now that the glitches of my unRaid appearing/disappearing from the network are solved, i need to attend to the task of dealing with the fact that my array is near capacity.

 

i am on unRaid 5.0-rc5, i have 15 slots, 14 of which are used, all 2TB disks, with the exception of one 1TB drive, for a total of 25TB, with a total of only 2.96TB available...since i need to back up about 4TB of data, i just purchased two 3TB WD Red drives, and intend to do the following:

 

- replace my parity drive with one of the new 3TB Red drives

 

- put the former 2TB parity drive in place of the sole 1TB drive (making all data drives 2TB, and gaining me 1TB in total array size)

 

- replace one of my 2TB drives that are 100% full with the other new 3TB Red drive (gaining me another 1TB of total array size)

 

i started doing what i always used to do, stopped the array, did a clean power-down, took out the first disk, and exchanged it with one of the new 3TB drives...however, when i started the unRaid up again, the web UI showed parity having no disk, and Disk 1 with a red dot...but the drop down menus, which i remember as giving you the option of selecting a drive by serial number, showed no content.

 

my question now is this:

 

since i am not entirely sure now whether my parity drive was in the first, or the last slot, how can i see which drive (by serial number) is in which slot?

 

for the time being i put everything back the way it was and started the server again in its current, nearly full condition...what do i have to do now to see which serial number drives are where? how can i be certain in which physical slot my parity drive is?

Far as I know there is no easy way to determine where in your machine a particular drive is.  I think most of us with large drive systems keep track on a sheet (or myself, an Excel file) where each drive is.

weird, maybe that was years ago, but i faintly remember a screen in the web-UI that showed all of the drives with their serial numbers in their respective slots...but now that i think about it, it's unlikely that ever existed, being that there is no way to know which mono or SATA connector goes to which physical slot, so with that said, i must have remembered incorrectly.

 

but in the meantime by trial and error, i was able to determine that my parity drive was in the next to last slot, and not in the first slot, as i initially thought...my memory must be getting worse.

 

what i have now is a situation where i physically moved the parity drive from the next to last slot (slot #14) to the last slot (#15), because i want to either add an additional data drive in slot #14, or a cache drive...i haven't started the array yet, because when i refresh the list of drives in the UI, parity shows as green, and the Disk 1 now shows as a red dot...is this normal behavior? can i start the array without the risk of data loss if i moved the parity disk to another slot? will it have to rebuild parity? if so, should i swap the 2TB parity drive for the 3TB one *now*, so that i don't need to rebuild parity twice?

 

what's a safe next step for me to upgrade my parity drive to the 3TB size?

Lets make sure we are talking about the same things here.  SLOTS in UnRaid are real and show on the menu in the gui (and should show serial numbers as well).  Slot 1 (or the top most drive in the list on the GUI) Has to be the Parity Drive.  It has no relation to physical location (unraid doesn't care where the drive is physically located or connected, meaning you can move it around inside your server, just make sure its assigned to the parity drive in the GUI).

 

To upgrade to a 3TB parity drive, I would replace my existing parity drive (remove it from the server, insert the 3TB, and start unraid).  Then have it build parity again.  Once that was done, i would shutdown, put my OLD 2TB parity drive back in the server, start unraid and assign the 'old' drive to a data slot and start the array).

 

Far as I know there is no easy way to determine where in your machine a particular drive is.  I think most of us with large drive systems keep track on a sheet (or myself, an Excel file) where each drive is.

type:

strings /boot/config/super.dat

The drives will be listed in order, with the parity drive listed first.

 

Joe L.

points well taken, but before i try any of that, i have to get out of the current state i'm in...i must have screwed something up here...i now have a web-ui Main screen that shows my 14 hard-drives and internal USB flash, starting with parity at the top, and the flash at the very bottom...all have a green dot in front, with the exception of Drive 1, which all of a sudden shows a red dot...i think i confused unRaid by having done what i did in my carelessness...the last couple of physical slot swaps i did while the server was left running, and then doing a refresh of the Main UI screen.

 

anyway, i don't see why drive 1 should all of a sudden be faulty, but the array won't start up right now...if i hit "Spin Up", nothing happens, and i also don't see any of the shares in my finder anymore...the array behaves as if it's not up and running, but it gives me the usual "Refresh" and "Stop" buttons, as if it is running.

 

is there a safe way to "rebuild" the array, assuming that everything is as it was before i started to switch disks around? everything is exactly as it was when it was last working right now, so shouldn't there be a way to start the array like this?

 

 

this is driving me bonkers!

 

does the red dot mean that unRaid disabled the disk? can i override this somehow? i am quite certain that nothing went wrong with the disk between an hour ago and now...it didn't show any errors before, and i think that my switching around of the first and parity disks triggered unRaid to just disable the disk to protect the array? but since parity is fine, shouldn't i be able to override this disabling and start the array again?

 

just attached the current syslog, for all it's worth.

 

i just noticed something else: when i go into the "Stats" page, i get an error message above Disk 1, that says "Warning: Division by zero in /usr/local/emhttp/plugins/simpleFeatures/DiskStats.php on line 117"...how did that get in there, and how do i get rid of it?

 

boy-oh-boy, did i get myself into a situation here!

i'm stuck

syslog-2012-07-24_01.txt

does my go script look suspicious in any way? looks like simpleFeatures is being installed on launch, no? here it is:

 

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

 

#!/bin/bash

# Start the Management Utility

/usr/local/sbin/emhttp &

 

#! /boot/boot/cache_dirs -w

#! -e "Downloads*" -e "_BAK*" -i "Media*" -i "Movies*" -i "Music*" -i "Production*" -i "Software*"

 

installpkg /boot/packages/simpleFeatures-0.4-unraid-speeding_ant.tgz

#! installpkg /boot/packages_global/powerdown-1.02-noarch-unRAID.tgz

#! installpkg /boot/packages_global/jre-6u20-i586-1.txz

 

cd /boot/unmenu/uu

 

#! sleep 30

 

#! /boot/scripts/subsonic_start.sh

#! /boot/scripts/shutdown_jobs_add.sh

 

#! cd /boot/packages && find . -name '*.auto_install' -type f -print | sort | xargs -n1 sh -c

 

unraid_notify start

 

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

 

if somebody could tell me what to try next, i'd really appreciate it...right now i'm dead in the water again and really can't afford to lose another night without it at least rebuilding parity on the 3TB drive while i sleep...i lost the last week due to the pesky network cable problem, and now this.

 

Ask in the User customizations forum.

this is driving me bonkers!

 

does the red dot mean that unRaid disabled the disk?

Yes.

can i override this somehow?

No, a write to it failed, therefore, it is disabled.  It must be re-constructed.

i am quite certain that nothing went wrong with the disk between an hour ago and now...it didn't show any errors before, and i think that my switching around of the first and parity disks triggered unRaid to just disable the disk to protect the array? but since parity is fine, shouldn't i be able to override this disabling and start the array again?

Possibly, but remember, a write to it failed... It is guaranteed to be incorrect.  look for the word "trust" in the wiki, but read all the warnings about the risks.

thanks for that pointer, Joe...i should add that Disk 1 holds old DVD backups, so if one (or a few) or them end up not being retrievable, it wouldn't be the end of the world for me...i keep pretty tight control over what files are on what disk, and since i started unRaid initially as a media server, the entertainment stuff is occupying Disks 1-6 for the most part.

 

i'll read through that Wiki info you told me about and then decide whether to go for it, or do a quick warranty exchange on that drive...WD has been great about swapping my Green drives out for new ones when errors would happen...but it is unfortunate that these Green drives had to be swapped out as much as they did.

 

i am beginning to replace them with Red drives now, hoping that their claims are true...we'll see.

 

Joe, i just did a search on "Trust" in the wiki, and the only articles i could find refer to a procedure to override unRaid's actions on a *parity* drive...am i missing something, or are the procedures the same for a data drive?

Why don't you post a smart test first? 

thanks mbryanr...since i'm pretty inept at using the terminal, i first went into the web GUI to try to see what methods are available to get some Smart Test results...what i found gave me the following results (admittedly, they don't look encouraging...and btw. all the drives were spun up before i did this):

 

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

 

SMART Status:

Failing

Get data off disk and run advanced tests below for more information

SMART Extended Test: Disk must be spun up before running tests

SMART Short Test: Disk must be spun up before running tests

SMART Information:

/dev/: Unable to detect device type

Smartctl: please specify device type with the -d option.

 

Use smartctl -h to get a usage summary

 

SMART Log:

/dev/: Unable to detect device type

Smartctl: please specify device type with the -d option.

 

Use smartctl -h to get a usage summary

 

Disk Information:

/dev/: Unable to detect device type

Smartctl: please specify device type with the -d option.

 

Use smartctl -h to get a usage summary

 

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

 

i'll probably have to go into the terminal to get some of the missing info, right? i looked into the wiki's "Obtaining a Smart Report" section, but already i'm unclear on how to do some of what i'm instructed to do...for example, it says "Look at the Devices page for the device identifier (within the parentheses) for each disk, and substitute that for 'sda' on the command line."...but where is the Devices page?

 

First the "devices page" is the first page of the unRAID web interface

 

Assuming you have kept your disk assignments from the latest syslog posted,

Jul 24 18:07:54 unRAID emhttp: WDC_WD20EADS-00R6B0_WD-WCAVY0035734 (sdb) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EADS-00R6B0_WD-WCAVY0416376 (sdc) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EARS-00MVWB0_WD-WCAZA7226740 (sdd) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EARS-00MVWB0_WD-WCAZAC986143 (sde) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EADS-00R6B0_WD-WCAVY0179096 (sdf) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EADS-00R6B0_WD-WCAVY0216140 (sdg) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EADS-32R6B0_WD-WCAVY0716487 (sdh) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EADS-00R6B0_WD-WCAVY0122453 (sdi) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EARS-00MVWB0_WD-WMAZA0700430 (sdj) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD10EACS-00ZJB0_WD-WCASJ1019333 (sdk) 976762584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EARS-00MVWB0_WD-WCAZA5919059 (sdl) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EADS-00R6B0_WD-WCAVY0248388 (sdm) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EARS-00MVWB0_WD-WMAZA4714338 (sdn) 1953514584 Jul 24 18:07:54 unRAID emhttp: WDC_WD20EARS-00J99B0_WD-WCAWZ0729087 (sdo) 1953514584

 

IIRC and from your syslog... you are using a mac ?

If so, follow this...<I only have a Windows Laptop so I don't know if this is accurate for your OS:>

http://www.wikihow.com/Use-Telnet-on-Mac-OS-X

 

This should get you to connect via terminal/telnet.

 

Once you've connected to your server...

enter the commands below <modifying sdX - where "X" is the correct disk identifier that you wish to run the smart test on:

 

http://lime-technology.com/wiki/index.php/Troubleshooting#What_do_I_do_if_I_get_a_red_ball_next_to_a_hard_disk.3F

 

This short test takes 1 to 3 minutes (remember to substitute your drive's identifier for 'sda' as described above)

smartctl -d ata -tshort /dev/sda

i assume the first entry in that list is the parity drive, right?

in any case, i know that the second one in that list is the one with the red ball in front of it...when i try to read out the Smart test from that first data drive (sdc), i get "Smartctl open device: /dev/sdc failed: No such device"...

 

as per the wiki's warning that "Some newer drive controllers do not like the "-d ata" command line option. If you do not get results from the form above, try running the command with the "-A" option instead (e.g., smartctl -a -A /dev/sda)." i then tried "smartctl -a -A -tshort /dev/sdc", but got the same result "Smartctl open device: /dev/sdc failed: No such device".

 

be that as it may, the fact that the GUI-based Smart test shows an immediate FAILED, i went ahead and entered an advanced warranty replacement on the WD website...i faintly remember reading something about the EADS family of Green drives being the worst of them all...frankly, i've had to replace about 5 or 6 of them in the past few years, some of them being EARS as well...maybe that is why WD has now introduced the "NAS savvy" Red series? i sure hope so, because it really doesn't instill confidence to have to replace drives at the rate of one/two per year in a 14 drive RAID system...or is that normal?

so i got the replacement for the 2TB WD drive with the red ball in front of it today, replaced it with the disabled one, and it's rebuilding as i type this (about 400mins to go).

 

once this drive is rebuilt, the array should be where i had it before this whole mess started.

i want my next milestone to be the upgrade of my 2TB parity drive with a 3TB parity drive, in preparation of expanding my nearly full array with some 3TB drives.

 

my question is this:

 

- when i remove the 2TB parity drive, can i simply put the new 3TB drive into the slot next to it (which would be the last slot in my array, and where i'd like it to be), then re-assign it as the new parity drive, and start the rebuild?

so i got the replacement for the 2TB WD drive with the red ball in front of it today, replaced it with the disabled one, and it's rebuilding as i type this (about 400mins to go).

 

once this drive is rebuilt, the array should be where i had it before this whole mess started.

i want my next milestone to be the upgrade of my 2TB parity drive with a 3TB parity drive, in preparation of expanding my nearly full array with some 3TB drives.

 

my question is this:

 

- when i remove the 2TB parity drive, can i simply put the new 3TB drive into the slot next to it (which would be the last slot in my array, and where i'd like it to be), then re-assign it as the new parity drive, and start the rebuild?

 

Yes.

thanks for confirming, dgaschk!...did it last night, and it's been going all night...6hrs to go...i have to leave for a trip to Portland for a few days, but when i get back, i'll continue with integrating the other two 3TB drives i have...oh what fun

just got back to my system last night after 3 days on the road...something got "stuck" in the rebuilding process of that red-balled drive i replaced with a warranty swap, and i had to brutally reset the server in a non/clean way...but just now i got it all back to normal, with the new drive rebuilt and parity re-calculated.

 

i don't wanna start a new thread for this next question, because it kinda relates to the topic of this thread (which i asked to find an answer to what i'm about to ask):

 

> i want to mount my server on its side, rather than vertical, as it has been...because of this i'd like to remove all of the drives, and re-insert them in a physical order that makes sense to me when looking at it (starting with drive 1 in the upper left slot, and going down and over in drive number, with parity being last in physical slot 15)...can i just stop the array, do the drive re-shuffle, and then start the array again and expect unRaid to know what's what, or do i have to re-assign each drive through the drop-down menus? i faintly remember having tried this years ago, and unRaid complaining that the array/drive-order has changed, or something like that.

 

> if i can't just re-order, what's the correct procedure to change the order of every physical slot?

 

marcusone said in an earlier reply:

unraid doesn't care where the drive is physically located or connected, meaning you can move it around inside your server, just make sure its assigned to the parity drive in the GUI.

but like i said, i think i remember this being not as easy as that.

With version 5 it is that easy. But record the desired locations to be safe (take a screen shot).

thanks dgaschk...when you say "record the desired locations to be safe (take a screen shot)", what screen exactly do you mean i should capture (and in what state), and what info on that screen am i looking for? since i have yet to move them into their desired locations, how could i take a screen shot of them? (sorry if i sound confused/ignorant)

 

in any case, since just moving the parity drive to a different slot (one over from where it was, from 14 to 15) in an earlier step, immediately ended up with a red-balled drive (which may have been totally unrelated) and some sort of error message, i am still reluctant to believe that it is that easy, but i'll keep my hopes up that this was just a "comedy of errors".

 

i am still in the middle of upgrading another 2TB drive to a 3TB one (re-calculating parity as we speak), then i still have to add one of the replaced 2TB drives into slot 14 to add to the array, and *then* i can finally try to re-order all of the drives...will report back with my findings.

The main page shows serial number and disk number.

so you mean that i should take a screen-cap of the main page, with the array stopped and the serial numbers for each slot showing, so that if things go wrong, i'll have an easier time getting things back to the way they were?

Started or stopped.

maybe this is just the case in the simpleFeatures GUI, but if the array is started, the serials don't show...not even sure that they show when the array is stopped (do they, anyone?).

 

so if i disable the simpleFeatures GUI in my go script before i do this, i should see all the drives' serial numbers?

 

sorry for all the questions...i'm hitting the sack for the night...thanks for all your help dgaschk!