WHMCS backend incredibly slow after update to 7.10.1

[DennisHermannsen]

Figured I would post in this community, despite we've fixed the issue.

On 27th of April, we upgraded to WHMCS 7.10.1. I didn't notice anything immediately, but a few days after my boss said the backend had been loading really slow for a couple of days. With that in mind, I started navigating the WHMCS backend, and it was incredibly slow. Loading the front page took between 5-7 seconds.

Just today I was able to dump all MySQL trafic and saw something that started like this:
 

UPDATE tbladminlog SET logouttime = lastvisit WHERE adminusername='admin' AND lastvisit < '2020-05....

This ran for seconds. I couldn't reproduce on our other installations.

Checking the table, I could see it no longer had any indexes, primary key or AUTO INCREMENT. A lot of rows had id 0. Deleting all rows with id 0 and setting indexes, primary key and auto increment fixed the issue.

I'm really interested if anyone else has had this issue?

Edited May 25, 2020 by DennisHermannsen

[bear]

Though this is just a random thought, having an admin login log suddenly show it's been changed like that is concerning. I, for one, would be making quite sure it was the update when this started and not some other cause like unwanted access that attempted to hide his tracks...

[DennisHermannsen]

It definately was the update. I don't know how, though.

The last visible admin log in WHMCS was a few minutes before upgrading WHMCS.

[DennisHermannsen]

See this:

(72866, 'admin', '2020-04-27 12:46:23', '2020-04-27 12:46:23', 'xx.xx.xxx.xxx', 'xxxxxxxxxxxx', '2020-04-27 12:46:23'),
(0, 'admin', '2020-04-27 13:09:24', '2020-04-27 13:09:24', 'xx.xx.xxx.xxx', 'yyyyyyyyyyyy', '2020-04-27 13:09:24'), 

After logging in on April 27th, I initiated the upgrade. I don't know the exact time, but it was before 12:00. Logging in after this is logged as id 0.

[WHMCS Dan]

1 hour ago, DennisHermannsen said:

See this:

(72866, 'admin', '2020-04-27 12:46:23', '2020-04-27 12:46:23', 'xx.xx.xxx.xxx', 'xxxxxxxxxxxx', '2020-04-27 12:46:23'),
(0, 'admin', '2020-04-27 13:09:24', '2020-04-27 13:09:24', 'xx.xx.xxx.xxx', 'yyyyyyyyyyyy', '2020-04-27 13:09:24'), 

After logging in on April 27th, I initiated the upgrade. I don't know the exact time, but it was before 12:00. Logging in after this is logged as id 0.

This sounds like something which would be good for us to double check.

I would suggest raising a ticket with our team, including the version you upgraded from and to and your cPanel login details so we can check the update logs and see if we can spot any issues which may have caused this.

[DennisHermannsen]

Nevermind. I was wrong about the update dates. We had already updated on April 22nd.

We had another issue, and WHMCS took a look at it. One of the supporters made a lot of changes and moved our current installation to a new directory on the exact date we started experiencing issue. I guess there's a possibility something went wrong when he was investigating. The site was completely offline for more than an hour, and sometimes it gave an error where the database connection could not be established. I'm starting to think he may have taken a backup of the database (just in case), done something that messed everything up and imported a backup of the database - but somehow the backup must've not include indexes for that one table.