Check out Atomic Chess, our featured variant for November, 2024.


[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Earliest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Comments/Ratings for a Single Item

Earlier Reverse Order Later
We're back[Subject Thread] [Add Response]
🕸Fergus Duniho wrote on Sat, Apr 15, 2017 07:22 PM UTC:

Earlier today, my hosting account was suspended for using too many resources, and I had to switch to a VPS account. I might still have to fix some things before the site is in working order. This is the first time this site has come back since this morning, and I haven't checked things over yet.


🕸Fergus Duniho wrote on Sun, Apr 16, 2017 04:16 PM UTC:

I'm not sure a lot is going to get done on Easter. The play subdomain is still not working, my webmaster email address is unavailable, I have not set up ftp, and I have to learn how to use webmin/virtualmin instead of cpanel. On the bright side, I have a command shell now, and this means I may be able to install Jocly. But that's for the future. The first thing is to get everything back in running order.


Glenn Nicholls wrote on Sun, Apr 16, 2017 05:49 PM UTC:

I can't seem to do anything on this site, even sign in.


🕸Fergus Duniho wrote on Mon, Apr 17, 2017 12:16 AM UTC:

Okay, I guess I'm posting by entering a password with my post.


🕸Fergus Duniho wrote on Mon, Apr 17, 2017 04:39 PM UTC:

It looks like play.chessvariants.com is currently pointing to the old server instead of the new one. In fixing this, it is possible that some game logs will lose their latest moves. Hopefully, that won't happen. This is a forewarning just in case it does.


Omnia Nihilo wrote on Mon, Apr 17, 2017 08:15 PM UTC:

Omnia Nihilo wrote on Mon, Apr 17, 2017 08:17 PM UTC:

I can't sign in using the front page login screen. It keeps giving an http 500 server error. 


JT K wrote on Tue, Apr 18, 2017 01:42 AM UTC:

I'm not seeing some variant pages anymore on the site.  Is that part of the error that's been discussed?


🕸Fergus Duniho wrote on Tue, Apr 18, 2017 02:32 AM UTC:

I made some progress today.

  • Logging in now works.
  • Some semantic URLs are now working, though not all are yet.
  • I have ftp access, though I haven't yet set up an ftp account for only this site.
  • Headers and footers are working.
  • The play subdomain is now pointing to the new VPS site.
  • Game Courier seems to be working for most games, though circular boards and maybe others will need fixing.

It's time to stop working on this for tonight*, but if you notice anything else that is broken, don't hesitate to mention it here.

*This message is dated according to the GMT time, but I am in America, where it is still April 17.


🕸Fergus Duniho wrote on Tue, Apr 18, 2017 02:33 AM UTC:

I'm not seeing some variant pages anymore on the site.  Is that part of the error that's been discussed?

Very probably. Some pages have semantic URLs, and these are not all functioning yet.


V. Reinhart wrote on Tue, Apr 18, 2017 07:14 PM UTC:

Thanks Fergus for all your hard work and getting it all running well again! Everything seems fine right now (from where I am and at this point in time).
I know you probably want a rest, or some diversion now. As soon as everything is stable, will you please make sure to remember to take some action or leave me some comment about this:

A submission to CVP


Thanks for all your hard hard work. You are keeping the great tradition of variant chess and fairy chess pieces alive and strong!:)


🕸Fergus Duniho wrote on Wed, Apr 19, 2017 02:12 AM UTC:

Today I went through files in index, indexmaint, and play/pbm, changing hardcoded paths to use the new path with either a variable or a constant. I started assigning the path value to a variable called $ROOT, but I later started defining it as a constant called CVP_ROOT. I'll go back later and rename instances of $ROOT as CVP_ROOT, but it's not critical. I also introduced some other constants in Game Courier with the aim of making its code more portable in case the site ever has to move again. I also changed instances of $_ENV to $_SERVER or getenv(), since $_ENV is not used as much on this server, and getenv() seems to return the $_SERVER value if there isn't a $_ENV value.

Although I corrected a game in which my opponent lost on time, I immediately lost on time after he moved, which is weird. I'll have to look into that, but I'm not sure what happened.

I tried using indexmaint/edititem.php, but it wouldn't work. Looking at the timestamp, I have not altered this file since last year, and it has no incorrect paths or instances of $_ENV that need correcting. But it does use mysql functions instead of PDO methods. So I'll have to look into whether mysql functions are still working. Although I have been working on converting from mysql to PDO since early last year, there is still a lot to do on that front. So I'm hoping the problem is something else.


🕸Fergus Duniho wrote on Wed, Apr 19, 2017 02:23 AM UTC:

Some more things I did today were to get member-submitted pages to display, to get the script-generated diagrams to display properly, and to fix some of Game Courier's rendering methods. The graphic image method of drawing circular boards has a mysterious glitch in it that makes the board look weird, but at least it is functional.


🕸Fergus Duniho wrote on Wed, Apr 19, 2017 01:02 PM UTC:

Since quickedititem.php works, and it also uses mysql functions, that is not the reason edititem.php is not working. Also, phpinfo indicates that mysql is supported.


🕸Fergus Duniho wrote on Wed, Apr 19, 2017 01:28 PM UTC:

I traced the problem with edititem.php to a couple identical SELECT requests. It is a very simple request that should not be causing any problems. Maybe rewriting the whole script to use PDO will fix it.


Greg Strong wrote on Wed, Apr 19, 2017 03:50 PM UTC:

If you'd like to email me the code, I'd be happy to take a look - see if anything jumps out.


JT K wrote on Wed, Apr 19, 2017 04:30 PM UTC:

Thank you administrators for getting the site back up and running so quick!  Much appreciated.


🕸Fergus Duniho wrote on Wed, Apr 19, 2017 06:09 PM UTC:

Greg,

I just sent you and other active editors new information on accessing the site. You should now be able to use the File Manager or ftp to take a look at it.


🕸Fergus Duniho wrote on Wed, Apr 19, 2017 08:51 PM UTC:

I tested the SELECT statement in a test script, mysql-test.php, and it worked. I beautified the code to make it easier to read, but I'm still not sure what the problem is. The relevant code uses the query "SELECT * FROM Person WHERE 1 ORDER BY LastName". However, the problem may be code that precedes it.


🕸Fergus Duniho wrote on Thu, Apr 20, 2017 01:19 AM UTC:

Greg,

I sent a test email to my gmail account, and it landed up in my spam folder. Since you're using gmail, check there for the email I sent you.


🕸Fergus Duniho wrote on Thu, Apr 20, 2017 02:00 AM UTC:

I fixed edititem.php. The problem wasn't with the mysql query but with what it did in a while loop that cycled through the results of the query. When the query was commented out, it never went through the while loop. Within the while loop were some lines of the form

$string = $string . "bla bla bla" . "more stuff" . "etc";

The problem went away when I changed these to use something of the form

$string .=  "bla bla bla" . "more stuff" . "etc";


Greg Strong wrote on Thu, Apr 20, 2017 02:33 AM UTC:

Got it.  I didn't see it before.  I had checked spam, but the message didn't come from the email address I was expecting, so it didn't come up in my search.  I've verified that I can get it now.  Thanks!


🕸Fergus Duniho wrote on Fri, Apr 21, 2017 09:35 PM UTC:

What I've done today is fix scripts in the login directory, including the scripts for changing email and for changing a forgotten password. I also fixed the logout script to return to the page the user was on, and I fixed the main login page to accurately know when someone is logged in.


23 comments displayed

Earlier Reverse Order Later

Permalink to the exact comments currently displayed.