I've even tried using a different computer.

concrete5 creates around 250 tables on install
I don't have an answer to your problem, but maybe you can tell us a bit more ?

Are you trying to install 5.7.4.2 or the release candidate 5.7.5rc1 ?

Are you sure that you have an empty database for concrete5 when you start the install ?

What host are you on ? Does it set limits ?

Do you have specs on your server ?

good luck !

I am getting the same error. It seems as though C5 is trying to install the database tables twice and errors out because all the tables were created the first time. I am having this same problem with both 5.6.3.3 and 5.7.4.2.

I am hosting many concrete5 websites on my own private server and this is the first time I have seen this problem. The mySQL server is on the localhost. Yes, each time I try to install it, I remove all the tables that were created the last time.

Have you heard anything about them fixing this or a possible solution?

I am just now starting this new install and this is the first time I have run across this issue.

I re-installed an already existing c5 site about a month ago and used 5.6.3.3 without this issue.

You are having the same problem both with the 5.6 and the 5.7 branch ?

That would point to a server setup problem...

Yes.

5.6.33 and 5.7.4.2 both do the same thing for me although they create a different number of tables.

You would think, but I have a half a dozen other concrete 5 web sites running on this same server and with the same mySQL service.

I even have a WHMCS site running on this machine.

UPDATE:
It looks like the installer is trying to install the database tables twice because when you select the "back" link, it starts the instalation all over again and that's where the error, "there are already tables in this database", is coming from.

Watching the progress reporting, the installation gets up to the "installing dashboard" and fails at this point. The progress bar is showing not even half way complete.

This is not just a server problem since everything else including other C5 web sites are all good on this server. It's also a Concrete5 thing.

I am running Centos6.6 on a remote private server. MySQL is on the same machine and access is on the localhost. My file permissions are set to owner/group ownership are set correctly. I have tried permissions 755, 775, and 777 so it can't be a file permissions problem.

If anyone has any suggestion as to what could be causing this problem I really would like to hear you.

Thanks

'because when you select the "back" link...'
I'm not sure if I understand you correctly:

You mean the problem is not that concrete5 installs double tables by itself, but it only does this when you reload the install page ?

But there is a problem with the install: it halts halfway on installing the dashboard ?

If so, here are a few links that might help you:
https://www.concrete5.org/community/forums/customizing_c5/installati...
https://www.concrete5.org/community/forums/installation/stuck-at-ins...
https://www.concrete5.org/community/forums/installation/installation...

Be sure to know what you are doing when you change any settings for your server. Especially when you have more sites on it already.
Hope this helps.

Yes, when the install fails, the progress field becomes empty with a light red background and under it there is a link "back". This link takes you back to the point where it tries to install C5 again instead of taking you to a page that details what the failier was. Might I suggest that this be changed in the installer?

Thanks for the links that might help me. I'll go through them know.

Thanks again.

I set my php.ini to max_execution_time = 120 seconds and ran the install again, only this time I have Firebug open in Firefox and monitoring the network activity. The "install_dashboard" process throws a "504 gateway time-out" error after 45 seconds. Is there some preset variable that I can change to make the installer wait for a longer time before timing out? I think this may be caused by my really slow connection through ATT wireless. This is the only link I have to the Internet right now.

here's one that google came up with:
http://www.h3xed.com/web-development/php-and-apache-504-gateway-tim...

I just checked timeouts. PHP's time outs are set for 60 seconds as well as my httpd.conf's time out. The gateway time out is happening at about 45 seconds. I did, however set the memory limit from 128M to 512M but I don't think that this is the problem.

By the way, I checked all logs and there is no errors being logged when it happens.

Maybe someone who knows more about server settings can help here ?

I think the bottom line here is the issue that there is a apache gateway time out of 45 seconds happening on the server that I can't find and also consider that "install_dashboard" shouldn't be taking that long anyway.

what do your settings show in a phpinfo? Is it possible that you are not altering settings in the correct place?

The installer process uses quite a bit of memory - despite how you are allocating it is there enough memory on the VPS/Server with the other stuff on there - you said a number of C5 sites and WHMCS - if you are running WHMCS are you also running Cpanel or are you running WHMCS APIs for other servers?

What are your resources like on the server? When you run the installer what happens to utilisation?

My server has enough memory available for all processes.
This is not a VPS. It is a dedicated server.
The site Memory usage will be dependent on the traffic to these sites and right now, the load is very light and everthing is running slick.
I do not use cPanel. I don't like what Plesk, WHM, and cPanel does to the file system. I use Webmin when needed.
phpinfo does reflect the changes I made. the following is from my php.ini.

;;;;;;;;;;;;;;;;;;;
; Resource Limits ;
;;;;;;;;;;;;;;;;;;;

; Maximum execution time of each script, in seconds
;http://www.php.net/manual/en/info.configuration.php#ini.max-executi...
max_execution_time = 120

; Maximum amount of time each script may spend parsing request data. It's a good
; idea to limit this time on productions servers in order to eliminate unexpectedly
; long running scripts.
; Default Value: -1 (Unlimited)
; Development Value: 60 (60 seconds)
; Production Value: 60 (60 seconds)
;http://www.php.net/manual/en/info.configuration.php#ini.max-input-t...
max_input_time = 60

; Maximum input variable nesting level
;http://www.php.net/manual/en/info.configuration.php#ini.max-input-n...
;max_input_nesting_level = 64

; Maximum amount of memory a script may consume (128MB)
;http://www.php.net/manual/en/ini.core.php#ini.memory-limit...
memory_limit = 512M

Does mysql have enough resources?

What do your error logs show immediately after the install fails the first time?

As I said before, there are no errors thrown in messages, httpd/error_log, or mysqld.log.
mySQL is on the same localhost and there are pleanty of resources to go arround. What I am looking for is what can be causing Apache to be throwing the "504 gateway time-out" error after 45 seconds. I can't find that timming anywhere in my configs. I even just got a reply back from my server provider that thay have no such proxy server on port 80 on their network.

Sorry AndyJ. In my haist, I hadn't noticed the change in helpers. I thought I was communicating with William.

ok fair enough - I'll wind my neck in and get back to spending my time more valuably. Was just thinking that it might not be a hard limit of 45 seconds: that might just happen to be when a certain resource maxes out.. Hope you find whatever the issue is and report back..

Please excuse my ancious response. I appreciate any help I can get but I didn't notice that the thread changed to a new helper and we are covering ground already tread upon.

no worries. I can understand your frustration - is a bit late in the day here and I had not read as thoroughly as I should have.

If you do a traceroute to that particular website does it show anything noteworthy compared to other sites on the server?

Traceroute shows the same path from my desktop to both sleepyvalley.net and countylineatv.rodeo which is the site in question.

If any resources maxed out while php was processing a request, wouldn't php error out with a message?

generally yes. What does your php log show leading up to the timeout error?

I think I know where the problem lies. My connection to the internet is through ATT cellular and my experience with them has been unbelievably bad. Now, in addition to my data flow being randomely interupted and lots of packet loss through their network because of they're being grossly overloaded at times, I can now add a Proxy server on their front end that is throwing gateway timeout errors. It should have registered that 504 is a gateway proxy service error, not an Apache generated error. I'm sure that's where the 45 second limit is comming from. I will call them again tomorrow with yet another complaint.

I live out of town in the middle of the Ozarks. That's why I am locked into using ATT Cellular as my connection for now as the only game not in town. I am in the process of putting up a 60 foot tower with mast on top to be able to connect to my ISP who runs radio links on towers throughout the southwest part of Missouri. They are scheduled to connect me with my tower on the 17th.

Tomorrow, I guess I'll have to go into town and connect with WiFi to complete this installation.

Hope this will work for you. I wonder if the originator of this thread has a similar problem.

Hope this will work for you. I wonder if the originator of this thread has a similar problem.

I went into town yesterday and while having lunch at the B&D dinner, I started this install process again on their WiFi and it completed without any issues. While this shows that this is truly an action of a proxy server on AT&T Cellular's network timing out after 45 seconds with a 504 error, I can't help thinking that C5 also has a hand in this problem by not writing their CGIs to take into account of any possible time constraints placed on the client's Internet connection.

Suggestion for future installer updates; On the client side, trap any errors that occure when the http connection closes and then report the errors with suggestions in the empty red error bar being displayed.

I'm glad it worked out for you. Thanks for reporting this back.
Hope your connection upgrade will do it's job

Once I get re-established with my Radio linked ISP, all of these problems will go away.

I tried to install 7.5 on one of my usual hosts (as a veteran of 5.6x installations)

I'm got the same issue as these guys, I'm based in the UK with a standard broadband connection (Sky) - pretty fast and stable but still kept getting the multiple table issue.

On install I get the error message that C5 needs a clean database and that there are already 13x tables there. Which there are not - Just the empty DB and the Information Schema.

I asked my IT support to run it from their computers on a super fast connection and it took a while but worked! So perhaps there is something in the above timeout issue and C5's attempt to start adding tables again. However I just wanted to let you all know that it's not just very poor connections that may suffer.

As my first 7.5 installation I am a little underwhelmed. Can't wait to try my next one.

Hope this helps.

THis is not an issue of poor connections to the internet, as souch, it's an issue with ISP providers who use proxy servers for outgoing http requests on their network. By having your IT guy run the process for you from a different Internet connection, you circomvented the the proxy server on your connection.

You get the database not empty error because when the install fails, it doesn't show what the error was, just an empty red field and a link to go back. You would expect that this would take you back to a previous display page in the install process with an explaination as to what the problem was. Instead, that back link takes the browser back to the beginning of the install process and starts the install all over again in silance and then you get the error message that the database isn't empty because on the first pass, it filled the database with tables.

To see what process is failing durring installation, first remove all the tables from the database and use a browser like Firefox with Firebug. In Firebug, click on the net tab and in that tab, it will show each process that the installation is performing. C5's install process, uses Ajax, from the browser side to issue a series of commands to the server CGIs and then waite for a result code which is also displayed. Each process should return a 200 ok code. If your problem is the same as mine, when my installation got to the install dashboard process, after 45 seconds I got the error 504 "gateway time-out" and the install process just died. This was as a result of AT&T Cellular through which I was connecting pushes the http protocal through a proxy service and this proxy times out after 45 seconds. If you see this 504 error in Firebug while the install is running, it is a good bet that you also have a proxy server at the head end of your ISP that is timing out and slamming the door on you to.

The install dashboard process takes well over a minute maybe two to complete. The easiest sollution is to find an open hotspot somewhere in town and run it again. If that hotspot gives the same error, move on to the next untill you find one that is not using a proxy that times out.

I had a similar issue trying to install concrete5 on Rackspace (now liquidweb) Cloudsites. I think they run behind a proxy due to how it's a high availability environment.

The install said it was done, but lots of rows were missing from tables.

If we install locally and then upload the files and db, we have no problems running c5 on cloudsites, just can't do an install there.

I assumed it was something messing up with the install scripts, but I think you're on to something with proxy and timeouts.

Just encountered this as well

Your timeout suggestion helped me track this down, as in my case the http request from the installer returned sucesfully.

This was due to my site being in a linux container behind a proxy (Apache Trafficserver)

The solution here was to increase the origin server connection timeout in Apache Trafficserver to 4 minutes.

Without this thread I would have wasted a lot more time on this .... So, thanks!

I am running into the same issue on install. V 8.0.2.

I have attempted install on 3 separate ISP's. Has anyone figured this one out?