Eigentlich lief es ja seit Jahren gut, dass ich andere DNS-Server als die von #Vodafone eingestellt habe, z.B. auch meinen Server seit einiger Zeit. Nun aber ist es eben zweimal hintereinander - und das war nicht das erste Mal - vorgekommen, dass urplötzlich die #Namensauflösung nicht mehr funktionieren möchte, oder soll.

Ich war gezwungen, dann einmal die Standard-DNS-Server (also von Vodafone) zu nehmen, die Einstellung zu speichern, meine DNS-Server ("Andere DNSv4-Server verwenden") auswählen und schon klappte es wieder.

Liebe anderen #Follower da draußen, kennt ihr das Problem auch? Ich glaube, das bettelt hier nach Kündigung. Was ich nicht kapiere ist, wieso ich dann überhaupt noch in der Router-Software (Fritz! Box) dies abändern darf, wenn es doch eh nur zeitweise klappt?

I have observed here on my Friendica node, that the worker queue gets fuller and fuller over time and I was really wondering why is that so. I checked the queue table and found no real issue with it.

Then I checked throught the PHP code and also saw no logic mistakes (related to this, e.g. in Friendica\Core\Worker class).

Here are some typical lines at the start of my investigation:

2021-02-20T23:24:16Z worker [NOTICE]: Load: 4.34/50 - processes: 409/4/115055 - jpm: 6/4/8 (0:1, 20:0/6, 30:0/6, 40:3/115452) - maximum: 4/5 [] - {
"file":"Worker.php","line":782,"function":"tooMuchWorkers","uid":"343740","process_id":17119}
2021-02-20T23:24:17Z daemon [NOTICE]: Load: 4.34/50 - processes: 409/4/115056 - jpm: 6/4/8 (0:1, 20:0/7, 30:0/6, 40:3/115452) - maximum: 5/5 [] - {
"file":"Worker.php","line":782,"function":"tooMuchWorkers","uid":"163d4b","process_id":17116}
2021-02-20T23:24:18Z worker [NOTICE]: Load: 4.34/50 - processes: 409/4/115056 - jpm: 6/4/8 (0:1, 20:0/7, 30:0/6, 40:3/115452) - maximum: 5/5 [] - {
"file":"Worker.php","line":782,"function":"tooMuchWorkers","uid":"343740","process_id":17119}
2021-02-20T23:24:19Z worker [NOTICE]: Load: 4.36/50 - processes: 409/5/115057 - jpm: 6/4/8 (0:2, 20:0/7, 30:0/7, 40:3/115452) - maximum: 6/5 [] - {
"file":"Worker.php","line":782,"function":"tooMuchWorkers","uid":"3f65a2","process_id":17123}
2021-02-20T23:24:19Z worker [NOTICE]: Load: 4.36/50 - processes: 409/5/115057 - jpm: 6/4/8 (0:2, 20:0/7, 30:0/7, 40:3/115452) - maximum: 6/5 [] - {
"file":"Worker.php","line":782,"function":"tooMuchWorkers","uid":"343740","process_id":17119}
2021-02-20T23:24:20Z daemon [NOTICE]: Load: 4.36/50 - processes: 409/5/115057 - jpm: 6/4/8 (0:1, 20:1/7, 30:0/7, 40:3/115452) - maximum: 4/5 [] - {
"file":"Worker.php","line":782,"function":"tooMuchWorkers","uid":"163d4b","process_id":17116}

As you can see maximum: 4/5 and even maximum: 6/5 is present. So I checked at the code where the 5 is coming from and I found it here:

$queues = DI::config()->get("system", "worker_queues", 10);

$maxqueues = $queues;

So I checked that configuration entry in config table and ... Bingo! There it is: 5. I then disabled debug log again (not to have the hard drive flooded with debug log) and restarted it.

Since then the daemon is working happily on the queue. So as it seems to me, there are indeed more queues (spawned PHP processes) needed to work on a public Friendica instance. And this one here has only a few local users while one (me) is most of the time active.

As you might have noticed, I have been called insecure about my #sexuality while I'm very much secure about (I know what I want and I know what I don't want). From today on, I will block all these idiots, calling me insecure about my sexuality or that I have an "odd opinion" about it.

The block list just got longer:
- Judging me about my feelings, relationship, penis size or whatever they want to frustrate me with (while it doesn't frustrate me at all)
- generalizing men while they don't like it when women are being generalized
- Calling me a homophobe, transphobe (see first reason)
- Calling me a sexist, "enableist" or "schwurbelnd"
- Or talking anything bad about my romantic relationship or "preference" (#sexuality is not a choice, obviously we #straight people have there a choice according to some radical fedizens)
- to many cats/dogs/whatever-animal pictures, quoted memes and other delusional views, huge red flag

Why they do it, because isn't it fun to mess around with complete strangers over the Internet? Yeah, isn't it!

Sorry for my English audience but I decided to make this episode in German language.

Die erste Aufnahme mit Webcam, gerichtet auf mich und leider war das ganze Layout durcheinander. Keine Ahnung, was #OBS dort durcheinander gebracht hat. Ungefaehr 7 GB Aufnahme in voller Qualitaet (kommt so direkt aus OBS heraus) sind leider vergeigt. Das Spiel war die ganze Zeit nach rechts unten in die Ecke verschoben und das moechte ich keinem anbieten.

Diese Episode war mal zur Abwechselung auf deutscher Sprache. Mittlerweile habe ich wieder vergessen, was ich dort alles gemacht habe und meine alten Episoden sehe ich mir nur selten selbst an. Also viel Spass!

Zum Download

PS: Der Web-Player kann offensichtlich nicht Matroska-Videos abspielen. Ich habe daher mal den Download aktiviert, uebertreibt es aber nicht, 10 TB Datenvolumen sind schnell aufgebraucht und dann darf ich zahlen.

[ CA_default ]
dir = /etc/ssl/zulu289 # Where everything is kept
certs = $dir/certs # Where the issued certs are kept
database = $dir/zulu289.index # database index file.
new_certs_dir = $dir/newcerts # default place for new certs.
certificate = $dir/certs/ca.pem # The CA certificate
serial = $dir/zulu289.serial # The current serial number
crl = $dir/crl.pem # The current CRL
crlnumber = $dir/crlnumber # the current crl number must be
private_key = $dir/private/ca/zulu289-ca-key.pem # The private key
default_days = 3650 # how long to certify for
default_md = default
policy = policy_match

[ req ]
default_bits = 8192
default_keyfile = cloud-key.pem
distinguished_name = req_distinguished_name

[ ca ]
default_ca = CA_default

[ req_distinguished_name ]
countryName = DE
countryName_default = DE

stateOrProvinceName = Northrhine-Westphalia
stateOrProvinceName_default = Northrhine-Westphalia

localityName_default = Krefeld

0.organizationName = Roland Häder
0.organizationName_default = Roland Häder

organizationalUnitName = private
organizationalUnitName_default = private

commonName = cloud.mxchange.org
commonName_default = cloud.mxchange.org

emailAddress = webmaster@mxchange.org
emailAddress_default = webmaster@mxchange.org

#unstructuredName = An optional company name

[ server ]
basicConstraints=CA:FALSE
nsCertType = server
nsComment = "OpenSSL Generated Server Certificate"
subjectKeyIdentifier=hash
authorityKeyIdentifier=keyid,issuer:always

[ client ]
basicConstraints=CA:FALSE
nsCertType = client
nsComment = "OpenSSL Generated Client Certificate"
subjectKeyIdentifier=hash
authorityKeyIdentifier=keyid,issuer:always
keyUsage = nonRepudiation, digitalSignature, keyEncipherment

[ policy_match ]
countryName     = match
stateOrProvinceName = match
organizationName    = match
organizationalUnitName  = optional
commonName      = supplied
emailAddress        = optional

I would like to introduce you to my upcoming series of playing #Factorio under Linux. I'm exclusively around with #Linux for now over 20 years and I'm happy to say that it was largely a success - honestly.

So here are my first episodes shared on my #Nextcloud instance for free downloading.

However, please be nice! The total size of the 1080p version (maximum resolution) is already +18 GB large and thats only "one" episode ... So please download with respect to my limited bandwidth of my server: You are not the only one utilizing my server's bandwidth. I have 10TB/month included in my contract and every additional GB must be payed extra and I don't get any money back from these downloads + I donate to the @FSFE every year + I don't ask for a donation, but you can if you wish. Just ask me for methods.

And for later episodes, I may have to cleanup some old episodes as also disk space is not bottom-less here. 2x1 TB in a RAID1 is already almost full:

me@zulu289:~$ df
Filesystem     1K-blocks      Used Available Use% Mounted on
udev             3914372         0   3914372   0% /dev
tmpfs             792044      1600    790444   1% /run
/dev/md2         9712160   7731828   1463932  85% /
tmpfs               5120         0      5120   0% /run/lock
tmpfs            1584080         0   1584080   0% /dev/shm
/dev/sda1          91387     38001     46484  45% /boot
/dev/md3       950374844 559739048 342336464  63% /var
tmpfs            3960204        60   3960144   1% /tmp
me@zulu289:~$

This server (zulu289) holds all (!) I do on my server:
* Tracker for FLOSS: tracker.mxchange.org (primary FlightGear Flight Simulator's GIT repository)
* Debian-multimedia mirror: debian.mxchange.org/
* Databases for all my instances I'm running + full SQL dumps from them, e.g. Nextcloud, this very Friendica instance and GNUSocial are the most demanding databases:

root@zulu289:/var/backups/dumps# ls -l friendica*31* social_mailer-2021-01-31.sql.bz2 owncloud-2021-01-31.sql.bz2 full-backup-2021-01-31.sql 
-rw------- 1 root root 13708573349 Jan 31 07:48 friendica-2021-01-31.sql.bz2
-rw------- 1 root root    50290723 Jan 31 09:33 friendica_dir-2021-01-31.sql.bz2
-rw-r--r-- 1 root root 18307805562 Jan 31 06:42 full-backup-2021-01-31.sql
-rw------- 1 root root    22672102 Jan 31 09:33 owncloud-2021-01-31.sql.bz2
-rw------- 1 root root   118700237 Jan 31 09:38 social_mailer-2021-01-31.sql.bz2
root@zulu289:/var/backups/dumps#

And maybe more I forgot to mention. :)

I have found out that for the default directory at
the timeout of one second is simply to short. I have raised it's minimum value now to 3 seconds.

Plus I replaced '\Friendica\Some\Foo' with \Friendica\Some\Foo::class so your used IDE/PHP editor can find it easier + it must really be there.

Pull-request is out:
github.com/friendica/friendica…