Brand new installation More than slow

SuiteCRM Forum - English Language SuiteCRM General Discussion
1

Hello,

I did a post few days ago, and did not have any reply. I can not edit it anymore and my title was not be really relevant. https://suitecrm.com/suitecrm/forum/suitecrm-7-0-discussion/20038-installation-really-really-really-slow).

By the way, I believed that my speed issue could come from old settings, so I decided to make a new installation on my OVH shared hosting. You can access it here https://www.septentriontours.com/SUITE/ .
You will see that it can take up to 3 minutes just to see the login page. On PhpMyAdmin, I can see that there is up to 3000 requests/ 5 secondes sent to the database just to display this login page.

My database has been just implemented, and I did not install any language pack or other moduls, neither any datas.

I am a kind of newby in computoring, but this one makes me crazy.

If anybody have a clue, I will be delighted.

Thx in advance

Php 7

2

Hello,

Up
Any help, please ?
Just a clue, may be?

Thanks

3

:whistle: I presume that August is not the better time of the year to have this kind of issue.

Just for your information, Joomla and Owncloud work perfectly under the same envirronement…

4

Hello,

Have you checked the permission & correct ownership of the SuiteCRM folder?

  1. Yes, it took a huge time to load login page. Maybe multiple redirect or issue with https?
  2. Can you try to access directly with IP?
  3. Also, check apache error log & suitecrm.log this may lead you to Investigate to the correct direction.
1 Like
5

Are these any errors in your Apache Log? SuiteCRM logs?

1 Like
6

Hello,

Here is what I have done with no results :

Reset Cache Custom Data Module Upload Theme and ConfigOveride on 775, everything else is on 755.
About the ownership, Filezilla give me a 845386 100 (this is the same for all the files I have on my shared host)
If you want me to change it, please tell me wich one I have to set, it is going to be my first Chmod.

I tried to disable https, by addind this kind of code in my .htacces, all I had it is redirection errors in Firefox.

 RewriteEngine On
RewriteCond %{HTTPS} on
RewriteRule (.*) http://%{HTTP_HOST}%{REQUEST_URI} [R=301,L]

The IP adress I collected using ping or any other web services (http://213.186.33.40) sent me to a non relevant page of my host.

Regarding logs, the suiteCRM.log gave me about 500 different " [FATAL] Slow Query".

The error log of my provider have several critical errors as this one:

 [host www.septentriontours.com] (13)Permission denied: AH00529: /homez.642/DATABASENAME/www/SUITE/cache/jsLanguage/EmailMan/.htaccess pcfg_openfile: unable to check htaccess file, ensure it is readable and that '/homez.642/DATABASENAME/www/SUITE/cache/jsLanguage/EmailMan/' is executable, referer: https://www.septentriontours.com/SUITE/index.php?module=EmailMan&action=index

However, as I said my rights are set properly, and my .htaccess is readable and active ( The redirections error in firefox). I tried to play with its permissions but nothing change. What is is its proper setting by the way ?

As you can see, I am really a newbie. If you need any more details just tell me where to find them.

Thank you very much for helping me on this issue.

7

Which version of SuiteCRM do you have?

Can you access it from phpMyAdmin? If so, please give the results of the first query in this post:

https://pgorod.github.io/Database-tables-size/

I suspect your Alerts table is overgrown. There was an old bug that caused performance problems with this.

Your server is taking 64 seconds to answer this request, and this request is in every screen of SuiteCRM…

https://www.septentriontours.com/SUITE/index.php?module=Alerts&action=get&to_pdf=1

1 Like
8

Hello,

I am on Suite CRM 7.10.

The first query on PHPmyAdmin gives me that :

[table][tr][td]
CONCAT(table_schema, ‘.’, table_name)[/td][td] rows[/td][td] DATA[/td][td] idx [/td][td]total_size [/td][td]idxfrac [/td][/tr][tr][td]
septentrbbsuite.job_queue [/td][td] 0.001M [/td][td] 0.000G[/td][td] 0.000G[/td][td] 0.000G [/td][td]0.355 [/td][/tr][tr][td]
septentrbbsuite.emails_text [/td][td] 0.000M[/td][td] 0.000G [/td][td]0.000G [/td][td]0.000G [/td][td]0.032 [/td][/tr][tr][td]
septentrbbsuite.relationships [/td][td]0.000M [/td][td]0.000G [/td][td]0.000G [/td][td]0.000G [/td][td]0.766 [/td][/tr][tr][td]
septentrbbsuite.acl_actions [/td][td]0.000M [/td][td]0.000G [/td][td]0.000G [/td][td]0.000G [/td][td]1.791 [/td][/tr][tr][td]
septentrbbsuite.campaign_log [/td][td]0.000M [/td][td]0.000G [/td][td]0.000G [/td][td]0.000G [/td][td]1.112 [/td][/tr][tr][td]
septentrbbsuite.email_templates [/td][td] 0.000M [/td][td] 0.000G[/td][td] 0.000G [/td][td]0.000G [/td][td]0.145 [/td][/tr][tr][td]
septentrbbsuite.prospects [/td][td]0.000M [/td][td]0.000G [/td][td]0.000G [/td][td]0.000G [/td][td]2.636 [/td][/tr][tr][td]
septentrbbsuite.email_addresses [/td][td] 0.000M[/td][td] 0.000G [/td][td]0.000G [/td][td] 0.000G [/td][td]2.504 [/td][/tr][tr][td]
septentrbbsuite.emails_beans [/td][td]0.000M [/td][td]0.000G [/td][td]0.000G [/td][td]0.000G [/td][td]0.882 [/td][/tr][tr][td]
septentrbbsuite.emailman [/td][td]0.000M[/td][td] 0.000G [/td][td]0.000G [/td][td]0.000G [/td][td]1.322 [/td][/tr][/table]

Nothing about the alert table…

I hope it can help.

Thank you again

9

Definitely no overgrown tables there…

Though I still suspect the Alerts because of the delays (you can see for yourself on your browser’s netwrok tab while you wait for pages to load).

The alerts code uses the date to check a condition. Can you check if your server’s date is correct, and if your PHP’s timezone is correctly defined?

10

Another thing: you said you had lots of “slow query” FATAL errors in suitecrm.log: which query is that? Is there a particular query that appears to be repeated most often?

11

Thanks for all theses advices :

In PHPMyAdmin,

SELECT CURTIME(), CURRENT_TIME(), CURRENT_TIME

give me the same time than the following php script I put in the SuiteCRM folder

<?php
$date = date("d-m-Y");
$heure = date("H:i");
Print("Nous sommes le $date et il est $heure");
?>

That’s the Paris one, the one set my users profile, and it looks the same as the one in my logs files.

Slow queries fatal logs are about anything starting by SELECT or INSERT in any table. It is just about what I am trying to do.

Sometimes I have

   [FATAL] log call at: modules/Administration/index.tpl:54 - MY_FRAME  is not set

but not so often.

Thank you so much for helping me on this one.

12

Does your PHP date script give the same results when you run it from the browser, and from the command-line?

php -f yourDateScript.php

The MY_FRAME message is a known issue, it is fixed in the next version which can be coming out as soon as this week, if tests go well. It is not a relevant message.

If you can’t solve the mystery, sometimes it’s a good idea to try and bypass it altogether, upgrading every software you can. So you could try going up to PHP 7.2 and MySQL 5.7 and hope the problem magically goes away. As a side-effect you get safer, more robust and efficient software in place.

13

Thank you about that.

My database is on a shared host, so I can not upgrade SQL.
About php, I went to the 7.2, nothing change.
The time is the same by command line or by the browser.

I hope the “magic effect” will take place once the new release you mentioned will be available.

It is a really a pity because I wanted to make my team use SuiteCRM that delighted me untill this issue. :S

Thx again

14

Your issue is very strange, and rare. I wonder if your hosting is providing you with a server that is fit for business. Maybe you can try getting your hosting support involved, say it’s behaving very slow, and ask them if they get some diagnostic of possible reasons…

Sometimes they’ll discover a failing disk or one of the other accounts on the server was hacked and is draining resources…

1 Like
15

Well,

I will try but at the end of the day, I have a Joomla website and a Owncloud storage on the same hosting that don’t look impacted by the same issue. That’s why it makes me mad.

I come back here if any kind of advice is given by them.

Thanks again,

16

Yes I admit it is weird.

Maybe try some repairs to the database form within MySQL. The problem seems to be in the database.

17

Hello,

Here is a small update about my issue that is still on.

My provider did not see any problems about the database.
I waited for a new release to install what I did last week-end.
SuiteCRM works but it is still really slow… (I let you try to display the login page : https://www.septentriontours.com/SUITE/ )

As my database is on a shared environment, I don’t have access to all its settings, all the brand new tables had been repaired and optimised. However something is telling me that I should try to index some tables but I don’t know which ones. (I am still a newbie…). I would be very delighted if anyone could help me with this one.

Just to remind you, Joomla and Owncloud work perfectly on the same server.

Thanks again

18

I don’t think this is solvable without some more general view on your server, you could be having all sorts of delays, DNS lookups, caching systems, etc. Only your hositng can probably figure out why the other apps run well but SuiteCRM doesn’t.

Note that the work SuiteCRM is doing just to show you the login page is pretty simple, and is tested a million times. I don’t think the problem is with SuiteCRM.

You could try tracking your log activity, first raise your log level to DEBUG (in Admin / System settings) and then connect an SSH console with Putty or another terminal and run this:

tail -f suitecrm.log

Then refresh the login page and see the lines scrolling by. If you notice a clear delay, press return 10 or 15 times to leave a space before any other lines print out. Then in the end you can examine your full scrollback and see what suitecrm was trying to do when it got delayed.

But really, this is the sort of thing that you shouldn’t be torturing yourself with, if I were you I would just move to a different hosting or a different server…

1 Like
19

Hi,

So with

 "tail -f suitecrm.log"

I did not detect any particular long execution time.

However every 30 sec or so, the system looks like checking cache and then it does that


Mon Nov  5 16:11:24 2018 [28223][-none-][DEBUG] Unable to find SugarController:: get
Mon Nov  5 16:11:24 2018 [28223][-none-][DEBUG] Hook called: ::server_round_trip
Mon Nov  5 16:11:24 2018 [28223][-none-][DEBUG] Calling MySQLi::disconnect()

I have got no clue of what it means, but if it can helps…

By the way, I was thinking to go on another server, but I am scared to not being able to mass email anymore (Today, my email accounts, hosts and servers are all in the same provider…)

Thanks again

20

I don’t see any particular issue with those messages, debug messages usually aren’t significant.

If you SuiteCRM handling of the request doesn’t include a long wait in the middlle, and yet your users have to wait 30 seconds or more to see a page, then the delay is before SuiteCRM gets called, or afterwards… right?

If you move to a different server, it shouldn’t affect email at all, SuiteCRM just uses SMTP, it definitely doesn’t have to be on the same server or on the same hosting company.