ratmice

Over the last week I have had the opportunity to upgrade 2 disks in my array. Each time I had the array (or at least the GUI) become non-responsive and had to sneak in (via telnet) and attempt to powerdown the array. I was eventually successful at the powerdown and everything eventually went OK. Is there something I'm doing wrong (procedure follows). 1. Preclear new disk 2. Stop server 3. Place precleared disk in my enclosure 4. restart server 5. reassign the disk I want to upgrade. 6. Press the start array button (indicates that the data will be rebuilt and array status says upgrade disk) This is where everything dies, GUI non-responsive, and takes a few times to get the powerdown command to shut down the server. I've waited at least a 1/2 hour to get the GUI back, too. /mnt/dev/ only lists the disks up to the new one (if I'm replacing disk 4 then only disks 1-3 are listed) 7. Power up the server. 8. Press the Start array command (indicates that the data will be rebuilt and disk status says valid configuration) 9. Data is rebuilt and everything appears normal. Why does the first rebuild attempt bog down the server and blow up the GUI?

Yes, that one Samsung is probably on the chopping block.

Thank you for taking the time to reply. I think I'm going to try the bjp script option to speed up the post zero read. It really helped cut down the excessive time to preclear the 6TB drives I got recently.

Is there any danger of the zeros not being written correctly, I see this mentioned occasionally but not sure if it's anything to really worry about. Maybe using bjp's modified pre-clear script with the fast post read would be a good compromise.

So I just replaced my parity disk. The old one is fairly new (8 mos.) and I would like to use it as a new data disk. Since it was precleared originally and has been in use, without issue, does it need to be fully precleared again? Is there a quick way to get it added to the array? I guess I'm just not totally comfortable with which parts of the preclear script I can skip. And I don't want to put undue stress on the drive if I don't need to. It seems like invoking preclear with the -W switch would get the job done, correct?

Unmenu isn't picky like emhttp, you can start/restart it no problem. Just type /boot/unmenu/uu in a telnet prompt and you should be back in business as far as unmenu goes. Thanks. I'm going to not tempt fate and let the rebuild complete before I do anything. Then i'll play with unmenu.

I decided that since I had the data disk I would just go ahead. Looks like it's working fine. One other issue, that I don't know if it is related: I can't get to unmenu anymore. Tower:8080 doesn't respond.

Here you go:

So, before I do something I shouldn't I thought I'd make sure I'm doing the right thing. Backstory: last week I replaced my old 4TB parity with a new 6TB drive, everything went fine. Today I wanted to do a little maint. and stopped my array. I removed an old 2TB drive and placed in it's place a new, pre-cleared, 6TB drive. At the same time I added another drive to be pre-cleared (the old 4TB parity) onto an open SATA slot on the MB. Good so far. I restarted the server and everything looked good. The unassigned drive slot I reassigned the new 6TB drive. Now the GUI says "Stopped - upgrade disk" and the start array button indicates that the array will start and the data be rebuilt. Press the button and GUI becomes non-responsive, went on for about a 1/2 hour without signs of life. I was able to telnet in and reboot the server. Now: on the disk status page all disk assignments are fine (including the new 6TB). The New disk has a yellow ball. The Array stays says "Stopped. Configuration valid". And the text near the array start button says "Start will bring the array on-line and start Data-Rebuild." Everything looks OK. What I'm worried about is the aborted array start doing something bad. The "configuration valid" seems odd, shouldn't that still be "upgrade disk" I'm probably being way overcautious, but I've been burned before. Thanks for any help.

Thanks for the replies. I was planning on going FIFO unless the errors from the OP are more concerning to anyone. There are a bunch of 2TB ones so going by size won't help right now, but would have been my first criteria.

I have recently picked up a couple of 6TB drives that I want to put into use as my array is getting tight on space. I have already replaced my older 4TB parity drive with one of the 6TB drives. So I have another 6TB and the old Parity to put into the array. Most of my other drives are 2TB. I am trying to figure out which ones to upgrade. None are showing any overt signs of failure. SMART reports are showing some warnings by using MyMain and unmenu, but as I read it these are not really operationally problematic, see here: Seagate drives showing high load cycle count values - this seems to be an issue with the head parking routine on these drives. A few drives showing some high fly writes, but not increasing. A Samsung drive showing a multi zone error rate that is not changing. A Hitachi drive that shows an ATA error, but that was when it was first added many moons ago and not recurring So assuming that my assessment of these errors is that they are not immediately indicative of a bad drive, what is the most useful criteria to pick out which drive to replace? Just Power on hours so I get rid of the oldest drives first, or should I look for something else? Thanks for any light that can be shed.

Oooh, picked up 2 as well. Been waiting for a while. Thx.

That's ignoring the 2-week badblocks test and preclear

I've had a ton of problems with recent plex plugins and getting them to actually get the server started. Granted I'm on 5.0.5, but I decided to wipe my flash and start from scratch. I used PhaZes plug in this time and it loaded up the latest Plex Pass version with no issues. If you want to use the "official" plug-in make sure you're using the 64-bit version for UnRAID 6+, not the 32-bit one.

Typically it's better to leave the run as user set to 'nobody'. Next, click on the word "RUNNING" and it should open the page to your server. From there you just need to look int eh configuration to make sure its connected to an outside port. Settings -> Server -> Connect You need to be signed in, and the port needs to be mapped. I originally had it set to 'nobody' but that wasn't working either. What ended up working, before I saw your reply, was accessing the setup page directly on the server (http://tower:32400/web/index.html#!/dashboard) this allowed me to get it up and running. Thanks for help.

You suspect correctly. Thanks for pointing that out. I've just actually dumped Dynamix for the moment as I wanted to go back to a completely Vanilla UnRAID to get rid of some nagging issues. When I get around to installing it again I will definitely redo the sensor config. Thanks.

At a command line, try typing sensors, and report to us what the response is. Also, it may be helpful to know what are the contents of your sensors.conf. root@Tower:~# sensors coretemp-isa-0000 Adapter: ISA adapter Core 0: +35.0 C (high = +89.0 C, crit = +105.0 C) Core 2: +35.0 C (high = +89.0 C, crit = +105.0 C) not sure where to find sensors.conf

Hi, I was having some trouble with my old UnRAID (5.0.5) install and ended up starting fresh. The first thing I'm trying to get up is PLEX. I used to use the old plugin, but was having lots of trouble getting it to work. I have installed this plug in and everything went well. However, when I attempt to access from the plex.tv web page the UnRAID server does not show up. It seems that I am having problem as the user a few pages ago. The attachment is a pic of the settings page and the plex web interface.

I am constantly getting error notices while using Dynamix (2.2.10), all plugins updated via Plugin Control, on my UnRAID (5.0.5) server: PHP Notice: Undefined offset: 1 in /usr/local/emhttp/plugins/dynamix/SystemTemp.php on line 20 The offset can be 0 or 1 and the line seems to be 19 or 20. Not sure if this is a real problem, or not, as everything seems to be working OK.

This sounds fantastic.

Any chance of rolling in a badblocks routine, could be optional, instead of the typical read and write tests? This would give an awesome workout to a new disk.

For anyone that care,s my choice to move to XFS is based on the fact that there is a tool that will allow Macs to read XFS, whereas there is not a simple way to do so for RFS or BTRFS. see: FUSE-xfs http://sourceforge.net/projects/fusexfs/

I am late to the party and finally upgraded my Snow Leopard box. This looks really cool and I am scheming ways to get home early to try it out. How's the iPhone app doing?

Another vote here for the Dune SMART series. I've had it for about three years and it plays everything I've thrown at it. Important to me is the ability to play video folder rips and iso's, all with full menus. I have a bunch in this format and don't really want to bother with transcoding. I use emplexer to hook up to the Plex Media Sever I have installed in my UnRAID box and it's been working great. The one thing that's a bit of an annoyance is that there is a constant audio sync issue with mkv's. I (as well as some others) have to adjust the audio sync to -0.22 sec to get things lined up. I have a macro programmed on the Harmony One to do that with a button press, so no big deal I also have a Pivos AIOS which is cheap and pretty good. I chose the AIOS for the folder rip/iso thing too. The other streamer from Pivos, the XIOS, is not as versatile. The AIOS does not have a PLEX client for me to use, but I don't mind so much. I am about to purchase another Dune SMART unit and think it is one of the best around, it better be for the price!

I've use IBVPN for a while now, reasonable price, servers all over the place, responsive tech. support, torrent only service (if that's what you need) many options for subscriptions, works with OpenVPN. Oh, and did I mention "free trial" - I think you get 6 hours to check it out. http://www.IBVPN.com/