I opened my case while attaching a monitor and upon restarting the array I get an orange ball on md2. I have some reiserfsck errors in my sys log. I will attach it. I tried doing the --check and the --fix-fixable. Both abort and say bread: Cannot read the bloc (2): (Input/output error).

 

I don't mind replacing the disk 2 and rebuilding it but disk 5 being red for the same reason is preventing that.

 

Before I try any other reiser commands I would like some advice. Thanks in advance. If you need anymore infer please let me know.

 

Disk 5 is empty BTW.

 

Version 5.0 rc11

syslog.txt.zip

When I try and replace one or the other it says, too many errors, cannot start array. I starts, mounts and is usable. I am not sure what files I am missing from disk 2 yet. I don't want to mess around on iTunes or iPhoto too much and corrupt my libraries.

 

Is there a way I can shrink my array by removing disk 5, then possibly, hopefully replace/rebuild disk 2 and finally add a new disk 5.

 

 

When I try and replace one or the other it says, too many errors, cannot start array. I starts, mounts and is usable. I am not sure what files I am missing from disk 2 yet. I don't want to mess around on iTunes or iPhoto too much and corrupt my libraries.

 

Is there a way I can shrink my array by removing disk 5, then possibly, hopefully replace/rebuild disk 2 and finally add a new disk 5.

Its possible. But don't. That's likely to add more complications, and make the issue more difficult to resolve.

Give folks a chance to read the syslog.

 

This started when you opened the box to do maintenance.

Odds are that you've knocked a cable loose. (power or SATA)...

Powerdown cleanly and double check all connections. If you don't use them, this is a good time to get some LOCKING SATA cables.

Make sure you've not wired two SATA cables together to make your box look 'neater'. SATA cables should be random like spaghetti!

---

[Edit:  Sorry, I have to go to work, but I quickly scanned your log. As you know you have problems...DO NOT change up the disks until you hear back from someone with time to read and interpret all the errors in your SYSLOG.  if the box is still open, powerdown as cleanly as you can and check the cable connections as suggested, but DO NOT swap or remove disks.

...Is there a way I can shrink my array by removing disk 5, then possibly, hopefully replace/rebuild disk 2 and finally add a new disk 5.

You can only rebuild one disk if parity and all the other disks the parity is valid for are good, so no. Shrinking the array would require rebuilding parity and then no rebuild of the other disk would be possible.

 

It's probably just as well that any attempt you made at reiserfsck aborted.

 

Your screenshot shows both drives unformatted. That means that unRAID cannot mount the reiser partition. Drive5 is also red. That means unRAID failed to write to the drive, and so now thinks it needs to be rebuilt.

 

Go ahead and check all of your cables and power connections and see if you can get at least drive2 recognized again. Don't do anything else without further advice.

 

I have checked and double checked all SATA and power cables. It still boots to the same scenario. Disk 2 did not say "unformatted" at first. Disk 5 did not have any data on it as of this failure, disk 2 is my only concern.

 

I will try another SATA cable on the 2 drives.

 

 

No changes when I replace the SATA Cables. Still starts the array in the same status.

 

Since unRaid can see the drives and I was not writing or reading at the time this took place. Can I possibly do a Trust my array restore?

 

Man I've never been more anxious.

No changes when I replace the SATA Cables. Still starts the array in the same status.

 

Since unRaid can see the drives and I was not writing or reading at the time this took place. Can I possibly do a Trust my array restore?

 

Man I've never been more anxious.

There is no array to trust here. 2 of your disks are "unformatted".

 

Maybe you need to tell us more details about your attempt with reiserfsck. I'm beginning to wonder if you didn't make your disks unformatted yourself.

the only reiserfsck commands I ran were 'check' then 'fix' but 'fix' aborted. That was only on disk 2.

 

It was just disk 2 at first. While trying to get 2 up then 5 had the red ball. And then I came here. 

 

The more I look at my files, disk 2 may have more pictures than I thought.

 

Since your last post I tried new SATA cables and different power connector. I even tried moving 5 to a different SATA slot. still red balls.

 

 

the only reiserfsck commands I ran were 'check' then 'fix' but 'fix' aborted.

What command did you type, EXACTLY?

reiserfsch --check /dev/md2

 

came back = bread: Cannot read the block (2): (Input/output error)

 

reiserfscheck --fix-fixable /dev/md2

 

came back with same results then says to replace drive to save your time or run the -B block remap. But I came to the forums and never did that.

The more I look at my files, disk 2 may have more pictures than I thought.

If disk2 is still showing as unformatted, then unRAID can't be reading from it. Maybe it is simulating it from parity and the other drives. Is disk5 still showing as unformatted as well, or just redballed?

 

If anything is different from your previous screenshot, post a new one.

Screen shot still looks the same.

 

I assume its running off of parity.

 

Should I just accept my loss? Or is there hope.

 

 

Screen shot still looks the same.

 

I assume its running off of parity.

 

Should I just accept my loss? Or is there hope.

I don't see how it can be simulating one drive with another drive still missing. It takes the parity data plus all of the corresponding data from all of the other disks to be able to recreate the data for a single drive.

 

What makes you think you can still read disk2? Are you actually reading from the disk2 share, or are you reading files on a user share and the files you are seeing are actually on a different disk?

 

 

No I can't see disk 2. I just thought I had more data on disk 2 than what is missing.

 

I thought I was seeing disk 2 because my iTunes Library looks OK so far. Nothing encountered missing yet. It seems my iPhoto Library took a big hit.

 

I haven't really poke around too much. I didn't want much reading and writing going on in such a fragile state.

 

My shares span all drives. iTunes hasn't encountered a missing file yet. My daughters played a couple movies before I realized they were on iTunes. I don't want to open iPhoto until this is resolved or settled.

 

I can only see what I have on disk 1. I can't tell what I am missing from disk 2. If that makes sense.

Disk 5 has a bad or loose SATA cable. Replace the disk5 SATA cable and reboot. Wait 20 minutes then copy and attach a new syslog.

Replaced disk 5 SATA cable again with no luck. Same screen shot. So I uploaded the syslog. I will attach another syslog at 20 minutes as requested.

syslog.txt.zip

Anything else you can tell us about the hardware that might be relevant? Are disk2 and disk5 on the same controller? What is your power supply?

 

Are you absolutely sure that both ends of the sata connectors and power connectors for both disks are OK?

 

Rebuild is not possible if you can't get to a state where only one disk has a problem. Maybe some file recovery with reiserfsck will be possible but I don't have any experience with that.

Corsair TX650 / fairly new. single rail

 

Disk 2 is on MoBo SATA port

 

Disk 5 is on a 4 port controller card along with disk 4 and Cache

 

I tried using the 4th available port on the controller card on one of the SATA cable swaps. Still same results.

 

I tried my one extra available power connecter on disk 5 during a SATA cable change.

 

I have a 3rd SATA cable I can try. I'll try again.

Just tried another SATA cable. No luck.

 

They both pass SMART and self test.

 

SysLog after 20 minutes was exactly the same so I didn't post it.

STOP!

 

Orange Ball icon means you have data on the cache drive that has not been moved to the data drive, via the Mover script. That is all. This is NOT an issue.

 

Red Ball means that drive has at least 1 failed write. That is an issue.

 

STOP!

 

Orange Ball icon means you have data on the cache drive that has not been moved to the data drive, via the Mover script. That is all. This is NOT an issue.

 

Red Ball means that drive has at least 1 failed write. That is an issue.

 

This is incorrect. See here: http://lime-technology.com/wiki/index.php/Un-Official_UnRAID_Manual#Colored_status_indicator

Replaced disk 5 SATA cable again with no luck. Same screen shot. So I uploaded the syslog. I will attach another syslog at 20 minutes as requested.

 

Disk 5 appears to be responding correctly. You'll need to set a new config and then rebuild disk2. See here: http://lime-technology.com/forum/index.php?topic=32491.msg298738#msg298738

STOP!

 

Orange Ball icon means you have data on the cache drive that has not been moved to the data drive, via the Mover script. That is all. This is NOT an issue.

 

Red Ball means that drive has at least 1 failed write. That is an issue.

 

This is incorrect. See here: http://lime-technology.com/wiki/index.php/Un-Official_UnRAID_Manual#Colored_status_indicator

 

No. What I posted is correct. That wiki page is not correct. Let me dig up the exact quote from Limetech.

STOP!

 

Orange Ball icon means you have data on the cache drive that has not been moved to the data drive, via the Mover script. That is all. This is NOT an issue.

 

Red Ball means that drive has at least 1 failed write. That is an issue.

 

This is incorrect. See here: http://lime-technology.com/wiki/index.php/Un-Official_UnRAID_Manual#Colored_status_indicator

 

No. What I posted is correct. That wiki page is not correct. Let me dig up the exact quote from Limetech.

 

Orange on the shares page is what you say. On unRAID main it is a partially rebuilt disk.

Hrm... Now I'll have to experiment with that. I thought content invalid was yellow, not orange.  Odd choice of colors indeed. 

 

I do know that Orange on the Parity drive means the Parity is being currently being (re)built.