Uptade 3.5.5 to latest

[MilorES]

Do I have to do something special?

I updated the docker and it stopped working, I failed to connect to the BDD, a rollback, that is, going back to the previous version, fixed it.

[data-tangles]

Same for me.

I'm using Kubernetes instead for the image and I thought I was going mad when I could'nt get the container to connect to my MariaDB instance.

Specifying the 3.5.5 image from 2 months ago works

[miguelwill]

Hi,

I'm having the same problem. With version 3.5.5, everything works fine, but with the current "latest" version, it gets stuck in a loop showing the message that it can't connect to the database service.

At that point, I can access the container (also via Kubernetes), and it is able to resolve and connect to the database. Even using the MySQL CLI client, I can connect to the provided data, but it seems that the connection check script is not reading them correctly.

Do you have any idea why this could be?

Thanks,

[tiredofit]

There is a 3.5.6 and a 3.5.7 but they are just version bumps to OSTicket. I don't believe I have triggered any builds as of late to overwrite latest, so I'm not sure.

You could inspect the bad image and get me some details from the labels which would be useful.

Otherwise you could go into the bad image and type head -v -n1 /assets/.changelogs/* and get me the output as well..

Maybe Github went on the fritz..

[miguelwill]

hi i run the image with the command "docker run --rm -it tiredofit/osticket:latest /bin/bash" , and show this:

Image:  tiredofit/osticket | Version  3.5.5 Type 'image_changelog' for details
some logs...

[tiredofit/osticket:3.5.5 02:05:26 /assets/.changelogs] $ head -v -n1 /assets/.changelogs/*
==> /assets/.changelogs/tiredofit_docker-debian.md <==
## 7.8.6 2023-06-27 <dave at tiredofit dot ca>

==> /assets/.changelogs/tiredofit_docker-nginx-php-fpm.md <==
## 7.6.10 2023-07-06 <dave at tiredofit dot ca>

==> /assets/.changelogs/tiredofit_docker-nginx.md <==
## 6.3.4 2023-07-15 <dave at tiredofit dot ca>

==> /assets/.changelogs/tiredofit_docker-osticket.md <==
## 3.5.5 2023-06-05 <dekat@github>

[tiredofit/osticket:3.5.5 02:04:40 /assets/.changelogs] $ ls -lh
total 80K
-rw-r--r-- 1 root root  33K Jul 22 01:36 tiredofit_docker-debian.md
-rw-r--r-- 1 root root  20K Jul 22 09:12 tiredofit_docker-nginx-php-fpm.md
-rw-r--r-- 1 root root  20K Jul 22 06:23 tiredofit_docker-nginx.md
-rw-r--r-- 1 root root 3.5K Jul 24 17:13 tiredofit_docker-osticket.md

[miguelwill]

testing in image 3.5.7 , show error

2023-07-27.22:36:45 [WARN] ** [osticket] MySQL/MariaDB Server 'db-mariadb' is not accessible, retrying.. (5 seconds so far)
2023-07-27.22:36:50 [WARN] ** [osticket] MySQL/MariaDB Server 'db-mariadb' is not accessible, retrying.. (10 seconds so far)
2023-07-27.22:36:55 [WARN] ** [osticket] MySQL/MariaDB Server 'db-mariadb' is not accessible, retrying.. (15 seconds so far)
2023-07-27.22:37:00 [WARN] ** [osticket] MySQL/MariaDB Server 'db-mariadb' is not accessible, retrying.. (20 seconds so far)
2023-07-27.22:37:05 [WARN] ** [osticket] MySQL/MariaDB Server 'db-mariadb' is not accessible, retrying.. (25 seconds so far)
2023-07-27.22:37:10 [WARN] ** [osticket] MySQL/MariaDB Server 'db-mariadb' is not accessible, retrying.. (30 seconds so far)

with image 3.5.5, work fine

** [osticket] Configuring mail settings
** [osticket] DB - Connecting to database mysql://osticket@db-mariadb/osticket
** [osticket] Database already installed
** [osticket] DB - Using installation secret from INSTALL_SECRET environmental variable
** [osticket] DB - Updating configuration file
2023-07-27.22:45:13 [NOTICE] ** [osticket] Applying configuration file security
2023-07-27.22:45:35 [STARTING] ** [nginx] [1] Starting nginx 1.25.0
2023-07-27.22:45:35 [STARTING] ** [php-fpm] [1] Starting php-fpm 8.1.18

[tiredofit]

Thanks. Can you go inside the container again and give me the command of mysql -u$DB_USER -p$DB_PASS -h$DB_HOST $DB_NAME

[PaddyPat]

in app container: bash: mysql: command not found

in db container:

/ # mysql -u$DB_USER -p$DB_PASS -h$DB_HOST $DB_NAME ERROR 2005 (HY000): Unknown server host 'osticket' (-2)

` docker log: osticket-app | 2023-07-28.11:39:12 [NOTICE] [nginx] Detected Legacy templates in /etc/nginx/sites.avaialable. Please change '' to {{NGINX_WEBROOT}} osticket-app | 2023-07-28.11:39:12 [NOTICE] [nginx] Detected Legacy templates in /etc/nginx/sites.avaialable. Please change '' to {{NGINX_LISTEN_PORT}} osticket-app | 2023-07-28.11:39:15 [ERROR] [php-fpm] Requested enabling openssl however it doesn't exist! osticket-app | 2023-07-28.11:39:16 [ERROR] [php-fpm] Requested enabling session however it doesn't exist! osticket-app | 2023-07-28.11:39:22 [WARN] [osticket] MySQL/MariaDB Server 'osticket-db' is not accessible, retrying.. (5 seconds so far) osticket-app | 2023-07-28.11:39:27 [WARN] [osticket] MySQL/MariaDB Server 'osticket-db' is not accessible, retrying.. (10 seconds so far) osticket-

`

in old container: app: mysql -u$DB_USER -p$DB_PASS -h$DB_HOST $DB_NAME works and mysql is connected

` DB_NAMEfit/osticket:3.5.5 11:44:49 /] $ [tiredofit/osticket:3.5.5 11:44:49 /] $ mysql -u$DB_USER -p$DB_PASS -h$DB_HOST $D Reading table information for completion of table and column names You can turn off this feature to get a quicker startup with -A

Welcome to the MariaDB monitor. Commands end with ; or \g. Your MariaDB connection id is 6 Server version: 10.11.2-MariaDB Source distribution `

mysql is missing in new app-container?

[tiredofit]

OK I'm now seeing the problem. So, we actually use MariaDB inside the container instead of Mysql and foryears (since inception) the old mysql command worked for cross compatibility. Now it doesn't. I'm going to fix it in one of my base images as opposed to this downstream one by adding a few "mysql" aliases. This will take a few hours.

[tiredofit]

3.6.0 is building and it should resolve the issues with the mysql command being deprecarted from the MariaDB client. Let me know if this works for you.

[miguelwill]

hi

I just tested the new 3.6.0 image and it works correctly.

thanks

[PaddyPat]

Is 3.6.0 osticket 1.74? Or did I deploy wrong? Thought it should be 1.8.0? Thanks four your work!

[tiredofit]

3.6.0 should be 1.18

[tiredofit]

As mentioned in another issue - Github has been stuck for the past little bit and it seems it's pushing false releases due to stuck PRs in the system which my upstream git server doesn't know about. Github tries to build the state at when the PR was submitted. This fix now will rebuild 3.5.6 3.5.7 and 3.6.0 properly.

[PaddyPat]

Docker image 3.6.1

[40] Starting nginx 1.25.1 osticket-app | nginx: [emerg] directive "listen" is not terminated by ";" in /etc/nginx/sites.enabled/osticket.conf:2 osticket-app | 2023-07-30.06:14:06 [STARTING] [nginx] [41] Starting nginx 1.25.1 osticket-app | nginx: [emerg] directive "listen" is not terminated by ";" in /etc/nginx/sites.enabled/osticket.conf:2 osticket-app | 2023-07-30.06:14:08 [STARTING] [nginx] [42] Starting nginx 1.25.1 osticket-app | nginx: [emerg] directive "listen" is not terminated by ";" in /etc/nginx/sites.enabled/osticket.conf:2 osticket-app | 2023-07-30.06:14:09 [STARTING] ** [nginx] [43] Starting nginx 1.25.1 osticket-app | nginx: [emerg] directive "listen" is not terminated by ";" in /etc/nginx/sites.enabled/osticket.conf:2 ^Ccanceled

[MilorES]

What would be the procedure?

• For the service
• TAR of all files and folder
• Modify the Docker-Compose
• Update So you could easily go back?

Note: I want to not have to start over or recover copy! :D