azzilla

Members
  • Posts

    37
  • Joined

  • Last visited

Posts posted by azzilla

  1. 17 hours ago, skois said:

    Ok, i made some edits that improved a little the "Error when assembling chunks".
    Edited the file in appdata/nextcloud/php/www2.conf adding the following UNDER the [www]

    
    ; Tune PHP-FPM https://docs.nextcloud.com/server/latest/admin_manual/installation/server_tuning.html#tune-php-fpm
    
    pm = dynamic
    pm.max_children = 150
    pm.start_servers = 15
    pm.min_spare_servers = 10
    pm.max_spare_servers = 30

    Also on appdata/nextcloud/php/php-local.ini i have the following 

    
    date.timezone = Europe/Athens
    upload_max_filesize=16G
    memory_limit=8G
    max_execution_time=7200
    max_input_time=7200
    post_max_size=16G
    max_file_uploads = 200
    default_socket_timeout = 7200
    
    
    ; Enable PHP OPcache https://docs.nextcloud.com/server/latest/admin_manual/installation/server_tuning.html#enable-php-opcache
    opcache.enable=1
    opcache.interned_strings_buffer=8
    opcache.max_accelerated_files=10000
    opcache.memory_consumption=128
    opcache.save_comments=1
    opcache.revalidate_freq=1


    Note that some of the above options are redundant. But i decided to group the together so i know what's going on.

    Tried to upload a 1,5GB file and completed fine (before i was getting the error). Trying now a 4GB file. Will edit soon to let you know how it goes.

     

    Aaaaand i got the error again :D
     

    Have you tried to disable chunking completely? That made a huge change for me.

  2. I don't get this. I tried plenty of settings and tweaks without trying to nag in this forum but now i am at a loss for ideas.

    The performance of uploading files through the webgui is soooooo slow. Around 30MB/s which i deem slow on a gigabit LAN and WAN.

    I totally understand that NC isn't perfect ootb in the sense that i needs some tweaking to get the best performance, but i tried a enough of the solutions without significant results that i am close to surrender my attempts.

     

    Share is cache enabled and according to "Diskspeed" i have no issues with read/write. Similar share settings for my "Downloads" share and speeds in nzbget is around 100MB/s.

     

    Has anyone managed to make any changes or tweaks that performs better than 30MB/s or have any suggestions as to what i should focus on?

    It's not as if i can't live with 30MB/s via the gui, but if higher speeds are possible i would really like to know.

     

    What i've tried.

    Using Postgresql instead of MariaDB. (Marginally improved CPU load when heavy use)

    Enable OPcache (No measurable difference)

    Tuned php-fpm (No measurable difference)

     

    What i haven't tried

    Redis backend

     

    NC Documentation links

    Link 1

    Link 2

     

    Linuxbots

    Link 1

     

  3. On 11/28/2020 at 12:40 PM, skois said:

    So to enter nextcloud docker console, click on the docker icon and select console option

    unfortunately nextcloud docker doesn't include rsync, but since you know that with krusader did that in about 10 sec. 

    While on docker's console do 

    cp input_path output_path
    e.g. cp  /data/admin/files/testfile.mp4 /data/test/testfile.mp4

    if you want to cp folder you have to do it like this

    cp -r /data/admin/files/ /data/test/
     

    If it takes lot more that 10 seconds for the same file. Probably something is up with the docker engine? Not sure what you cause that. 
    But this will help you narrow it down

    I have also followed Spaceinvader One's guide and having sort of the same issue and i can't understand why. Copy within the docker console works as expected, 5gb file copies in 8-10 seconds.

     

    I am getting slow uploads only (10MB/s), downloads are fine (90-120MB/s). There are som topics concering this issue over at Nextclouds forum but I am not getting anywhere with the solutions suggested over there.

     

    https://help.nextcloud.com/t/slow-upload-speed-i-need-advice/60909/14

     

    I can't remember this being an issue looking back on using Nextcloud for almost 2 years. I had some trouble uploading large files and had to make some edits to the .conf file for swag to fix that.

    Other from that, nothing has changed.

     

  4. On 10/18/2020 at 5:32 PM, Tabraxis said:

    Hello All,

     

    Does anyone know where to change the date in the docker? for some reason mine thinks its Oct 25, 2020 around 3:29pm (as of today's date.) i went looking for .env and managed to change it to EST from PST, but i'm not seeing a line for the date, my Unraid is verified correct date/time.

    image.png.68bcb147662c0c2f7874ae274b259d08.png


    Thanks,

    I have this same issue. Where did you find the .env file though?

  5. On 9/11/2020 at 11:10 PM, argash said:

    So here's an interesting one. I setup reverse proxy with the LE/SWAG container and everything is working fine. However, when I go to generate new labels the QR codes are still pointing to the internal IP address. I've updated the App URL variable with the external URL and restarted the container but new codes still point to the internal IP not the URL.

    Do you mind sharing your proxy conf file? :)

  6. I can not for the life of me get the Mattermost docker working with remote proxy. The documentation on Mattermost's site has an example of the config file for a subdomain but it is not working with SWAG (letsencrypt) and NGINX.

    Does anyone have this up and running and could share their configuration file?

     

  7. 13 minutes ago, johnnie.black said:

    There's filesystem corruption on the cache pool, best bet is to backup any important data there and re-format the pool.

    Okay. Do you know why it starts every night at midnight?

    I read in this thread that loop2 refers to the docker image. It made me think that there could be an error with just the docker image.

  8. I have been having an issue where my log fills up with BTRFS error codes. It starts every night at midnight and then fills the log at around 7.30 in the morning causing execution errors for my docker container.

     

    I have tried my best at finding a solution by myself but i can't seem to fix it.

     

    Things i've tried:

    Disable docker and delete the image.

    Increased the log partition but it fills up eventually anyways.

     

    The only thing that helps to get rid of the execution errors is a reboot of the server.

    bighoss-syslog-20200720-0539.zip

  9. Hey Azzilla,
     
    Did you find anything?  I am just bored and I want to see if I can install something on my server.. I have a HDHomeRun Extend already but still..
     
    Let me know!
    -n 
     
     

    I got Xteve working but I didn’t like Plex’s gui for IPTV so I got rid of it completely.
  10. I have tried tvheadend with proxy, worked like crap. Xteve never got it working. There is a docker for telly but I can’t seeem to find anything on how to set it up. 

    Do you have a setup that works flawlessly? Could you share the setup you have? 

    Please and thank you

  11. I had a BTRFS error on my cache pool so i hade to format. I backed up my app data with this guide but something went wrong and I was not able to save the docker images. 

    I set up my docker when I installed Unraid a long time ago and managed to set the docker image to rest on a share that i named "Docker".

    This (at least i think) resulted in the docker programs acting up. Radarr isn't finding any media, can't Unifi Controller loses it's config.

     

    Reading the guide for setting up Docker and of course, the recommendation is to set the Docker image directly to the Cache device.

     

    Can I move the Docker image to the cache drive with midnight commander or some other way without loosing the docker apps?

     

     

    • Like 1
  12. Any unclean shutdowns? Run a scrub on the pool, if that can't fix it best to backup cache, re-format and restore.

    Never performed a scrub before. How would I go about doing that?


    Skickat från min iPhone med Tapatalk
  13. Don't know if this issue is something that I should panic about but my server has been starting to act strange. 

    CPU is sometimes (rarely) at 100% and the only thing that then works is to take the array offline and restart. 

     

    Haven't seen this behavior at all before.  

    Skärmavbild 2019-01-14 kl. 20.24.09.png

  14. How do you do when you edit the file?
    Do you take out the memory o put it in another computer or shut down your server first then edit the file or do it from cli?


    For Windows Use “Putty” or Terminal on Mac and SSH to your server. Type mc to open midnight commander and copy the go file to a share and edit in Notepad++ for windows or Atom on Mac. Then copy the file back to the original location. Ezpz


    Skickat från min iPhone med Tapatalk
  15. Just wanted to updatea bit and say that the temperatures has decreased drastically since i did some modifications to my Silverstone DS380. I 3d printed a "cooling duct" that i found from someone who had similar problems. https://blog.briancmoses.com/2017/04/creating-a-cooling-duct-for-the-silverstone-ds380.html

    I also installed two Noctua NF-F12 Industrial 3000rpm fans. Loud af but as my server is in the basement, thats not an issue.

    Thanks for all the advice!

    unraidtemp.thumb.JPG.8f10cc0bc70466070965a738f62e4ca8.JPG

    • Like 2
    • Upvote 1