fotd

You may want to check around on the b660 chipset with 13th gen cpu. I thought I read there were some issues with power consumption and overheating. Something about not being able to adjust pl1/2 settings. Unless you already have it I’d think going for a z790/690 would make more sense for the added pcie lanes.

Your power bill is going to be ridiculous lol. No clue if you can install it, it's been probably 8+ years since I last had to direct connect to a vnx like that. Everything was done via java app and never really looked at the underlying system that much. I still have a couple locations with dormant vnx racks that have been powered off for a long time. I'm trying to remember how those work, they have a main power DA with battery back up that each feeds 2 pcs that do the controlling if I remember right. You obviously could only run one. There might be a secondary controller/cache system, I'm just having a hard time picturing it. I believe the current OS runs on the first 5 (vault)drives of the first shelf. There's usually a sticker from the factory don't touch and the system defaults to boot from there. Whether you could boot from usb I can't remember.

So my server isn't booting up anymore, unfortunately not getting much clues from the on screen halt. Anyone know how to turn on the syslog to flash drive through the config files? I know I turned it off via the gui. I assumed it was just add a "1" to syslog_flash in rsyslog.conf, maybe I just need to put in a path logs\syslog? local_server="1" server_protocol="udp" server_port="" log_rotation="" log_size="1M" log_files="1" remote_server="" remote_protocol="udp" remote_port="" syslog_flash=""

That sounds like what's happening, thanks! Not sure I even have room for a new nic but I'll check. If not maybe I'll have to downgrade. For the Samsung SSD, it works great if I have it on one of the Raid controllers. I know I've swapped cables and ports and the issue has followed that device. I guess I'll try again.

So I've been having an issue the past 2 days with network dropping out. The server stays up and appears to be working as normally but no external network access in or out. It seems to happen randomly. Last night booted at 10:43 and it went out at 11:14 (syslog7-1-19), booted backup and at best guess it went out around 3:00am. Booted it around 7:45am and it's been up since (syslog 7-2-19). It's an older AMD fx-8350, at least 6+ years old. I was on 6.71 yesterday and updated to 6.72 and the issue persisted. When going to 6.7 I did have odd networking issues and had to get rid of static and go dhcp to get it to work. If I manually added a default route at the time it would work with static. unraid-diagnostics-20190702-1418.zip syslog7-2-19 syslog7-1-19

Had the same issue with the route. Worked when I added it manually but after starting array it would clear it out. Ended up just setting it to dhcp with a reservation and it worked through the reboots.

I run Unraid as a guest to ESXi still. I've been running it this way for years so before unraid had kvm. I was hesitant to switch because I used to have issues with Unraid locking up during large file transfers and becoming totally unresponsive and was worried that would affect the VMs as well. Also the idea of Unraid needing to be up kind of bothered me. They've resolved the unresponsive UI issue and I think there's actually more flexibility with KVM/Unraid as far as device pass thru. My next upgrade I'll probably try it with KVM just to see but won't be surprised if I go back to this way. If I were you I wouldn't build a secondary server, maybe just rely on cloud backup if you're concerned with the data. Google drive unlimited is only $10 a month which is probably at least what a new server would cost in power every month. Just add another DC to Unraid and see how it performs, if it works well shutdown the old one.

If you're looking for more feedback.... I agree there is a scaling issue on lower resolutions, on my 1080p laptop I need to scale down to 75% to get more then one post on the page. Also I appreciate the white theme, it's much easier for me to read. But if you're going to look at more theme options I would consider lightening the dark black stripes on the white theme. To me it's almost as jarring as the black on white theme on a laptop. Probably due to reflectiveness of touch laptop screen.

These are down to $160 for Black Friday, I assume through Saturday.

You may want to check this thread, I've only used SSD cache drives but stopping using btrfs solved my issue. This was plaguing me for at least a year or more, if multiple things going and moving a file over 10GB would hose everything up until it was complete. I only have 1 cache drive so going to xfs was easy.

After seeing that guy on reddit who got 22 I scooped one. Just shucked it and it's a WD80EFZX.

Maybe this one http://lime-technology.com/forum/index.php?topic=53153

Probably not the right place to put this. But after starting docker with unraid and now using it on other systems, Unraid has spoiled me. The interface they have setup is awesome and works great. Not that setting up dockers and updating is hard, but it is kind of a pain. You can't just update an image, you have to pull it, clone it, delete it, rename it, rerun run command, blah, blah, blah.... Having all that stuff done via automated scripts makes everything so quick and easy. Maybe I'm doing something wrong or have missed it but is there a decent web interface like unraids for docker on other systems? I'm going to give docker-ui a try, but at first glance nothing appears to be as seamless as unraids built in interface.

Happened to see your post in the 1.1 thread and looked at your log. You are getting the avahi not found error. Nov 3 10:36:23 TMCINAS02 root: Starting Avahi mDNS/DNS-SD Daemon: /usr/sbin/avahi-daemon -D Nov 3 10:36:23 TMCINAS02 avahi-daemon[9528]: Failed to find user 'avahi'. Nov 3 10:36:43 TMCINAS02 root: Timeout reached while wating for return value If you follow John_M's advice I bet it fixes it for you. If they don't exist just add the 2 avahi lines at the bottom. http://lime-technology.com/forum/index.php?topic=50635.msg488473#msg488473

Check your logs for avahi errors. Aug 5 13:52:02 DarkTower emhttp: shcmd (20782): /etc/rc.d/rc.avahidaemon start |& logger Aug 5 13:52:02 DarkTower root: Starting Avahi mDNS/DNS-SD Daemon: /usr/sbin/avahi-daemon -D Aug 5 13:52:02 DarkTower avahi-daemon[20325]: Failed to find user 'avahi'. If you see the failed to find user error, you need to add it to your /etc/passwd file. http://lime-technology.com/forum/index.php?topic=52628.msg509201#msg509201

Had the same issue, it's the avahi. Check this thread http://lime-technology.com/forum/index.php?topic=50635.msg488473#msg488473 I added the avahi and avahi-autoipd lines to my /etc/passwd file and it instantly fixed the issue. I upgraded from 6.12 to 6.21 and every action in the gui took 30-40 seconds.

Thanks

Seems my plexpass docker reverted back a few versions on restart. It was at 1.21 at some point because streaming brain was up and working. Now that 1.22 is out it's now at 1.14 and won't update on restart. Not sure if it's just me or something changed from plex on how the latest version is listed. Anyone recommend a different docker version if this one is dead? I always like this one because you always got the plexpass version right away without having to wait for someone to update their servers. I can't remember what it was but there was something going on there that for some reason other versions weren't doing it that way.

Thanks for looking into it! I could just go the nginx route, maybe that's better in the long run. I reloaded and I'm still having issues, now I'm getting an error on the cert. I reset the perms of the directory to 777 but didn't matter. The cert and key exist in the root of /config 07-Jul-2015 23:52:01 - INFO :: MAIN : Retrieving latest version information from github /opt/mylar/lib/requests/packages/urllib3/util/ssl_.py:90: InsecurePlatformWarning: A true SSLContext object is not available. This prevents urllib3 from configuring SSL appropriately and may cause certain SSL connections to fail. For more information, see https://urllib3.readthedocs.org/en/latest/security.html#insecureplatformwarning. InsecurePlatformWarning 07-Jul-2015 23:52:01 - WARNING :: MAIN : Could not get the latest commit from github 07-Jul-2015 23:52:01 - INFO :: MAIN : Remapping the sorting to allow for new additions. 07-Jul-2015 23:52:01 - INFO :: MAIN : Sucessfully ordered 3 series in your watchlist. 07-Jul-2015 23:52:02 - ERROR :: MAIN : Error creating SSL key and certificate: [Errno 13] Permission denied: u'server.key' 07-Jul-2015 23:52:02 - WARNING :: MAIN : Unable to create certificate and key. Disabling HTTPS 07-Jul-2015 23:52:02 - WARNING :: MAIN : Disabled HTTPS because of missing certificate and key. 07-Jul-2015 23:52:02 - INFO :: MAIN : Starting Mylar on http://0.0.0.0:8090/ 07-Jul-2015 23:52:02 - INFO :: MAIN : Initializing the DB Updater. Edit: it appears even with ssl disabled I'm now also having trouble pulling nzbs from mylar->sab gets stuck fetching. Was working before update, I'll look at it more tomorrow. I see the same things in my log. Not being able to pull the latest github version doesn't really matter to me. I have to update the container in order for the version to update anyway as updating from github never worked for me. As far as SSL I think you have to provide the files because they won't generate for me either. Although reading again it sounds like you are setting the path. So I'm not sure... I just removed the container and image and started over. I was able to add my comic vine api and set the comics and downloads directory. I set it up to use a blackhole. The permissions on the folder created on my cache drive are drwxrwxrwx 5 nobody users 83 Jul 19 20:43 mylar Furthermore that directory looks like this: root@unRaid:/mnt/cache/apps/appdata/mylar# ls -hal total 44K drwxrwxrwx 5 nobody users 83 Jul 19 20:44 . drwxrwxrwx 38 root root 4.0K Jul 19 20:36 .. drwxr-xr-x 2 nobody users 56 Jul 19 20:02 cache -rw-r--r-- 1 nobody users 3.7K Jul 19 20:44 config.ini drwxr-xr-x 2 nobody users 58 Jul 19 20:26 logs -rw-r--r-- 1 nobody users 33K Jul 19 20:44 mylar.db drwxr-xr-x 4 root root 95 Jul 19 19:57 post-processing root@unRaid:/mnt/cache/apps/appdata/mylar# I wouldn't think you'd need to check permissions since this is how it seems to default for me but it's worth a shot. EDIT: One thing if you haven't done so yet is to make sure you delete the config folder from your cache drive when you're starting fresh. Before I did this test I mode sure to move my config folder off so that any past permissions or other junk weren't going to interfere. Thanks Pinion. I did dump it and start fresh (new directories too). As we were discussing in the PM's I think the issue is more of a sabnzbd issue that started with some of the recent code changes. The suggested config.ini change for the host_return (or almost any config change) would cause me to get an odd permissions like issue that I wasn't able to get around couldn't load pictures/write files/etc. Would cause me to have to completely wipe. I'll double check the perms against the ones you posted, but I had reset all the perms to wide open. I haven't checked the main mylar github in a few days to see if they've fixed the issue that caused the sabnzbd problems. As far as the ssl once I can get sabnzbd working again I'll try again. Thanks again for your help.

Thanks for looking into it! I could just go the nginx route, maybe that's better in the long run. I reloaded and I'm still having issues, now I'm getting an error on the cert. I reset the perms of the directory to 777 but didn't matter. The cert and key exist in the root of /config 07-Jul-2015 23:52:01 - INFO :: MAIN : Retrieving latest version information from github /opt/mylar/lib/requests/packages/urllib3/util/ssl_.py:90: InsecurePlatformWarning: A true SSLContext object is not available. This prevents urllib3 from configuring SSL appropriately and may cause certain SSL connections to fail. For more information, see https://urllib3.readthedocs.org/en/latest/security.html#insecureplatformwarning. InsecurePlatformWarning 07-Jul-2015 23:52:01 - WARNING :: MAIN : Could not get the latest commit from github 07-Jul-2015 23:52:01 - INFO :: MAIN : Remapping the sorting to allow for new additions. 07-Jul-2015 23:52:01 - INFO :: MAIN : Sucessfully ordered 3 series in your watchlist. 07-Jul-2015 23:52:02 - ERROR :: MAIN : Error creating SSL key and certificate: [Errno 13] Permission denied: u'server.key' 07-Jul-2015 23:52:02 - WARNING :: MAIN : Unable to create certificate and key. Disabling HTTPS 07-Jul-2015 23:52:02 - WARNING :: MAIN : Disabled HTTPS because of missing certificate and key. 07-Jul-2015 23:52:02 - INFO :: MAIN : Starting Mylar on http://0.0.0.0:8090/ 07-Jul-2015 23:52:02 - INFO :: MAIN : Initializing the DB Updater. Edit: it appears even with ssl disabled I'm now also having trouble pulling nzbs from mylar->sab gets stuck fetching. Was working before update, I'll look at it more tomorrow.

Having issues trying to load ssl on the mylar docker. Says "WARNING :: MAIN : The pyOpenSSL module is missing. Install this module to enable HTTPS. HTTPS will be disabled." Is this something I can update or does it need to be added to the docker? Still new to this docker stuff. Thanks!

to bad this doesn't work yet seems like it would be perfect

I know people have gotten unraid working under vmware workstation, but run into the 4 disk issue. Now that ESXi is free, I'm wondering how that would work. Boot ESXi, use a small VMFS volume to host RDM's that point to physical disks in the system for UNRAID to use. I'm not exactly sure what the limit of disks are for ESX, but I know I can add at least 6 vmdk's in ESX3.5, I'm assuming RDM disks are the same (looks like there's at least 20 SCSI addresses in the options. I don't see why it wouldn't work, I don't think ESXi allows USB pass through, so that might be a problem but solvable. I've never used ESXi, I know they've stripped some functionality, but not sure of the top of my head. The only real issues I see with doing this is whether ESX will allow regular SATA/IDE controllers to host RDM's. There are issues with ESX support on SATA/IDE controllers hosting VMFS volumes, but there are a few hacks I believe you can do to get it to work. Also a lot of Sata raid cards work. If I can get some time maybe I'll give it a try.

Have you done the windows audio fix? I'm pretty sure I ended up having to do it regardless of sp1. http://lime-technology.com/forum/index.php?topic=1162.msg8269#msg8269 http://blogs.technet.com/markrussinovich/archive/2007/08/27/1833290.aspx#1841934 http://courtneymalone.com/2007/08/28/a-note-on-vista-network-speed/

Ya you're right, I always forget about the B/C designation. I think it comes from years of references to "Class C" networks, which is what the majority of people tend to call their Class B network. Since the common practice is to take your class b and break it into individual C's or subnet it down, most people get the C stuck in their head. In any case I'm just glad I haven't had to deal with the stuff in years, I was never very good at math .