Excessive CPU with Recurring Events

[Sayward]

I’ve got a client using The Events Calendar Pro with a significant number of recurring events (typically, 5-10+ events/day Monday to Friday, plus a few on the weekends). The schedule has been set up through June of 2017, resulting in about 2000 posts.

I started noticing timeout errors just after the site went live – usually when I was doing something in the Events Calendar. But I didn’t pinpoint it specifically to TEC at that point – I was just adding lots of events so that’s the screen I was usually on. After asking my tech support about the frequent timeouts, they checked the logs and found some issues. Part of their Email response is below:

it IS clear that at least a small part of the performance problem on the existing server is the sayward site itself.

[2016-10-08 11:20:07]: longrun: sayward/php5 30.51 seconds
(http://sayward.ca/event/age-friendly-program-drop-in/2016-10-11/)

[2016-10-08 11:25:07]: longrun: sayward/php5 30.38 seconds
(http://sayward.ca/event/thanksgiving-day-closed/all/)

[2016-10-08 11:29:47]: longrun: sayward/php5 29.85 seconds
(http://sayward.ca/event/open-swim-friday-am/2017-02-17/)

[2016-10-08 11:37:27]: longrun: sayward/php5 30.10 seconds
(http://sayward.ca/event/age-friendly-dominoes-2/2016-10-08/)

[2016-10-08 12:42:48]: longrun: sayward/php5 29.47 seconds
(http://sayward.ca/event/public-swim-friday-pm/2016-10-07/)

[2016-10-08 12:42:59]: longrun: sayward/php5 29.65 seconds
(http://sayward.ca/event/age-friendly-dominoes-2/all/)

Every one of those hits took approximately 30 cpu seconds to render. That
is NOT right.

and as far as I can tell, that appears to be repeatable.

ahh, it is actually failing…

<br />
<b>Fatal error</b>: Maximum execution time of 30 seconds exceeded in
<b>/raid/home/sayward/sayward/wp-includes/taxonomy.php</b> on line
<b>1532</b><br />

(That 30 cpu second limit would be controlled by your php.ini
file, but I highly doubt increasing it would help)

hhmm, tracing one of those processes… it appears to be stuck in a
loop… at least one mysql query (a complex one) got run 26 times before
the script ran out of CPU time.

SELECT wpvos_posts.*
FROM wpvos_posts
INNER JOIN wpvos_postmeta ON ( wpvos_posts.ID = wpvos_postmeta.post_id )
WHERE 1=1
AND ( \n wpvos_postmeta.meta_key = ‘_EventStartDate’\n)
AND wpvos_posts.post_type = ‘tribe_events’
AND ((wpvos_posts.post_status = ‘publish’))
GROUP BY wpvos_posts.ID
ORDER BY wpvos_posts.post_date DESC

Obviously, the common thread to the long running processes appears to be a
trailing date parameter?

I did find a couple of posts about excessive CPU issues with recurring events in the forums, but the latest, from June 2016, appears to be a different issue than this, as it was creating extra posts. That is definitely not the case here.

The client has a facility that runs recurring events from September to June. That was the key reason we chose TEC. Making events only recur for a month or so at a time is really not an option – this needs to be as hands-off as possible.

Thanks in advance for your reply.

Terry (Village webmaster)

[Cliff]

Hi Terry. Sorry you’re experiencing this. Yes, 30 seconds is very bad.

I’d suggest installing/activating this plugin to make sure you’re not affected by a “runaway recurrence” bug we had a short while back: tribe-extension-recurring-cleanup.zip (Recurring Events fixer plugin wp-admin page preview)

Please let me know how it goes for you.

[Support Droid]

Hey there! This thread has been pretty quiet for the last three weeks, so we’re going to go ahead and close it to avoid confusion with other topics. If you’re still looking for help with this, please do open a new thread, reference this one and we’d be more than happy to continue the conversation over there.

Thanks so much!
The Events Calendar Support Team

[Author]

Posts