This section is only for discussing issues pertaining to the PerlMonks web site itself.
For example, asking about how things work, or offering ideas on how the site could be made better.
See this recent post for an example: converting binary to decimal
All voting buttons for all nodes on the page are placed way off the screen because of the long lines in the initial node.
It seems the Function Nodelet links are broken: Instead of leading to substr, they lead to ?node=doc://substr which doesn't work (and even logs me out). First reported by adamsj in Chatterbox.
Sometimes it responds immediately, but mostly it takes many seconds (into the 10s sometimes) to load the main site, then again to load "Recent Threads".
This is going to push people away, including long time users of the site.
Is this a bandwidth issue? A DB issue? An issue with an overwhelming number of bots swamping the site?
A discussion needs to be had about what can be done to keep this very valuable resource usable.
If you are browing Perlmonks using Google Chrome and are not logged in, your browser needs to send the sec-ch-ua header, otherwise your anonymous access to the site is blocked.
We are constantly being hit by crawlers claiming to be Google Chrome, so this is a first step in the rat race.
Several times today (11:24, 18:11, 20:43 Europe/Prague) I got "Unable to connect" from firefox when trying to connect to RAT. After several minutes, the page was back up. Is it a new problem, or one of the old problems reappearing?
I noticed this new behaviour today -- I logged in, posted a Meditation, then used the same tab to go somewhere else. Half an hour later, I decided to come back to the site .. and I had to log in again.
So, just now, I visited the site, logged in, then opened a new tab (in the same browser window), and once again I was prompted to log in again. That's odd -- has anyone else noticed this behaviour? I'm using Firefox 136.0 (64-bit) on Ubuntu.
For a long time, I had a link in my .sig going to Groklaw. I heard that as of December 2024, this link is dead. Still, thanks to PJ for all your work, we owe you so much. RIP Groklaw -- 2003 to 2013.
This is because the CGI variables are emitted in a random order - I expect there's a `keys` call.
I believe it would be better to sort those, so that semantically identical URLs with CGI args will always have the same canonical "spelling", so one can most easily see at a glance whether one has previously visited them.
Over the past few days I see 20 - 40 second page load times for any page I've tried to access through https://perlmonks.org. For a typical page fetch (Perl Monks Discussion in this case) Firefox's Developer Tools reports:
Request Timing /?node=Perl Monks Discussion
Blocked: 8.22 s
DNS Resolution: 0 ms
Connecting: 203 ms
TLS Setup: 8.10 s
Sending: 0 ms
Waiting: 13.93 s
Receiving: 405 ms
Request Timing /?node=list replies toggle javascript
Blocked: 6.77 s
DNS Resolution: 3 ms
Connecting: 204 ms
TLS Setup: 6.56 s
Sending: 0 ms
Waiting: 394 ms
Receiving: 0 ms
Finish: 36.20s (The two fetches above are somewhat overlapped. Other f
+etches are ignored.)
When I go to Seekers of Perl Wisdom (https://www.perlmonks.org/?node_id=479 or https://www.perlmonks.com/?node_id=479), and scroll down to the bottom looking for the box into which I can initiate a new thread, I find that said box is no longer present.
Brilliant though this is, could someone please fix it.
I take it that many outages here are caused by misbehaving bots effectively causing a denial of service attack. Especially those bots ignoring no-follow settings.
Now often nodes here are linked not by ID but by name [name]
If the exact name is missing as node title, they cause a supersearch for all node titles containing the words (in any order).
These searches are very slow and I suppose expensive, since the results aren't cached.
Nodelets, which are constantly shown, contain many such links, especially behind ? marks, which until recently were missing real targets.
Now the question:
Could it be that such named-links turning into searches contribute to outages in the monastery?
I just realized that there 3 different PerlMonks URLS. I am logged in at PerlMonks.com and I was surprised when I clicked on a link that took me to Perlmonks.net/somewhere that I wasn't logged in. So, my question is is there some benefit to having three different sites other than just reserving the name to make sure no one else occupies another one of the main addresses? And why aren't all of the different URLS automatically forward you to PerlMonks.com?
