-
Posts
-
Hi there
We recently upgraded to 2.5 (we now have 2.6 running). After updating to 2.5 we’ve seen a large increase in resource usage across the board.Including network bandwidth, which is unusual. The cause looks to be an increase in resource usage by wp-cron.php.
Strace’ing one of the cron processes it looks like it might be w3tc. Specifically it looks like there’s some kind of issue regarding how it’s communicating with Redis.
I saw these in the logs during the plugin update process that seem to backup the issue being with w3tc/redis:
[31-Oct-2023 12:04:52 UTC] PHP Fatal error: Allowed memory size of 167772160 bytes exhausted (tried to allocate 225280 bytes) in /home/thefoldline-prod/public_html/wp-content/plugins/w3-total-cache/Cache_Redis.php on line 168
[31-Oct-2023 12:04:53 UTC] PHP Fatal error: W3TC\Generic_Plugin::ob_callback(): Cannot use output buffering in output buffering display handlers in /home/thefoldline-prod/public_html/wp-includes/blocks/blocks-json.php on line 3We’ve purged all caches to see if that helps but i wanted to raise with you in case its an issue further down.
All the best
Richard StarrHello Richard,
Thank you for reaching out and I am happy to help.
Nothing in the W3 Total Cache has been changed related to the redis_cache in the previous updates.
It seems like the key value is not fetching the old data or trying to do so.
Can you please try to flush redis databases and restart Redis service flushdb clears one database and flushall clears all databasesClear CURRENT
Delete default or currently selected database (usually 0) withredis-cli flushdb
Clear SPECIFIC
Delete specific redis database with (e.g. 8 as target database):redis-cli -n 8 flushdb
Clear ALL
Delete all redis databases withredis-cli flushall
FLUSHDB – clears currently active database
FLUSHALL – clears all the existing databases
- The topic ‘redis and output buffering’ is closed to new replies.
