October 1, 201510 yr A parity check was just spontaneously aborted with a report of a bad disk2. I am moving files off of it onto another disk using MC, but wanted to see if someone had a different idea from the syslog. Several files were reported as bad during the move, requiring me to skip them. Thanks, Lev tower-syslog-20151001-0020.zip
October 1, 201510 yr I have been having the same issue, my disk 5 connected via SAS card which apparently is not supported by the current version of unRaid 6 causes this issue. It keeps dropping the disk on and off the array.
October 1, 201510 yr Author That's interesting that you say that. I tried copying files from the "bad" drive to a brand new one, cleared twice, and the MC stalled. I rebooted the unraid and could not see it coming up in the browser. I telneted into it and it's there. MC only sees 4 disks out of what is supposed to be 11 +1 ! How is that possible?
October 1, 201510 yr A parity check was just spontaneously aborted with a report of a bad disk2. I am moving files off of it onto another disk using MC, but wanted to see if someone had a different idea from the syslog. Several files were reported as bad during the move, requiring me to skip them. Thanks, Lev Instead of the syslog, it is better on v6 to post the diagnostics zip file you get by going to Tools - Diagnostics. This will include the syslog as well as other information that helps us diagnose your issues. Since your syslog is showing a write error on disk2, I assume is it actually disabled. Since it is disabled, unRAID is not actually reading from the drive. Instead, it is emulating the drive by reading all the other data drives plus parity to calculate the data that would be on the disk. The usual way to deal with a disabled drive is not to copy all the data from it, but instead to rebuild it. In fact, I recommend quitting the copying you are doing and just post the diagnostics zip file I mentioned so we can get a more complete picutre.
October 1, 201510 yr Author A parity check was just spontaneously aborted with a report of a bad disk2. I am moving files off of it onto another disk using MC, but wanted to see if someone had a different idea from the syslog. Several files were reported as bad during the move, requiring me to skip them. Thanks, Lev Instead of the syslog, it is better on v6 to post the diagnostics zip file you get by going to Tools - Diagnostics. This will include the syslog as well as other information that helps us diagnose your issues. Since your syslog is showing a write error on disk2, I assume is it actually disabled. Since it is disabled, unRAID is not actually reading from the drive. Instead, it is emulating the drive by reading all the other data drives plus parity to calculate the data that would be on the disk. The usual way to deal with a disabled drive is not to copy all the data from it, but instead to rebuild it. In fact, I recommend quitting the copying you are doing and just post the diagnostics zip file I mentioned so we can get a more complete picutre. I wish I could! I can't even get into the web GUI now. How can I get that diagnostic file? Is there another way? I've rebooted the system several times to no avail.
October 1, 201510 yr Author Actually, I thought that I was rebooting it from a telnet window by issuing a reboot command, however, I noticed that it was just sitting there. Is there another way to reboot the tower remotely?
October 1, 201510 yr Type diagnostics from console or telnet, ignore any error message, find the diagnostics zip file on your flash drive, and post it.
October 1, 201510 yr Author Type diagnostics from console or telnet, ignore any error message, find the diagnostics zip file on your flash drive, and post it. OK. Created. Now, how do I get it out of the server? Do I need to physically remove the flash drive and copy it on another PC? I am remote right now from work so am not immediately available to do the actual hardware. I was hoping to remotely copy / reboot the tower.
October 1, 201510 yr I usually can access my flash drive over the network, since I have it shared (hidden). Can you? In any case, since you have already saved the diagnostics, it's OK to reboot now. Make sure you don't have your telnet session's current working directory on one of the data drives or user shares or it will not be able to stop the array. Probably you can just cd / then powerdown -r should reboot the server.
October 1, 201510 yr Author I've tried powerdown -r several times to no avail. The server will not reboot.
October 1, 201510 yr Author Probably have to do a hard shutdown with the power switch then. I was kind of afraid of that one. OK. I'll do it tonight and report back. Thank you.
October 1, 201510 yr Author By the way, are there issues with using Supermicro SASLP-MV8 cards with version 6.1.3? I have two of those in my case.
October 1, 201510 yr By the way, are there issues with using Supermicro SASLP-MV8 cards with version 6.1.3? I have two of those in my case. I am using the exact same card in my setup and yes these cards are having some issues with the latest unRaid version. Hopefully there will be a fix soon.
October 1, 201510 yr By the way, are there issues with using Supermicro SASLP-MV8 cards with version 6.1.3? I have two of those in my case. I am using the exact same card in my setup and yes these cards are having some issues with the latest unRaid version. Hopefully there will be a fix soon. I thought it was just the SAS2LP-MV8 that was having problems not the SASLP-MV8. I know I'm not having any problems with my SASLP-MV8 but I'm on 6.1.2 not .3 so maybe that is the difference?
October 1, 201510 yr Author By the way, are there issues with using Supermicro SASLP-MV8 cards with version 6.1.3? I have two of those in my case. I am using the exact same card in my setup and yes these cards are having some issues with the latest unRaid version. Hopefully there will be a fix soon. Do you know what kind of problems? Should I downgrade to the earlier version 6, like .1 or .2? By the way, I am running them on a Supermicro MBD-C2SEE-O
October 1, 201510 yr By the way, are there issues with using Supermicro SASLP-MV8 cards with version 6.1.3? I have two of those in my case. I am using the exact same card in my setup and yes these cards are having some issues with the latest unRaid version. Hopefully there will be a fix soon. Do you know what kind of problems? Should I downgrade to the earlier version 6, like .1 or .2? By the way, I am running them on a Supermicro MBD-C2SEE-O Some users with SAS2LP cards experience slower parity checks with them. Most (myself included) do not.
October 1, 201510 yr Author OK. Here is the diagnostics log. I've rebooted the server and can see all of the drives, including the one that failed and is virtually exported. Sorry, had to break the files in to two. Diag_2.zip Diag_1.zip
October 2, 201510 yr Author Now, without me doing anything at all, I got a message from the server - array turned good. Array has 0 disks with read errors. What does that mean? I'm attaching the latest syslog. tower-syslog-20151001-2103.zip
October 2, 201510 yr By the way, are there issues with using Supermicro SASLP-MV8 cards with version 6.1.3? I have two of those in my case. I also have two of these in my unRAID system and am having NO problems with them.
October 2, 201510 yr Author Over night I tried copying the data off of the "bad" disk, and the process stalled, errors showed up on a different disk, one that precleared twice before going into the system about a week ago, and now again, I cannot remotely reboot the server and only 4 disks out of 13 show up in the midnight commander. Should I downgrade to the earlier version of the unRaid? Everything was smooth with 6.0 Attached is what is on the screen of the monitor directly attached to the server.
October 2, 201510 yr Author It seems that this drive keeps producing errors. Can someone give me a hand with the SMART report? tower-smart-20151002-1403.zip
October 2, 201510 yr What is the current state of your array? Is the disk still being emulated? If so, you should forget about that disk and get your array back to normal by rebuilding, not copying, its contents onto a new drive. Then when things are working again you can try to determine if that disk is worth keeping.
October 2, 201510 yr It seems that this drive keeps producing errors. Can someone give me a hand with the SMART report? Bad sectors. Did you Preclear it before adding it to the array? Unfortunately that drive model (ST3000DM001) has the WORST reputation of all current drives! But they're also the cheapest per terabyte. You are probably going to have to remove the drive and Preclear it a couple more times.
Archived
This topic is now archived and is closed to further replies.