benyaki

Yes, it is doing a parity sync based on the drive after it was formatted (ie it thinks that the drive should be empty and is doing a sync). Should I just let that complete then move the files over, or replace with the old 1.5TB drive, re-do a sync, then replace the drive with the new one and do a rebuild?

Thank you. I (unfortunately?) restarted the array last night to re-insert the 1.5TB drive into a bay and forgot to stop the parity sync. It has already gone through about 25+%. Is it better at this point to just leave it and finish, then transfer the files over from unassigned devices (the 1.5TB drive)? Or will it be quicker to do as you suggested, then replace and rebuild the drive with the new one? Thanks again for the help.

Current array is 2 cache drive, 2 parity drives, 12 disks (slowly upgrading old disks). So I needed to replace an old drive (1.5TB). I stopped the array, replaced it with an already pre-cleared 4TB drive, and then started rebuild. I then noticed that it was formatted as reiserfs, and since I wanted it XFS, I stopped the rebuild, stopped the array, formatted the disk, started the array, then started rebuild again. (THE MISTAKE). The new drive is showing only 4GB used, and obviously when I started a parity sync/rebuild, it is doing nothing to add data back to that disk. It shows the appropriate time for a parity sync, writing data to the drive (approx 30MB/s) and reading from all other driver and parity drives, however there is nothing on that drive being created. I still have the old 1.5TB drive with the existing data on it, and I am wondering how to recover from this the best way. Any input would be greatly appreciated.

Sorry, should have been more clear. A modular server power supply. AC in one end (120V here), which would then slide into a 1/2/3/4/5U+ server case. Not a regular desktop PSU with cables etc.

Looking to pick up a server PSU. Not too picky on the size etc, it will be used as a benchtop 12V supply for RC use. Prefer something that doesnt have a huge footprint. Not a lot of output necessary, will only use about 300W or less of it. Shipping would be to Vancouver, BC Thanks!

Solved. Was looking through syslog, found a bunch of avahi errors, thought I would solve that problem, apparently that was THE problem. somehow didnt show up during my forum search. hope this helps someone down the line. https://lime-technology.com/forum/index.php?topic=48508.msg515980#msg515980

This appears to still be a problem. I just upgraded to 6.30 hoping that something would have been solved, but i am having very slow webGUI response times. Dockers: Plex Sabnzbd duckdns Plugins (all updated): CA autoupdate CA Backup CA Cleanup Community Applications Couchpotato Dynamix Active Streams Dynamix sys info Dynamix sys temp Fix Common Problems (currently running extended test to see if anything comes up) Nerdtools Sickbeard Statistics Hardware: M/B: Supermicro - H8DM8-2 CPU: 2419 - 6 core @ 1800 HVM: Enabled IOMMU: Disabled Cache: 768 kB, 3072 kB, 6144 kB Memory: 0 GB (max. installable capacity GB) - not sure why it has this. There is 8GB installed right now. Network: eth0: 1000 Mb/s, full duplex, mtu 1500 tower-diagnostics-20170206-1848.zip

It looks like the problem has been solved. 4Gb were added in addition to the existing 2GB and everything is working without a hitch after a week or so. Consider it solved.

If you had typed diagnostics in that telenet session, the diagnostics file would have been saved to the Flash Drive. (The diagnostics file is zipped so it is much smaller than a straight text file.) But I suspect that you may not have physical access the server at this point... I understand, but unfortunately as I said before, everything locks up. I cannot access the webgui, telnet (existing is unresponsive - anything entered after that is a new line with no response), new telnet, console, any of the plugins, SMB etc.

Just waiting until I can run out and pick up some more RAM The system has become unresponsive and I wont have a chance to physically check on it for a day or two. Here is a capture (had to remote desktop to the running telnet session and copy it over, sorry if the format is not correct. I did have to trim some of the beginning of the syslog as it was over 320Kb). But it looks to me like a memory issue to me. If that solves it, I may as well grab a better processor as well - any recommendations? syslog_capture.txt

Just was looking through the syslog and cane across some things that might be suspicious and causing problems - check items related to plex - notes out of memory (appears after library update)? I just ran a library update on one section, as well as playing a file remotely (transcoded) and ram usage stayed around 50%. Load was around 0.8 with the single core being used. Suggestions if it is hardware? - start with unlock the second core and up the ram to say 4GB? New diagnostics attached tower-diagnostics-20170120-0729.zip

It has been running fine for quite a long time on UNRAID 6 without issue. This is only a recent problem. Any suggestions or other information that I could provide. I will try to capture a syslog if possible after a crash, however everything seems to lock up.

Diagnostics for your viewing pleasure. Thanks! tower-diagnostics-20170119-1943.zip

Additional images

Having an issue with a server. The problem initially started with the USB drive not being recognized and not booting (just a halt on boot). I upgraded to 6.2.4 and tried to boot but was having problems still. Made the USB bootable on another mac (only have macs) that was running a different OS version and it finally boots (even though no error were given before when making the USB bootable etc). Checked that everything was working, fixed a few settings that were changed and the server runs fine for approx 24 hours, then becomes unresponsive. I cannot access the webGUI, cannot telnet (it will go as far as "Connected to tower. Escape character is '^]'." but does not go an additional line for login. The console also does not respond. One thing that I just noted was when plugging in a USB keyboard, it would not lower up any of the lights (caps lock, num lock etc) if that helps to diagnose the problem at all. I have also tried replacing the files (upgrade again with same version). UNRAID version 6.2.4 Motherboard: MSI 880-GMA-E53 (MS-7623) - upgraded to latest BIOS when the USB drive was not being recognized CPU: AMD Sempron 140 (with only 1 core right now, did not bother unlocking the second until things start working correctly) RAM - 2GB Gigabit ethernet All HDD's are running off the motherboard SATA ports. There is an additional SATA card installed but not being used at this time. All SATA cables checked for connectivity problems, changed RAM slots to ensure no problems there (this is all before I was able to boot the system) PSU - Antec 500w? HDDs 5x 2.0TB WD EARX (1 parity, 4 data) Cache 500GB Plugins: Sickbeard and couchpotato (PHAZE) Nerd tools (cannot install PERL for some reason...) See image for additional mostly stock plugins Docker: Plex SABnzbd Attached are some screenshots of the setup, plugins, docker, system info etc as well as a syslog (sorry, cannot capture a syslog once it stops working, this is just after it has booted up, wondering if there is something in there that someone can pick out). Have started a tail syslog.txt

Not sure exactly what the problem is/was, but I tried replacing bz files on the flash drive with a fresh download of 6.2, thinking something may be corrupt. It did nothing. I downloaded and replaced them with 6.19 version. The GUI is back to normal. It lost my docker for plex for some reason, just added it again and is working fine. Just need to replace some plugins that this broke, but I think all my issues should be solved now, fingers cross, knock of wood, etc.

You have to enable disk shares in Global share settings. You might also have to click the cache disk and enable the share. thanks, working as expected now. that doesn't make me feel like an idiot at all...

Thanks!

The thing is that it was like this from the start, new install, without anything else (plg, docker, etc) added

I just added a second drive to my cache, however it does not look the same as every other image/reference that I can find. The second drive does not show a file system, however it seems, at least to me, to be working right. Files are accessible on the cache drive (via appdata user share). I have also tried removing all drive from cache, changing the order, formatting the second one, and adding the other drive back in and it looks the same. Just wondering if this is something I should worry about or leave as is.

Just upgraded to 6.2 from 5.0 Did not change any Shares etc. Under 5.0 I was able to SMB to my cache drive via smb://IP/cache Under 6.2 I cannot. Not sure if this is a change, or if there is a way to set it up. I like to have access to the root of cache ... I can SMB to appdata (smb://IP/appdata) but it is setup as a share right now as I am setting things up Anyone else run into this problem/change and have any suggestions?

More hardware info attached (via lshw) hardware.txt

I recently upgraded (finally) from 5.0 to 6.2. Before upgrading, I had no issues aside from some plugins starting to fail with no support. On first boot (with nothing installed, no plugins, no dockers, no VMs) I noticed that it was slow. I thought it might just be slow to start. It did not get better. Clicking on a tab or stopping or starting the array, anything within the webGUI is extremely slow, as in it takes approximately 2-3 minutes before the page refreshes, at minimum. (I just waited about 15 minutes to stop the array) newperms has been run just to be sure Downloading through plugins or downloading a docker is quick. Plugins work quickly. No problem streaming anything from the array on the plex docker. Since initial install, I have installed some plugins, a docker, and changed from a single cache drive to a pool (which is another two problems) but nothing has changed for better or worse. Attached is diagnostics Hardware is: Model: N/A M/B: Supermicro - H8DM8-2 CPU: @ 1800 HVM: Enabled IOMMU: Disabled Cache: 768 kB, 3072 kB, 6144 kB Memory: 0 GB (max. installable capacity GB) Network: eth0: 1000 Mb/s, full duplex, mtu 1500 Kernel: Linux 4.4.19-unRAID x86_64 OpenSSL: 1.0.2h Note - the motherboard did initially have an IPMI board installed (and has a second ethernet port for this) however there is no cable attached and I have since removed it from the motherboard and (I think) disabled it in the bios - not very straightforward tower-diagnostics-20160917-1330.zip

Thanks PhAzE. Updated everything, following the test script from yesterday that worked well. Everything is still working. For some reason though, I had to start from scratch configuring SAB again - got the Wizard screen... Oh, and it was Lupus (SLE). Once.

Thanks for the quick reply. I completely agree about not opening up the plugin to a problem by disabling it as a dirty workaround. I have tried changing the URL in the settings in SB for usenet-crawler, however it will NOT save the change from https:// to http://. I have tried editing the URL in /mnt/cache/appdata/sickbeard/config.ini as well, with no luck - changes back to https:// Is there anything else I could try that I am not thinking of? EDIT: I have SB working by just adding a custom search provider for usenet crawler with the proper URL. Just digging into SAB right now, can't seem to find what I am looking for (ie SB will successfully find and send the link the SAB, but SAB won't take it - WAIT 60 seconds etc). I'd be glad to post some logs for any program, is there anything that needs to be removed from them before posting?