primeval_god

No, netdata does not at this time provide any Windows support.

Hi @theone, I am having some trouble getting SWAP to work on my setup running unRAID 6.6.7. I have the plugin set to make the swap file on a drive mounted with the Unassigned Drives plugin. The filesystem is ext4, and on hitting start the swap file is created. But the plugin never shows a check for Swap File in use and there is no sign in Netdata that the system is using swap. Checking the system log the only output produced by the plugin is as follows: Mar 11 20:59:24 Nexus emhttpd: req (7): cmd=%2Fplugins%2Fswapfile%2Fscripts%2Frc.swapfile&arg1=start&runCmd=Start&csrf_token=**************** Mar 11 20:59:24 Nexus emhttpd: cmd: /usr/local/emhttp/plugins/swapfile/scripts/rc.swapfile start Mar 11 20:59:24 Nexus rc.swapfile[19406]: Swap file /mnt/disks/Kingston_DataTraveler_20/swapfile re-used and started Mar 11 20:59:41 Nexus emhttpd: req (8): cmd=%2Fplugins%2Fswapfile%2Fscripts%2Frc.swapfile&arg1=start&runCmd=Start&csrf_token=**************** Mar 11 20:59:41 Nexus emhttpd: cmd: /usr/local/emhttp/plugins/swapfile/scripts/rc.swapfile start Mar 11 20:59:41 Nexus rc.swapfile[19941]: Swap file /mnt/disks/Kingston_DataTraveler_20/swapfile re-used and started Any help would be appreciated Edit: I managed to figure this out. Turns out due to the speed of the drive I was using and the size of the SWAP file I was trying to create, the start script was taking long enough that the web request was timing out. Running the script from the console allowed it to finish correctly and everything is working now.

So far as i am aware the S3 sleep plugin does not have any native support for waking up the server on a schedule. When I last used it, I had the "Before Sleep" commands feature to run a small script to program an RTC wakeup alarm. Support for RTC wakeup can be a pain though.

AFAIK it is still not possible to have SWAP on a BTRFS filesystem. I think i read somewhere that it may become possible in Linux Kernel 4.21, but I dont think that is available yet in the wild, much less in unRAID.

Slow in what way? I have been generally pleased with the performance of Traefik, however i have noticed that it recently seems slow to see and update new docker configurations and occasionally misses them all together. I am fairly certain that mine is a setup issue with the docker socket proxy i am using however.

I believe this is because in the Ryzen architecture the infinity fabric is clocked based on the RAM speed. Thus RAM speed effects not just how fast data can be moved to and from RAM but also how fast data can be moved around within the processor and to the peripherals.

I am not sure what the Djoss's official stance is on this, but I update the program in the container all the time. Jdownloader updates far more frequently than the Container it seems. I have never had an issue updating the program within the container.

I am just finishing up the build right now, planning to boot it up in the morning. I didnt really look for server motherboards, but I would suspect the answer is no since the 2700 isnt really a server processor (EPYC) or even the enthusiast grade (thread ripper). Based on my research the motherboard I chose does support ECC RAM, and linux does report that the RAM has multibit error correction available. One warning though if you go with a board like mine, that needs a bios update to support the 2700, be prepared for a headache. Even though i was assured multiple times by the retailer that the bios would and was updated for my processor, it was not.

Similar. As I mentioned i dont use the Chrome plugin only the my.jdownloader webapp and the android mobile app. I am making the assumption that they use the same backend communications channel as the browser plugin(s). Looking back at your original posts though I am not seeing errors on either app that tell me i need to log back in, rather I see occasional (more like often) terrible performance/connection drops/inability to connect to jdownloader. A further difference is that i have never found that bringing up the docker WebGUI (which always works fine) does anything to resolve the issues, they just eventually get better. I have not however found any indication that the Jdownloader application is crashing/freezing/or losing network connection within the docker container and i have maintained 24-7 uptime for upwards of a month at a time.

Have you tried seeking support from the makers of Jdownloader or maybe on its support forum? I am a user of this docker container, the MyJdownloader site, the android app, though not the browser plugins. My first instinct would be to say that the problem probably lies within the Jdownloader application itself and not the docker packaging. From my personal experience running Jdownloader 24/7 I have noticed that the Jdownloader remote access architecture, that connects the MyJdownloader site -> a Jdownloader instance (and likely is behind pluging connection), generally does not work all that well. I see connections from the site run slowly and fail all the time when i am certain the container and Jdownloader instance are running fine. In past version it has been more stable but lately it has been really hit or miss. I havent yet been bothered enough to seek support for it.

Cant you bind mount your svn working directory into the container using the shared option so that both the host and container have access to the same directory? https://docs.docker.com/storage/bind-mounts/#configure-bind-propagation

Sorry for shifting us slightly OT here, but this has managed to pique my interest somewhat. My reading of the comment you linked leads me to believe that I was correct in asserting that there is no special handling (or special case in the code) for mirroring. Rather it is the nature of the parity calculation that in the case of a single data disk the P parity bits end up being the same as the data bits. And in the case of the Q bits the same is true when the disk is placed in a specific slot. This is worth noting, as i have to assume there is a performance impact to doing a parity calculation that results in mirroring vs just sending bytes to both disks, although maybe that impact is negligible. I am not saying that they couldn't have added an optimization for this particular array arrangement, the comments @limetech made on the other side of the link just don't really sound like it to me.

@johnnie.black Interesting, that was not my understanding at all. Any chance you could point me to someplace this special behavior was mentioned/documented?

I think this answer may need clarification. It sounds like the original question is "Does unRAID have a Mirror mode for disks that offers more performance than the normal unRAID data array parity scheme." The answer to that question is no, the data array has only 2 modes single and dual parity. What @johnnie.black is pointing out above is that in the case of an array with only 2 drives parity is equivalent to drive mirroring. This is not a special mode, and as far as i know does not do anything to increase the speed of the array. If you have already tested a system with a 2 drive mirror vs an unRAID system with only two array drives then the speed difference is what it is.

Just an FYI there are at least 2 other threads requesting this feature in the "Feature Requests" forum already.

@johnnie.black Good to here, and just to confirm by your reading this mobo should support the selected RAM (i.e. conforms to the DIMM 1Rx8/2Rx8 memory module requirement)? Again thanks a lot for taking the time to look this over for me. Despite the research i have done on these parts I am always more comfortable with a second set of eyes on it.

Do you have a link to where you found this info?!? Because I found the following on the specs page for this mobo, And the supported memory list https://www.gigabyte.com/us/Motherboard/GA-AX370-Gaming-5-rev-10#support-doc has a couple of verified ECC modules. Additionally I am not 100% certain that of the DIMM 1Rx8/2Rx8 nomenclature, and how to be certain the modules I am looking at qualify. I understand, I have always used consumer grade or workstation parts for my NAS system, alternating between Intel and AMD based on what is the best in my price range at the time. If I can manage to untangle the ECC mess the Ryzen platform best fits my needs/price point at this time.

Thanks @johnnie.black. Do you, or anyone else, have a recommendation as to Memory or Memory/MOBO combinations that would work in ECC mode? It was my understanding that although Ryzen does not have official ECC support, ECC functionality has been achieved with certain mobos (and i was under the impression that this was one).

Bump

First step would be to look here https://docs.traefik.io/basics/ I think you are looking for AddPrefix. I will take a look at my setup when I get off work and see if i can provide a more informative answer.

I also would like to keep the possibility open for eventually adding a GPU and doing passthrough. If anyone can comment on the IOMMU grouping of the MOBO i would appreciate insight as to the viability of the layout.

I am planning major overhaul of my unRAID server in the very near future. I am planning on jumping on the Ryzen bandwagon. I have been following ryzen news here on the forums and I am pretty confident in the parts I am looking at. Never the less, I would like a second set of eyes on it, more if possible. The link to PC part picker is below, I have yet to finish the build the Mobo, RAM, and Processor are my current concerns. I am already aware that the motherboard may need a bios update before it will accept the processor, not sure if i can find/get one with the proper bios to start with. One primary area of concern is the RAM. For my server I really want ECC RAM. PcPartPicker suggests that the it may not be compatible, but based on the documentation i think it should be. If anyone knows definitively I would appreciate confirmation. https://pcpartpicker.com/user/primeval_god/saved/Z33mqs

Let me preface this by saying i know exactly 0 about routing tables in linux and how they are modified. Since the unRAID OS runs in ram and is unpacked fresh from the USB drive on boot, changes that you make to the underlying linux system at runtime do not persist. The only things that do persist across reboot are those things that the unraid GUI / Config files allow you to tweak. With that said I dont know one way or the other whether unRAID has a configuration option for static routes, but if you have looked and not found it then possibly not. In that case your only choice is to re-make the changes on every boot. I would suggest writing a script that makes the changes and then run it on startup using the CA User Scripts plugin.

Are you sure about that? 172.17.0.2 looks like an internal address that docker would assign to the container.

@Jclendineng As mentioned somewhere above, you can already setup traefik on unraid using the docker hub search feature in CA.