrow.
May 5, 2008 at 12:45 am
*more praying needed*
May 6, 2008 at 1:53 pm
I tried the ISO file I made out of the floppy’s image file and it didn’t work. There’s a PC with Nero, which I’ll use to create a bootable CD with the required applications on it.
After that, I’ll upload it to Adam and make it publicly available.
May 6, 2008 at 3:50 pm
Seems like “pvresize” took a long time and got stuck due to an error somewhere.
I rebooted the machine after over 24 hours passed and nothing happened. The 2nd array /dev/md3 started rebuilding for some reason, and this time, at 34MB/s !!! (compared to 9MB/s)
I waited for the rebuild to finish then proceeded and everything went smooth.
Now we have 3.7TB of total storage which is 38% used only.
ABUSE!
May 6, 2008 at 3:52 pm
I have kept a log of all the commands and outputs during the processes above.
I think I’ll dump them into a text file and make it available for those who wish to see the details.
For the expansion phase, I simply followed the exact steps mentioned in the section: “Expanding Storage”
May 6, 2008 at 8:46 pm
zip.php page has been updated to prevent the script from updating the zip file with in 10 minutes.
If the list of files is updated in less than 10 minutes of the previous zip file creation time, you’ll be served the old zip file.
Not that anyone is visiting the site now, but it’s just a precautionary step.
I’m thinking of removing the whole thing, and just run a cron job every 30 minutes to keep the file updated…
May 7, 2008 at 7:54 am
Should I be posting all the above in the blog, or here?
Hmmmm…
And where the hell did this weird image of mine come from!!!!???!!11??one!
May 7, 2008 at 3:14 pm
3baid,
What’s with the lame picture? -> http://dn0.no-ip.org/images/splash.jpg
/me receives 250x damage
May 7, 2008 at 10:09 pm
It sure looks better than the previous grendaizar picture heh. So how can we access the current files.. if possible? Also how’s verlihub coming? Anything we can idle on atm? I got my buntu machine just sitting there..
I got verlihub running, but figured since you’re working on the dc daemon you’d probably only need one server and I can probably find a way to set it up to idle on the server with the shares of stuff on my NAS.
May 7, 2008 at 11:26 pm
Hahaha XD
May 8, 2008 at 1:10 pm
I got a hard time trying to convert the floppy image into a bootable ISO CD image.
I tried all possible ways using mkisofs, yet failed. Weird enough, the resulting images boot fine using qemu, but not on vmware nor a normal PC.
My last solution is to use freeDOS; I’ll modify the base ISO file’s bootloader to boot into safe-mode by default and run the BIOS flashing tool automatically.
The original size of the base freeDOS ISO is 7MB, since it contains packages to install DOS. I’ll delete these to reduce the total size to 1MB.
I hope the safe-mode boot image allows the execution of EXE files…
May 11, 2008 at 4:35 pm
I hate the color green.
May 11, 2008 at 7:38 pm
How come all your images are green? :P
May 13, 2008 at 2:43 pm
http://lifehacker.com/389503/alliance-creates-private-p2p-file+sharing-networks
That said .. I don’t want to go through the above ^ :)
May 13, 2008 at 3:49 pm
K, we saw that but it’s got no CLI app as far as we know. :/
May 13, 2008 at 6:36 pm
N, we haven’t finished experimenting with verlihub, but you should be able to connect on port 4111. If you can, great! If you can’t, wait until we get it running properly or create a hub of your own so that it runs as a fallback/backup.
May 16, 2008 at 12:02 am
That reminds me .. after the protocol has been decided, we should add a section on setting up the router to forward packets.
May 16, 2008 at 12:13 am
K,
Putting details for each router brand is just too much :/
We can say things like set up a port forward rule, for both TCP & UDP. Then people should refer to their manuals on how to achieve that.
May 16, 2008 at 4:36 am
MBH, I confirmed that the hub can’t be accessed remotely and I think it’s because of the router :/
I can access regular hubs, but Adam gives me timeouts.
May 17, 2008 at 9:44 am
It could be requiring other ports than the one we setup.
Next time I go to the dewaniya, I’ll sniff the traffic when connecting to the hub and see which ports are being used.
May 25, 2008 at 7:46 pm
Sniffing the local traffic when only the Direct Connect client was trying to establish a connection, resulted in connections over port 4111 (TCP) only.
So I guess the problems of not being able to connect from outside is caused by the dumb SpeedTouch Wireless router.
June 13, 2008 at 3:26 am
Okay, I’ve literally spent HOURS trying to get the port forwarding to work with the new router (Belkin N1) and nothing! First of all, the WLAN IP doesn’t seem to be correct to begin with, (one time it was a myspace.com server?!) the hostname ‘Adam’ is not recognized by the router for some odd reason and there’s no way to assign static IPs from the router that I’m aware of. There are no firmware updates btw.
I’m going to email Belkin about this and if they don’t offer a solution, I think the best thing to do is make the server become a DHCP server as well :/
June 13, 2008 at 11:02 pm
See if the thing has a firmware update available. Usually such crap comes with the default firmware, but they fix it later on.
Let’s hope the router will be able to work as DHCP relay client so that those who connect wirelessly can obtain the IP from Adam (if we had to go that way).
By the way, it’s been about 4 days that the Internet traffic at home is very slow, when I have Azureus running (downloading at 10kB/s, uploading at 17 kB/s) — and my connection is 512kbps.
I hope QualityNet didn’t start throttling P2P traffic … that would be very bad for them. They wouldn’t be living up to their name…
June 13, 2008 at 11:56 pm
There’s no firmware update. The connection’s been fine on our side.
June 17, 2008 at 5:07 pm
IT WORKS! The Firmware updates WORKS! :D
The tech support sent us a beta firmware via email :]
We’re beta testers now :P
June 23, 2008 at 10:31 am
Where’d the join in post go? I’m using ShakesPeer on the Mac it is working wonderfully. To connect though, have to enter the host as HOSTNAME:PORT for it to work.
I’m working to get a client running on my pc at home to share off the NAS.
Awesome work guys.
June 23, 2008 at 12:07 pm
The DC Hub is on the main page.
June 24, 2008 at 6:56 am
When I got up this morning, I tried to browse your files, Verlihub kicked me off the server I think and sent me to a public hub lol. I couldn’t get back on, I had to close the client and reconnect. I was using (DC++) linuxdcpp. It worked fine after that.
Here’s a question.. Does DC++ count as P2P traffic?
So there wasn’t a problem with transferring files? Did you get the file list in one piece?
June 24, 2008 at 3:23 pm
Yes, everyone’s reporting the redirection and the same thing happens to me when I need to close the client and open it again.
Yes, DC++ is considered as P2P traffic unfortunately :/
Yes, I was able to get the entire file from you.