france1's questions

I am trying out pixray/clipit but cuda fails to allocate the remaining 1GiB of my graphics card. My graphics card is "Nvidia GTX 1660 super" which has the same amount of RAM as the "Nvidia GTX 1660 Ti" which belongs to somebody I know - and it works just fine there, probably using Windows though.

My system is 64 bit, Linux Manjaro-Desktop2 5.15.60-1-MANJARO #1 SMP PREEMPT Thu Aug 11 13:14:05 UTC 2022 x86_64 GNU/Linux.

CUDA:

Why does PyTorch (CUDA) crash even there's 1GiB of available vram? Here you can find information about my graphics card: https://www.nvidia.com/de-de/geforce/graphics-cards/compare/?section=compare-16

Normally the Linux Kernel flushes the write cache if a specified time limit is exeeded or the cache is full. This leads to stalling with HDDs since the writing process makes reading processes much slower.

That's why I want to avoid a flush of the write cache if there's only a bit read activity. Since I have 40 GB of RAM this should not be a problem. My thoughts on that would be to set the time limit to five minutes, tune the write cache size to your needs and write a script that flushes the cache ONLY if there is 0 read activity. I don't know how to do that, would the shell command sync offer good enough performance?

My goal is to have a write cache with a size like 1, 2 GBs. Flushing this would take a long time. Is there anything that can be stopped as soon there's read activity again or a higher disk PSI (would that even react fast enough??)

I have two Seagate (non-smr?) drives: ST1000DM010-2EP102 - but I guess that information isn't needed.

With dirty_ratio I currently set a dynamic cache size.

echo 360000 > /proc/sys/vm/dirty_expire_centisecs
echo 360000 > /proc/sys/vm/dirty_writeback_centisecs

This would set the maximum time to start flushing the write cache to 1 hour.

Reference: https://unix.stackexchange.com/questions/30286/can-i-configure-my-linux-system-for-more-aggressive-file-system-caching

Useful commands:

With the following command you can see how much is in the write cache:

cat /proc/meminfo | grep Dirty

Output:

Dirty:               104 kB

But this is not very important..

Where am I stuck?

I want to be able to almost instantly cancel the sync command (is this even possible) and I want to know if there is a process reading. That's all what I need to write the script.

I found an old script that's able to monitor read activity: awk '{print $1}' /sys/block/sdb/stat

How to use this?

if [ $old-value -lt $new-value ];then COMMAND;fi

I'm using clamonacc on my desktop computer and servers. I use a raspberry pi as clamd server. The clamd process only uses ~25% of all 4 cores so with htop I see it's using like ~100% and 400% is the maximum a process could use because the raspberry pi has 4 cores. RAM isn't full and there isn't much i/o wait (according to htop)

I'm currently testing clamonacc on one server but the clamonacc queue fills up until it's limit (too many opened files) I only have these problems if I use a dedicated clamd server - with the socket it doesn't fill up the queue often.

There's always only one clamd thread (two if I look with clamdtop because clamdtop needs another one) on the raspberry pi.

I think that the tcp connection makes all slower. The traffic goes through a switch from my server with clamonacc to my raspberry pi clamd server - but that shouldn't be a problem. htop says that there's ~7MB/s (3000/8000 packets)

The raspberry pi isn't the problem because I used another device and it still didn't work

How can I fix this?

Both systems have ubuntu installed. Clamav version: 0.103.3

btw, clamdscan doesn't seem to have problems but it's slower through tcp

edit: I got it.. The problem's like you would run clamav under root but you don't exclude root so it's scanning in a loop. I use clamonaccess mount-path / to scan my system. The problem is that even if I exclude both, clamav and root, it will still scan in a loop. If I change to a local socket it wont scan in a loop but if I use a dedicated clamd server it does

Is this a bug by clamav?

I'm using Apache with following configuration:

Timeout 60
KeepAlive On
MaxKeepAliveRequests 100
<FilesMatch \.php$>
        SetHandler "proxy:unix:/var/run/php/php7.4-fpm.sock|fcgi://localhost/"
</FilesMatch>
HostnameLookups Off
<Directory />
        Options FollowSymLinks
        AllowOverride None
        Require all denied
</Directory>

<Directory /usr/share>
        AllowOverride None
        Require all granted
</Directory>

<Directory /var/www/>
        Options Indexes FollowSymLinks
        AllowOverride None
        Require all granted
</Directory>
<Directory /var/www/html/>
   Options +FollowSymlinks
   AllowOverride All
</Directory>
<IfModule mod_expires.c>
ExpiresActive On
FileETag None
ExpiresByType text/javascript "access plus 1 month"
ExpiresByType text/x-javascript "access plus 1 month"
ExpiresByType application/javascript "access plus 1 month"
ExpiresByType text/html "access plus 1 month"
</IfModule>

<IfModule worker.c>
StartServers 2
MaxClients 50
MinSpareThreads 1
MaxSpareThreads 2
ThreadsPerChild 15
MaxRequestsPerChild 0
</IfModule>

<IfModule prefork.c>
          StartServers          2
          MinSpareServers       1
          MaxSpareServers       2
          MaxClients           60
          MaxRequestsPerChild 1000
</IfModule>

Then this is my external.conf (where the traffic from varnish's gonna be encrypted):

<VirtualHost *:443>
        Protocols h2 http/1.1
        ServerName cloud.privacyy.ch
        ErrorLog              /var/log/apache2/external-https_error.log
        CustomLog             /var/log/apache2/external-https_access.log combined

        Include /etc/letsencrypt/options-ssl-apache.conf
        SSLCertificateFile /etc/letsencrypt/live/cloud.privacyy.ch/fullchain.pem
        SSLCertificateKeyFile /etc/letsencrypt/live/cloud.privacyy.ch/privkey.pem


    ProxyPreserveHost       On
    ProxyPass               / http://127.0.0.1:8080/
    ProxyPassReverse        / http://127.0.0.1:8080/

    <IfModule mod_rewrite.c>
          RewriteEngine On
          RewriteRule ^/\.well-known/carddav https://%{SERVER_NAME}/remote.php/dav/ [R=301,L]
          RewriteRule ^/\.well-known/caldav https://%{SERVER_NAME}/remote.php/dav/ [R=301,L]
          #RewriteRule ^/\.aurora https://cloud.privacyy.ch/aurora-staefa.ch/ [R=301,L]                                                                                                                    
    </IfModule>

    Header set Cache-Control "max-age=84600, private"
    <FilesMatch "\.(js|css|php)$">
           Header set Cache-Control "max-age=63072000, private"
    </FilesMatch>


    Header always set Strict-Transport-Security "max-age=63072000; includeSubdomains"
    Header always set X-Content-Security-Policy "allow 'self';"
    Header always set x-xss-protection "1; mode=block"
    Header always set X-Robots-Tag "none"
    Header always set X-Download-Options "noopen"
    Header always set X-Permitted-Cross-Domain-Policies "none"
    Header set X-Content-Type-Options: "nosniff"
    Header set X-Frame-Options: "sameorigin"
    SSLCompression off
</VirtualHost>

This is internal.conf:

<VirtualHost *:8181>
        protocols h2 http/1.1

    DocumentRoot /var/www/html
    <FilesMatch \.php$>
        SetHandler "proxy:unix:/run/php/php7.4-fpm.sock|fcgi://localhost/"
    </FilesMatch>

    <IfModule mod_headers.c>
            Header always set Strict-Transport-Security "max-age=15552000; includeSubDomains; preload"
    </IfModule>

       Header set Cache-Control "max-age=84600, public"
        <FilesMatch "\.(iso|mp3|mp4|m4a|mov)$">
         Header set Cache-Control "max-age=42300, private"
        </FilesMatch>
</VirtualHost>

Please note that I didn't include comments or like log-settings here

When I download anything using the web browser the apache server's using more and more ram (and uses whole 12.3GB swap even if I don't download a 12.3GB file) and the download speed's decreasing. The same happens when using Nextcloud (desktop/phone) apps

The original download speed was 10MB/s

I've enabled following apache mods:

access_compat alias auth_basic authn_core authn_file authz_core authz_host authz_user autoindex deflate dir env expires fcgid filter headers http2 mime mpm_event negotiation pagespeed proxy proxy_balancer proxy_fcgi proxy_http reqtimeout rewrite setenvif slotmem_shm socache_shmcb ssl status unique_id

I'm using a maria database (recently switched but didn't do much)

Nextcloud's using redis-server as a php cache and I've enabled opcache with the nextcloud's recommended settings. I disabled that already to test but that didn't solve the problem.

Accessibility
1.8.0
Vorgestellt
Activity
2.15.0
Vorgestellt
Brute-force settings
2.2.0
Vorgestellt
Calendar
2.3.4
Vorgestellt
Camera RAW Previews
0.7.12
Checksum
1.1.2
Circles
22.1.1
Vorgestellt
Collaborative tags
1.12.0
Vorgestellt
Comments
1.12.0
Vorgestellt
Contacts
4.0.3
Vorgestellt
Contacts Interaction
1.3.0
Vorgestellt
Dashboard
7.2.0
Vorgestellt
Default encryption module
2.10.0
Vorgestellt
Deleted files
1.12.0
Vorgestellt
End-to-End Encryption
1.7.1
Vorgestellt
Federation
1.12.0
Vorgestellt
File sharing
1.14.0
Vorgestellt
 Group folders
10.0.0
Vorgestellt
Log Reader
2.7.0
Vorgestellt
Mail
1.10.5
Vorgestellt
Maps
0.1.9
Vorgestellt
Metadata
0.14.0
Monitoring
1.12.0
Vorgestellt
Nextcloud announcements
1.11.0
Vorgestellt
Notes
4.1.1
Vorgestellt
Notifications
2.10.1
Vorgestellt
Password policy
1.12.0
Vorgestellt
PDF viewer
2.3.0
Vorgestellt
Photos
1.4.0
Vorgestellt
Privacy
1.6.0
Vorgestellt
Quota warning
1.11.0
Recognize
1.6.10
Registration
1.3.0
Right click
1.1.0
Vorgestellt 
 Share by mail
1.12.0
Vorgestellt
Support
1.5.0
Vorgestellt
Talk
12.1.2
Vorgestellt
Text
3.3.0
Vorgestellt
Theming
1.13.0
Vorgestellt
Two-Factor Email
2.1.1
Update notification
1.12.0
Vorgestellt
Usage survey
1.10.0
Vorgestellt
User status
1.2.0
Vorgestellt
Versions
1.15.0
Vorgestellt
Video converter
1.0.3
Video player
1.11.0
Vorgestellt
Weather status
1.2.0
Vorgestellt

(vorgestellt means recommended)

Only apache's using that much RAM - everything else's fine Also the things don't get removed from the RAM(&swap) until I reload/restart the apache server

When downloading from the browser the download just stops at 383MB - and the browser doesn't say how large the file is to download (Onedrive has that issue too with less than 1MB/s download speed on larger folders, it's like: unknown time remaining - how-much-already-downloaded - download speed)

What did I configure wrong? Or is it something else?

My server's an apple macbook air 4GB RAM 2GHz - 2.2GHz(boost) Intel multithreaded 2 core cpu, xanmod kernel If there's something missing, notify me One thing: earlyoom kills the apache server if there's no free RAM Another thing: I'm on Ubuntu21.04 without x-server and 400MB RAM usage if there's nothing going on

When I upload a file to https://cloud.privacyy.ch I've to set File chunking to 1.5MB I also had a problem with file downloading but that was redis. I disabled redis cache..

Still, if I want to upload any file without chunking it would've 10MB/s and then goes down immediately to some KB/s like 700 and then slowly to 100KB/s and then 10KB/s. At the end you just stop it.

During uploading there's a time when the server has 2 apache processes using 100% of one core. Also later on the apache processes have to wait more and only use 10% but the hdd is loading. There's no special change on upload speed - it just slows down and once the HDD goes well again the apache processes use 100% again. My cpu has 2GHZ and 2 cores (but is multithreaded).

An update of nextcloud didn't help

feel free to ask - I just don't know what logs I've to put in (there's nothing in apache error log)

ah, you can log in to the server by the user: test and the password: lester12345678 if you want to take a look at that special error

btw: max execution time isn't the problem, max post size isn't the problem and max upload size isn't the problem