See here: http://lime-technology.com/forum/index.php?topic=28484.0

I ran the getfattr but i do not get any errors. System still crashes whenever I try to copy something to disk2

Since I cannot permanently leave the cache drive disabled to prevent mover from running and crashing my system, and since I need the data on the cache drive to be on a data drive, I did the following.

 

I changed the share settings for an existing share on the problematic disk 2 to exclude disk 2 retaining the split level of 1.

Then I ran mover manually assuming that now the mover would create a new share at the root level of a disk other than disk2 and start copying the contents to it. But this also crashed the server when the first file started copying.

 

I had made sure there is no other share on cache drive that would copy to disk 2.

 

I had pressed the log button on right hand side of UI and it abruptly stopped at the copy line. I am sorry I cannot press syslog because I tried to refresh the syslog page, but it did not refresh and it is not giving me an option of going "back" either

 

Please help!

I don't think it is a problem with writing to the disk2, since I am able to rename folder names on the disk. And since it contains movies, I am able to save subtitles that are being downloaded from the player that accesses it over the network!

 

run reiserfsck check on the cache drive.

Hi dgaschk,

 

Under cache settings, I have set "Use Cache Disk" to "No". After that I tried copying small amounts of data from my windows PC to the share on the problematic disk2. Obviously this must have gone directly to the disk rather than transiting on the cache drive. I was able to copy 5-6 files. On the 7th one, the copy dialog lingered, windows explorer hung and finally I got a message that the network location could not be reached.

 

I am really at my wits end. However I am very thankful for your help. What can I check next?

 

EDIT: Perhaps it is worth noting that every time I lose network access to the server, I have to do an unclean shutdown by keeping the power button on the server pressed. unRaid then boots, mounts the array and immediately starts a parity check. During parity check I can see in Main that there are writes to all disks including disk2. But the parity check always completes successfully without any errors!!!

every time I lose network access to the server, I have to do an unclean shutdown by keeping the power button on the server pressed.

So the unraid local console is locked up as well? If that's the case, I would think it's a hardware issue. PSU, motherboard, RAM or CPU. Have you run a successful memtest session for a long period? (24+ hours)

 

I am really at my wits end. However I am very thankful for your help. What can I check next?

 

EDIT: Perhaps it is worth noting that every time I lose network access to the server, I have to do an unclean shutdown by keeping the power button on the server pressed. unRaid then boots,

 

The is the wrong thing to do and will trash the server. A syslog is required that include the problem. Using telnet or an attached keyboard and display copy the syslog to the flash drive. See System LOG in my sig.

every time I lose network access to the server, I have to do an unclean shutdown by keeping the power button on the server pressed.

So the unraid local console is locked up as well? If that's the case, I would think it's a hardware issue. PSU, motherboard, RAM or CPU. Have you run a successful memtest session for a long period? (24+ hours)

 

No I have not run memtest nor tried from any console attached directly to server. I am doing everything over telnet or web GUI. As i mentioned it is a peculiar problem. And i have sort of ruled out hardware problem by exchanging power, sata cables and replacing the HDD in concern. If it was anything more global, like CPU, PSU, MoBo or RAM, would it selectively affect just one drive slot inspite of me having changed the drive assignment?

 

The is the wrong thing to do and will trash the server. A syslog is required that include the problem. Using telnet or an attached keyboard and display copy the syslog to the flash drive. See System LOG in my sig.

 

Sorry I know that but I get left with no choice once I lose telnet and web GUI to server. A single press of power button does not shut down this machine. So I have to kill it this way.

 

I have tried getting the syslog many times but it has always abruptly ended with the last line indicating cop process. I have got these logs by pressing the "Log" button on right side of Main page in web GUI. Do you think a telnet log would differ, since I lose network almost immediately.

 

I will connect keyboard and monitor to the server, and type tail -f /var/log/syslog I assume this is the correct procedure.

 

Thanks again!

The system should be stopped cleanly. See console commands in my sig.

Phew!!!

 

Finally by hooking up a monitor to the machine, I could take a pic of the syslog.

 

Please refer attachment.

By the way I am also stuck at this screen. Obviously web GUI or Telnet doesn't work. Neither can I type shutdown or any other command directly to the machine using the keyboard. The cursor at the bottom of the above snapshot is just blinking.

 

 

Sent from my iPhone using Tapatalk - now Free

Did you rum memtest?

No I have not. Will do it and revert.

 

 

Sent from my iPhone using Tapatalk

Memtest screen grab attached. Nothing amiss to my untrained eye.

 

But tell me, can a RAM issue so selectively affect a single drive slot?

 

I have obviously ruled out bad HDD or cabling by swapping all of it out.

 

Any recommendations anyone?

 

 

Sent from my iPhone using Tapatalk

Run reiserfsck --check. See check filesystems in my sig.

Reiserfsck results in the attached .txt

 

There does seem to be many errors.

reiserfsck_17_Nov_13.txt

Run with rebuild tree.

I ran rebuild tree. Sorry but I didnt know how to get a log. But the trailing text is attached.

 

My array is still in maintenance mode till I can hear back from you guys.

 

Thanks in advance!

 

reiserfsck_18_Nov_13.txt

Run with check again.

Reiserfsck check log attached.

reiserfsck_19_Nov_13.txt

Looks good.

Thank you. Will try copying data to the array and post back in a few hours. Fingers crossed.

 

 

Sent from my iPhone using Tapatalk

I have said it before and I have to say it again - I am with unRAID because of the extremely superior quality of support on the forums here (without taking any credit away from the fact that it is an absolutely superb product to begin with!). I am VERY glad to report that my issue with the array going offline when I copied to disk 2 is now SOLVED! It caused me a lot of frustration but sincere thanks to dgaschk and other contributors on this thread! You guys rock!!