For a side project, I just needed to download the favicons of brands to visually augment a web portal :)
Historically, that icon was named favicon.ico, and stored in the root directory of the website.
However, nowadays the icon is typically called shortcut icon, and there are tons of options on how to get it into the browers’ tab pane…
Very rarely it’s still named favicon.ico.
It’s often not even an ICO file, but a PNG image or an SVG graph.
And developers often refer to it from within a webpage’s HTML code using a <link ...> tag.
However, it’s pretty convinient to create a link to a brand as it properly resembles the brand’s official log!
Of course, downloading the remote web page, parsing the HTML code, and selecting the correct short cut icon (if any, otherwise falling back to $domain/favicon.ico including error handling etc) would be pretty expensive and error-prone.
In such cases it’s always good to outsource the job to someone who’s doing that anyway for their business..
Thus, the GET parameter domain carries the domain of the site of interest (here it’s twitter.com).
Pretty straight forward, isn’t it?
As a bonus, you’ll get a small PHP function to download the icon and store it on your disk:
This will retrieve the favicon for $url, store it in CACHE_DIR, and return the path to the stored file (the file name being the sha1 hash of the domain).
Just make sure you defined CACHE_DIR and enjoy your icons :)
So I heard you don’t like Google?
There is at least one alternative: https://api.faviconkit.com/twitter.com/144.
There is also PHP tool for that, if you want to self-host such a tool: https://github.com/ao/favicons.
As I’m working on multiple machines (two desks at work, one desk at home, laptop, …) I’ve always been looking for a way to sync my browsers…
Of course, I knew about Firefox’ sync, but I obviously don’t want to store my private browsing data in Mozilla’s cloud!
Every once in a while I stumbled upon articles and posts suggesting to run a private syncserver.
However, every time when looking into that project it left an uncomfortable impression:
(i) you need to manually compile some 3rd party software,
(ii) the whole thing seems very complex/unclean, as it requires an account server and a sync server and may work with Mozilla’s account server (but how?),
and (iii) the sync project was once already abandoned (Firefox Weave was discontinued because too complex and unreliable)…
Therefore, I never dared to give it a try.
Today, when I’ve again been frustrated with that fragmented situation, I saw that Mozilla’s syncserver sources contain a Dockerfile!
It probably has been there for ages, but I never recognised it..
Even if that project may be a mess, in a container environment it’s pretty easy to give it a try (and clean it, if unsatisfied)!
That changes everything! :P
Running your own syncserver using Docker is pretty straight forward.
This how-to is based on the project’s readme at GitHub:mozilla-services/syncserver, but I’m using docker-compose and I deployed the service behind an Nginx proxy.
You can of course skip the proxy settings and have it run locally or something.
Docker-Compose makes it easier to assemble and handle multiple containers in a medium complex environment.
My compose config looks like this:
This snippet encodes for a container named firefox-sync, which is based on the image syncserver:latest.
It mounts the host’s directory /path/to/mozilla-sync/share into the container as /syncshare (I’d like to store my stuff outside of the container).
In addition it declares some environment:
SYNCSERVER_PUBLIC_URL tells the service the actual URL to your instance.
SYNCSERVER_SECRET should be complicated as it is used to generate internal certificates and stuff.
SYNCSERVER_SQLURI tell the service which database to use. I point it to the directory (/syncshare) that was mounted into the container, so it will actually store the database on the host.
SYNCSERVER_BATCH_UPLOAD_ENABLED is, if I understand correctly, an option to allow for uploading everything immediately…?
SYNCSERVER_FORCE_WSGI_ENVIRON must be set to true, if SYNCSERVER_PUBLIC_URL doesn’t match the actual URL seen by the python tool. In my case, I would connect to SYNCSERVER_PUBLIC_URL, which is however the Nginx proxy, which forwards the traffic to the syncserver. However, the syncserver will see a different request (e.g. it’s internally not https anymore) and complain.
The last two variables (VIRTUAL_HOST and VIRTUAL_PORT) just configure the reverse proxy that I’m using.
Feel free to drop these lines if you want to expose the service directly to the network, but then you need to add a port forwarding for that container, such as
which forwards traffic at your machine’s HTTP port (:80, use a different port if you’re already running a web server) to the service’s port in the container (:5000).
If you have a porper Docker-Compose configuration, just run
to start the service.
Et voilà, you should be able to access the service at the configured SYNCSERVER_PUBLIC_URL :)
Configure Firefox to use your Private Sync Server
First make sure you’re signed out in the browser!
That means, about:preferences#sync should not show your identity and instead provide a button to sign in.
Then, open about:config and search for identity.sync.tokenserver.uri.
By default, it will be set to Mozilla’s sync server https://token.services.mozilla.com/1.0/sync/1.5.
Edit that field and point it to your SYNCSERVER_PUBLIC_URL plus /token/1.0/sync/1.5.
Thus, in our example above I’d set it to https://firefox-sync.example.com/token/1.0/sync/1.5.
Now go back to about:preferences#sync and sign in with your Mozilla account.
Yes, correct. You still need an account at Mozilla!
But that is just for authentication…
There is an option to also run a private account server (see Run your own Firefox Accounts Server), but that’s even more complicated.
And as I need a Mozilla account anyway to develop my AddOns, I skipped that additional hassling..
Some days ago, @email@example.com convinced me on Mastodon to give BTRFS a try.
That’s actually been a feature on my list for some time already, and now that I need to switch PCs at work I’m going for it.
However, this post wouldn’t exist if everything went straight forward.. ;-)
I have a 1TBSSD that I want to encrypt.
It should automatically get decrypted and mounted to certain places when I log in.
pam_mount can do that for you, and I’ve already been using that a lot in different scenarios.
However, with BTRFS it’s a bit different.
With any other file systems you would create a partition on the hard drive, which is then LUKS encrypted.
This has the drawback, that you need to decide on the partition’s size beforehand!
With BTRFS you can just encrypt the whole drive and use so-called subvolumes on top of it.
Thus, you’re a bit more flexible by creating and adjusting quotas as required at any point in time (if at all…), but (or and!) the subvolumes are not visible unless the device is decrypted.
Let’s have a look into that and create the scenario.
I assume that the SSD is available as /dev/sdb.
Then we can create an encrypted container using LUKS:
You’re not sure which cipher or key-size to choose?
Just run cryptsetup benchmark to see which settings perform best for you.
My CPU, for example, comes with hardware support for AES, thus the AES ciphers show a significantly higher throughput.
If you’re still feeling uncompfortable with that step, I recommend reading the sophisticated article at the
ArchLinux’ wiki on dm-crypt/Device encryption.
We can now open the encrypted device using
This will create a node in /dev/mapper/mydrive, which represents the decrypted device.
Next, we’ll create a BTRFS on that device:
That’s indeed super fast, isn’t it!?
I also couldn’t believe it.. ;-)
We can now mount the device, for example to /mnt/mountain:
So far, the file system is completely empty.
But as it’s a BTRFS, we can create some subvolumes.
Let’s say, we want to create a volume for our $HOME, and as we’re developing this website, we also want to create a volume called www:
So we have two subvolumes in that file system: home (id 258) and www (id 259).
We could now mount them with
But we want the system to do it automatically for us, as we login.
So unmount everything and close the LUKS container:
PamMount can Decrypt and Mount Automatically
I’m using pam_mount already for ages! It is super convenient.
To get your home automatically decrypted and mounted, you would just need to add the following lines to your /etc/security/pam_mount.conf.xml:
Given this, PAM tries to mount the respective subvolumes of the disk (identified by the UUIDa1b20e2f-049c-...) to /home/user and /var/www as soon as YOU logs in.
Here, I am using UUIDs to identify the disks.
You can still use /dev/sdb (or similar), but there is a chance, that the disks are recognised in a different sequence with the next boot (and /dev/sdb may become /dev/sdc or something…).
Plus, the UUID is invariant to the system – you can put the disk in any other machine and it will have the same UUID.
As said above, with BTRFS you’ll have your partitions (called subvolumes) right in the filesystem – invisible unless decrypted.
So, what is PAM doing?
It discovers the first entry in the pam_mount.conf.xml configuration, which basically says
mount a1b20e2f-049c-... with some extra options to /home/user when YOU logs in
PAM is also smart enough to understand that a1b20e2f-049c-... is a LUKS encrypted device and it decrypts it using your login password.
This will then create a node in /dev/mapper/_dev_sdb, representing the decrypted device.
And eventually, PAM mounts /dev/mapper/_dev_sdb to /home/user.
So far so perfect.
But as soon as PAM discovers the second entry, it tries to do the same!
Again it detects a LUKS device and tries to decrypt that.
But unfortunately, there is already /dev/mapper/_dev_sdb!?
Thus, opening the LUKS drive fails and you’ll find something like that in your /var/log/auth.log:
First it seems annoying that it doesn’t work out of the box, but at least it sounds reasonable that PAM cannot do what you what it to do..
… is quite easy, even though it took me a while to figure things out…
As soon as the first subvolume is mounted (and the device is decrypted and available through /dev/mapper/_dev_sdb), we have direct access to the file system!
Thus, we do not neet to tell PAM to mount /dev/disk/by-uuid/a1b20e2f-049c-..., but we can use /dev/mapper/_dev_sdb.
Or even better, we can use the file system’s UUID now, to become invariant to the sdb-variable.
If you run blkid with the device being decrypted you’ll find an entry like this:
You see, the new node /dev/mapper/_dev_sdb also carries a UUID, actually representing the BTRFS :)
This UUID was by the way also reported by the mkfs.btrfs call above.
What does that mean for our setup?
When we first need a subvolume of an encrypted drive we need to use the UUID of the parent LUKS container.
For every subsequent subvolume we can use the UUID of the internal FS.
Transferred to the above scenario, we’d create a /etc/security/pam_mount.conf.xml like that:
Note the different UUIDs? Even though both mounts origin from the same FS :)
Actually, I wanted to have my home in a raid of two devices, but I don’t know how to tell pam_mount to decrypt two devices to make BTRFS handle the raid..?
The only option seems to use mdadm to create the raid, but then BTRFS just sees a single device and, therefore, cannot do its extra raid magic…
If anyone has an idea on that issue you’ll have my ears :)
I’m running Thunderbird to read emails on my desktops.
And I’m using the Lightning plugin to manage calendars, evens, and tasks.
However, since I updated to Thunderbird 60 some weeks ago, Lightning strangely seems to be broken.
The Add-ons manager still lists Lightning as properly installed, but there the “Events and Tasks” menu is missing, as well as the calendar/tasks tabs and the calendar settings in the preferences.
As I’ve been pretty busy with many other things, I didn’t study the problem - hoping that the bug gets fixed in the meantime - but living without the calendar addon is cumbersome.
And today it became annoying enough to make me investigate this…
I then made sure that the versions of Thunderbird and Lightning are compatible (both are 1:60.0-3~deb9u1 for me):
Eventually, I stumbled upon a thread in the German Debian forums: Thunderbird 60 - Lightning funktioniert nicht.
And they figured out, that it may be caused by missing language packs for Lightning…
Indeed, I do have language packs for Thunderbird installed (de and en-gb), that are not installed for Lightning:
And it turns out, that this was a problem!
Thunderbird apparently wouldn’t run Lightning unless it has all required language packs installed.
After installing the missing language packs (aptitude install lightning-l10n-de lightning-l10n-en-gb), the extension is again fully working in Thunderbird!
All that may be cause by a missing dependency..?
Even though thunderbird recommends lightning, thunderbird-l10n-de (and similiar) do not recommend lightning-l10n-de.
Not exactly sure how, but maybe the dependencies should be remodelled…?
I strongly discourage everyone from buying a ShiftPhone.
The Phone was/is on Android patch level from 2017-03-05 – which is one and a half year ago!
Not to mention that it was running an Android 5.1.1 in 2018…
With soo many bugs and security issues, in my opinion this phone is a danger to the community!
And nobody at Shift seemed to really care…
However, I now have a OnePlus One, which is supported by LineageOS - the successor of CyanogenMod.
So, first action was installing LineageOS.
Immediately followed by installing SU to get root access.
Next, I’d like to have SSH access to the phone.
I did love the native SSH server on my Galaxy S2, which used to run CyanogenMod for 5+ years.
Using the SSH access I was able to integrate it in my backup infrastructure and it was much easier to quickly copy stuff from the phone w/o a cable :)
The setup of the native SSH server on LineageOS seems to be pretty similiar to the CyanogenMod version.
First you need a shell on the phone, e.g. through adb, and become root (su).
Then just follow the following three steps:
Create SSH daemon configuration
You do not need to create a configuration file from scratch, you can use /system/etc/ssh/sshd_config as a template.
Just copy the configuration file to /data/ssh/sshd_config;
First, we need to create /data/.ssh on the phone (note the .!) and give it to the shell user:
Second, we need to store our public SSH key (probably stored in ~/.ssh/id_rsa.pub on your local machine) in /data/.ssh/authorized_keys on the phone.
If that file exists, just append your public key into a new line.
Afterwards, handover the authorized_keys file to the shell user:
Create a start script
Last but not least, we need a script to start the SSH service.
There is again a template available in /system/bin/start-ssh.
Just copy the script to /data/local/userinit.d/:
Finally, we just need to update the location of the sshd_config to /data/ssh/sshd_config in our newly created /data/local/userinit.d/99sshd script (in the template it points to /system/etc/ssh/sshd_config, there are 2 occurences: for running the daemon w/ and w/o debugging).
You can now run /data/local/userinit.d/99sshd and the SSH server should be up and running :)
Earlier versions of Android/CyanogenMod auto-started the scripts stored in /data/local/userinit.d/ right after the boot, but this feature was removed with CM12..
Thus, at the moment it is not that easy to automatically start the SSH server with a reboot of your phone.
But having the SSH daemon running all the time may also be a bad idea, in terms of security and battery…
I’m consuming quite some input from the internet everyday.
A substantial amount of information arrives through podcasts, but much more essential are the 300+ RSS feeds that I’m subscribed to.
I love RSS, it’s one of the best inventions in the world wide web!
However, there are alarming rumors and activities trying to get rid of RSS…
We probably should all get our news filtered by Facebook or something..!?
The importance of RSS, which allows users to keep track of updates on many different websites, seems to get continuously ignored..
And so does the new website of our University, where official RSS feeds aren’t provided anymore :(
Apparently, many people were already asking for RSS feeds of the University’s webpage.
At least that’s what they told me, when I asked…
But the company who built the pages won’t integrate RSS anymore - probably wasn’t listed in the requirements..
And the University wouldn’t touch the expensive website.
“Fortunatelly,” they stayed with Typo3 as the CMS, which we’ve been using as well - before we decided to switch.
And this Typo3 platform can output the page’s content as RSS feed out of the box, you just need to know how! ;-)
And… I’ll tell you: Just append ?type=9818 to the URL.
That’s it! Really. It’s so easy.
Static websites are great and popular, see for example
They are easy to maintain and their performance is invincible.
But… As they are static, they cannot dynamically handle user input, which is an obvious requirement for every search engine.
Outsource the task
Lucky us, there are already other guys doing the search stuff pretty convincingly.
So it’s just plausible to not reinvent the wheel, but instead make use of their services.
There are a number of search engines, e.g.
Yippy, and more (list sorted alphabetically, see also Wikipedia::List of search engines).
They all have pros and cons, but typically it boils down to a trade between coverage, up-to-dateness, monopoly, and privacy.
You probably also have your favourite.
However, it doesn’t really matter.
While this guide focusses on DuckDuckGo, the proposed solution is basically applicable to all search engines.
The idea is, that you add a search form to your website, but do not handle the request yourself and instead redirect to an endpoint of a public search engine.
All the search engines have some way to provide the search phrase encoded in the URL.
Typically, the search phrase is stored in the GET varialble q, for example example.org/?q=something would search for something at example.org.
Thus, your form would redirect to example.org/?q=....
However, that would of course start a search for the given phrase on the whole internet!
Instead, you probably want to restrict the search results to pages from your domain.
Fortunatelly, the search engines typically also provide means to limit search results to a domain, or similar.
In case of DuckDuckGo it is for example the site: operator, see also DuckDuckGo’s syntax.
That is, for my blog I’d prefix the search phrase with site:binfalse.de.
Implementing the workaround is no magic, even though you need to touch your webserver’s configuration.
First thing you need to do is adding a search form to your website.
That form may look like this:
As you see, the form just consists of a text field and a submit-button.
The data will be submitted to /search on your website.
Sure, /search doesn’t exist on your website (if it exists you need to use a different endpoint), but we’ll configure your web server to do the remaining work.
The web server needs to do two things: (1) it needs to prefix the phrase with site:your.domain and (2) it needs to redirect the user to the search engine of your choice.
Depending on the web server you’re using the configuration of course differs.
My Nginx configuration, for example, looks like this:
So it sends the user to duckduckgo.com, with the query string site:binfalse.de concatenated to the submitted search phrase ($arg_q = the q variable of the original GET request).
If you’re running an Apache web server, you probably know how to achieve the same over there.
Otherwise it’s a good opportunity to look again into the manual ;-)
Furthermore, the results pages of DuckDuckGo can be customised to look more closely like your site.
You just need to send a few more URL parameters with the query, such as kj for the header color or k7 for the background color.
The full list of available configuration options are available from DuckDuckGo settings via URL parameters.
In conclusion, if you use my search form to search for docker, you’ll be guided to https://binfalse.de/search?q=docker.
The Nginx delivering my website will then redirect you to https://duckduckgo.com/?q=site%3Abinfalse.de+docker, try it yourself:
search for docker!
This of course also works for dynamic websites with WordPress, Contao or similar…
I’ve been using it for my calendars and adressbooks already for more than 4 years now.
However, I initially installed it as plain PHP application with a MySQL database.
The developers also announced quite early, that they are working on a Docker image, but there is nothing useful as of mid 2018.
So far they just provide a quite inconvenient how-to and a list of issues that apparently prevent them from providing a proper Docker image.
Thus, I just dockerised the application myself :)
Using the image is fairly simple.
Basically, you only need to mount some persistent space to /var/www/Specific
docker run -it --rm -p 80:80 -v /path/to/persistent:/var/www/Specific binfalse/baikal
Please make sure that the directory /path/to/persistent has proper permissions.
In the container an Apache2 is serving the contents, so make sure the user www-data (UID33) is allowed to rwx that directory.
To support encrypted connections you would need to mount the certificates as well as a modified Apache configuration into the container.
However, I recommend to run it behind a reverse proxy, such as binfalse/nginx-proxy, and let the proxy handle all SSL connections (as for all other containers).
This way, you just need one proper SSL configuration.
The default SQLite database is perfect for a first test, but is slow and just allows for a limited amount of SQL variables.
If you for example have more than 999 contacts, the first sync of a clean WebDAV device will result in an exception such as:
PDOException: SQLSTATE[HY000]: General error: 1 too many SQL variables
Thus, for production you may want to switch to a proper database, such as MariaDB.
Lucky you, the Docker image supports MySQL! ;-)
To reproducibly assemble both containers, I recommend Docker-Compose.
Here is a sample config with two containers baikal and baikal-db:
This assumes, that your Baikal configuration can be found in /srv/baikal/config.
The database will be stored in /srv/baikal/database.
Also note the database credentials for configuring Baikal.
If you’re not running a reverse proxy in front of the application, you also need to add some port forwarding for the baikal container:
In a typical Docker environment you’ll have plenty of containers (probably in multiple networks?) on the same machine.
Let’s assume, you need to debug some problems of a container, eg. because it doesn’t send mails anymore..
What would you do? Correct, you’d go and check the logs.
By default, Docker logs the messages of every container into a json file.
On a Debian-based system you’ll probably find the file at /var/lib/docker/containers/CONTAINERID/CONTAINERID-json.log.
However, to properly look into the logs you would use Docker’s logs tool.
This will print the logs, just as you would expect cat to dump the logs in /var/log.
docker-logs can also filter for time spans using --since and --until, and it is able to emulate a tail -f with --follow.
However, the logs are only available for exsiting containers.
That means, if you recreate the application (i.e. you recreate the container), you’ll typically loose the log history…
If your workflow includes the --rm, you will immediately trash the log of a container when it’s stopped.
Fortunatelly, Docker provides other logging drivers, to e.g. log to AWS, fluentd, GPC, and to good old syslog! :)
Here I’ll show how to use the host’s syslog to manage the logs of your containers.
Log to Syslog
Telling Docker to log to the host’s syslog is really easy.
You just need to use the built-in syslog driver:
Voilà, the container will log to the syslog and you’ll probably find the messages in /var/log/syslog.
Here is an example of an Nginx, that I just started to serve my blog on my laptop:
By default, the syslog driver uses the container’s ID as the syslog tag (here it is af6dcace59a9),
but you can further configure the logging driver and, for example, set a proper syslog tag:
This way, it is easier to distinguish between messages from different containers and to track the logs of an application even if the container gets recreated:
Here, I configured an nxinx that just serves the contents from /srv/web/default.
The interesting part is, however, that the container uses the syslog driver and the syslog tag docker/website.
I always prefix the tag with docker/, to distinguish between log entries of the host machine and entries from Docker containers..
Store Docker logs seperately
The workaround so far will probably substantially spam your /var/log/syslog, which may become very annoying… ;-)
Therefore, I recommend to write Docker’s logs to a seperate file.
If you’re for example using Rsyslog, you may want to add the following configuration:
Just dump the snippet to a new file /etc/rsyslog.d/docker.conf and restart Rsyslog.
This rule tells Rsyslog to write messages that are tagged with docker/* to /var/log/docker, and not to the default syslog file anymore.
Thus, your /var/log/syslog stays clean and it’s easier do monitor the Docker containers.
Disentangle the Container logs
Since version 8.25, Rsyslog can also be used to split the docker logs into individual files based on the tag.
So you can create separate log files, one per container, which is even cleaner!
The idea is to use the tag name of containers to implement the desired directory structure.
That means, I would tag the webserver of a website with docker/website/webserver and the database with docker/website/database.
We can then tell Rsyslog to allow slashes in program names (see the programname section at www.rsyslog.com/doc/master/configuration/properties.html) and create a template target path for Docker log messages, which is based on the programname:
Using that configuration, our website will log to /var/log/docker/website/webserver.log and /var/log/docker/website/database.log.
Neat, isn’t it? :)
Even though all the individual logfiles will be smaller than a combined one, they will still grow in size.
So we should tell logrotate of their existence!
Fortunatelly, this is easy as well.
Just create a new file /etc/logrotate.d/docker containing something like the following:
This will rotate the files ending in *.log in /var/log/docker/ and its subdirectories everyday and keep compressed logs for 7 days. Here I’m using a maximum depth of 3 subdirectories – if you need to create a deeper hierarchy of directories just add another /var/log/docker/*/*/*/*.log etc to the beginning of the file.
In a previous post I explained how to run a Contao website in a Docker infrastructure.
That was a good opening.
However, after running that setup for some time I discovered a few issues…
A central idea of Docker is to install the application in an image and mount persistent files into a running container.
Thus, you can just throw away an instance of the app and start a new one very quickly (e.g. with an updated version of the app).
Unfortunately, using Contao it’s not that straight-forward – at least when using the image decribed earlier.
The browser requests a file cron.txt, which is supposed to contain the timestamp of the last cron run.
If the timestamp is “too” old, the browser will also request a cron.php, which then runs overdue jobs.
If a job was run, the timestamp in cron.txt will be updated, so cron.php won’t be run every time.
Good, but that means the cron.txt will only be written, if a cron job gets executed.
But let’s assume the next job will only be run next week end!?
The last cron-run-time is stored in the database, but the cron.txt won’t exist by default.
That means, even if the cron.php is run, it will know that there is no cron job to execute and, therefore, exit without creating/updating the cron.txt.
Especially when using Docker you will hit such a scenario every time when starting a new container..
Thus, every user creates a 404 error (as there is no cron.txt), which is of course ugly and spams the logs..
A typical Docker infrastructure (at least for me) consists of bunch containers orchestrated in various networks etc..
Usually, you’ll have at least one (reverse) proxy, which distributes HTTP request to the container in charge.
However, I experienced a few issues with my proxy setup:
HTTPS vs HTTP
While the connection between client (user, web browser) and reverse proxy is SSL-encrypted, the proxy and the webserver talk plain HTTP.
As it’s the same machine, there is no big need to waste time on encryption.
But Contao has a problem with that setup.
Even though, the reverse proxy properly sends the HTTP_X_FORWARDED_PROTO, Contao only sees incomming HTTP traffic and uses http://-URLs in all documents…
Even if you ignore the mixed-content issue and/or implement a rewrite of HTTP to HTTPS at the web-server-layer, this will produce twice as much connections as necessary!
The solution is however not that difficult.
Contao does not understand HTTP_X_FORWARDED_PROTO, but it recognises the $_SERVER['HTTPS'] variable.
Thus, to fix that issue you just need to add the following to your system/config/initconfig.php (see also Issue 7542):
In addition, this will generate URLs including the port number (e.g. https://example.com:443/etc), but they are perfectly valid. (Not like https://example.com:80/etc or something that I saw during my tests… ;-)
URL encodings in the Sitemap
The previous fix brought up just another issue: The URL encoding in the sitemap breaks when using the port component (:443)..
Conato uses rawurlencode to encode all URLs before writing them to the sitemap.
However, rawurlencode encodes quite a lot!
Among others, it converts :s to %3A.
Thus, all URLs in my sitemap looked like this: https://example.com%3A443/etc - which is obviously invalid.
A more delicate issue are cache and assets and sitemaps etc.
Contao’s backend comes with convenient buttons to clear/regenerate these files and to create the search index.
Yet, you don’t always want to login to the backend when recreating the Docker container..
Sometime you simply can’t - for example, if the container needs to be recreated over night.
Basically, that is not a big issue.
Assets and cache will be regenerate once they are needed.
But the sitemaps, for instance, will only be generated when interacting with the backend.
Thus, we need a solution to create these files as soon as possible, preferably in the background after a container is created.
Most of the stuff can be done using the Automator tool, but I also have some personal scripts developed by a company, that require other mechanisms and are unfortunately not properly integrated into Contao’s hooks landscape.
And if we need to touch code anyways, we can also generate all assets and rebuild the search index manually (precreating necessary assets will later on speed up things for users…).
To generate all assets (images and scripts etc), we just need to access every single page at the frontend.
This will then trigger Contao to create the assets and cache, and subsequent requests from real-life users will be much faster!
The best hack that I came up with so far looks like the following script, that I uploaded to /files/initialiser.php to Contao instance:
The first 3 lines initialise the Contao environment.
Here I assume that ../system/initialize.php exists (i.e. the script is saved in the files directory).
The next few lines purge existing cache using the Automator tool and subsequently regenerate the cache – just to be clean ;-)
Finally, the script
(i) collects all “searchable pages” using the Backend::findSearchablePages() functionality,
(ii) enriches this set of pages with additional pages that may be hooked-in by plugins etc through $GLOBALS['TL_HOOKS']['getSearchablePages'],
and then (iii) uses cURL to iteratively request each page.
The first part should be reasonably fast, so clients may be willing to wait until the cache stuff is recreated.
Accessing every frontend page, however, may require a significant amount of time!
Especially for larger web pages..
Thus, I embedded everything in the following skeleton, which advises the browser to close the connection before we start the time-consuming tasks:
Here, the browser is told to close the connection after a certain content size arrived.
I buffer the content that I want to transfer using ob_start and ob_end_flush, so I know how big it is (using ob_get_length).
Everything after ob_get_length can safely be ignored by the client, and the connection can be closed.
(You cannot be sure that the browser really closes the connection. I saw curl doing it, but also some versions of Firefox still waiting for the script to finish… Nevertheless, the important content will be transferred quick enough).
In addition, I created some RewriteRules for mod_rewrite to automatically regenerate missing files.
For example, for the sitemaps I added the following to the vhost config (or htaccess):
That means, if for example /share/sitemap.xml not yet exists, the user gets automagically redirected to our initialiser.php script!
In addition, I added some request parameters (?target=sitemap&sitemap=$1), so that the initialiser.phpknows which file was requested.
It can then regenerate everything and immediately output the new content! :)
For example, my snippet to regenerate and serve the sitemap looks similar to this:
Thus, the request to /share/somesitemap.xml will never fail.
If the file does not exist, the client will be redirected to /files/initialiser.php?target=sitemap&sitemap=somesitemap, the file /share/somesitemap.xml will be regenerated, and the new contents will immediately be served.
So the client will eventually get the desired content :)
Please be aware, that this script is easily DOS-able!
Attackers may produce a lot of load by accessing the file.
Thus, I added some simple DOS protection to the beginning of the script, which makes sure the whole script is not run more than once per hour (3600 seconds):
If $dryrun is true, it won’t regenerate cache etc, but still serve the sitemap and other files if requested..
However, if there is also no $_GET['target'] defined, we don’t know what to serve anyway and can die immediately…
You could include the script at the footer of your webpage, e.g. using
This way you would make sure, that every request produces a fully initialised system.
However, this will probably also create unnecessary load every hour…
You could increase the time span in the DOS-protection-hack, but I guess it should be sufficient to run the script only if a missing file is requested.
Earlier requests then need to wait for pending assets etc, but to be honest, that should not be too long (or you have a different problem anyway…).
And if your website provides an RSS feed, you could subscribe to it using your default reader, which will regularly make sure that the RSS feed is generated if missing.. (and thus trigger all the other stuff in our initialiser.php)
– A feed reader as the poorest-man-cron ;-)
As I said earlier, my version of the script contains plenty of personalised stuff.
That’s why I cannot easily share it with you.. :(
However, if you have trouble implementing it yourself just let me know :)