Unfortunately no one can be told what fun_plug is - you have to see it for yourself.
You are not logged in.
Hi guys,
I've been getting the error "The hard drives have been installed incorrectly. Please power off and swap hard drive locations." in the DNS admin page for two days solid now. Swapping the drives doesn't fix anything. Both drives seem to flash when writing to the device, it's just that with that error I can't get into the settings pages via the web page.
I'm new to all this so please let me know what I shoudl do. I read today about the mdadm command to check the existing raid config and got this :
root@nas:~# mdadm --detail /dev/md0
/dev/md0:
Version : 00.90.01
Creation Time : Mon May 19 20:43:15 2008
Raid Level : raid1
Array Size : 486544512 (464.01 GiB 498.22 GB)
Device Size : 486544512 (464.01 GiB 498.22 GB)
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0
Persistence : Superblock is persistent
Update Time : Fri May 30 19:13:51 2008
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Events : 0.104280
Number Major Minor RaidDevice State
0 8 18 0 active sync /dev/sdb2
1 8 2 1 active sync /dev/sda2
root@nas:~#
Offline
I have been working for a while on doing manual raid configuration. If it is not done exactly right, you will get this and I have not found the fix directly for it. I am very close to a more complecated fix, so hang tight for a little bit.
To access the settings, after logging in and getting the message, type the URL in of one of the known pages in the web tool. It will take you in and you will probably see that the RAID 1 array is probably fine and healty.
For a little more info, what H/W and F/W is you unit? Are you using the full amount of the drive for the array? Have you formatted the drives via the web interface under F/W 1.04 or 1.05 (assuming you are using one of these)?
Offline
Hi. The firmware is 1.05 (the latest one from the US site as of 2 weeks ago - I cant go in and check it as i don't know any known page names in the web interface). The hardware is B1.
I formatted it with 1.03 that was delivered with the unit , then installed 1.05 over the top, no reformat.
Offline
That is the problem. Upgrades from 1.03 to 1.04 or 1.05 require a reformat (or more correctly a re-partition) to set-up the config partition. Without this partition, the unit will keep thinking the drives are in the wrong order because it cannot read the config data. The good news is that your array is healthy and working. Here is one of the URLs you can use to get past login: /web/status/device_status.asp I'll leave it to you to know how to correctly copy it into the browser.
I do have more good news for you. I have very sucsessfully been able to break and rebuild the arrays on the device. As I said in the last post, I have been finishing the testing. I will be posting the instructions for F/W 1.04/1.05 shortly.
For preliminary, are you using the entire drive for the raid 1 (no JBOD at end)? One I know the info, I will post step be step instructions to fix this. While I have never lost any data doing this, 1 mistake in typing a command can wipe it out, so I recommend a backup if you have it.
Offline
Yep, the whole drive is raid 1.
Thanks for the link. I'll see if I can google "cut and paste" :-)
Offline
Just keep the paste off your fingers, the keyboard doesn't like it...
Do you have backup? How much room on the array do you have used up with data?
Last edited by bq041 (2008-05-30 19:34:50)
Offline
I have a backup for all data except the new utils installed on the device (twonky, funplug, etc)
I have only used up 40gb of the 500 with data so far.
I think you're right, the raid array is fine. We just need to fool the device into thinking it has been formatted under 1.05. How's testing going?
Offline
I think I've used the same drive too many times. The data that is flashed into mtdblock0 remembers them even if they have been fdisked and formatted. Anyway, I have gotten it right a couple of times, but not repeatable. Sometimes it likes the right disk, sometimes the left. Anyway, I don't find out until after sync, which takes about 2 hours each time, so I don't get too many chances during the day.
I know how to set it ever time in mtdblock0, but I am intentionally finding a surefire way to do it without having to mount the flash (don't want anyone to brick the thing). Anyway, I write out how I do it each time, and then I repeat it once I have it right, but the repeat is where I have issues. Anyway, I hopefully wil have it going this weekend. F/W 1.03 was way easier to do this.
Offline
Any progress on the hack to get around the fw 1.05 requirement for the extra partition ?
I still get the same error and yes the unit is still working fine apart from that.
Seeya
Offline
I'm getting there. I found that the web script looks for a specific file in the /tmp directory to display that message. However, removing this file instantly prompts to re-format, though the raid is still up and running. Since it takes me 120 minutes each time a I build the array, it is slow going. I did just complete my automated script for breaking the raid 1. Later today, when I get home from work, I will start building the array again.
Also, I have been slow because the unit does not like to continually format and reuse the same drives over and over. It stores the drives hardware serial number in flash and some other settings. Using this, it makes several assumptions that may or may not be true about how they are set-up, which has led to some re-formatting issues.
Anyway, I will hopefully have what replacement file is needed, so a temporary fix may be to use fun_plug to erase some files in the /tmp directory and adding the ones so the admin tool sees it correctly. This will only be a temporary fix until I can get the conditions right so the start-up processes will complete corectly. I think I know where my last mistake was, but I get about 1-2 trys each day.
Offline
reebar
post your output from:
ls -all /tmp
Do this after a fresh reboot before logging into the admin tool and again after logging in and getting the error. Also, how big are your drives? Are they identical? How much space is used on them? Are they RAID1 only, or do you have JBOD at the end?
Good news, I now have got the array correct under F/W 1.04 and 1.05. I also have a sucessful way to break the array. I need to add security into the scripts, though, as errors are ignored and the script will coninue to run. This is potentially bad. It will be done soon. The next thing is to work on the issue for the 1.03 guys running 1.04 or 1.05. I will get an interem fix for you, but I will also get it so you can upgrade to the new partition style without losing your data.
Offline
Okay, I found out that we cannot temporarily patch yours, as the web admin tool does a direct comparison of certain files when you log in. Unfortunately, the location for these files are not on your computer. It can be repaired, but it will take a little bit of patience. I will need some info from you,though.
Do you have everything in RAID1, or do you have a JBOD, also? Are the 2 hard drives identical?
A synopsis of what will get done is this:
1) We break the array, this will give you 2 identical disks.
2) We partition and format (using telnet, not the web) 1 drive in the format of F/W 1.05.
3) We copy all your data over to the fresh drive.
4) We partition and format the second drive.
5) We rebuild the array (using a script I wrote, or if you want to learn the process, we can go the long route, also).
Offline