I’ve got a number of accounts set up in Cyber Panel. Occasionally some will stop loading - producing a 500 error. Usually I can fix this issue by resetting the PHP on the account by either modifying the account or using the CLI cyberpanel changePHP command in terminal for the account. I haven’t been able to figure out why this is happening, but today I have a site that is having this issue, but the PHP reset isn’t fixing the problem.
I’ve done the following:
Set PHP to 7.4 via CLI
cyberpanel changePHP --domainName realhomesofgpab.ca --php 7.4
Modified the Website account via CyberPanel and setting the package and php and saving.
restarted Litespeed via command line:
restarted nghttpx via command line:
restarted PHP via command:
killall -9 lsphp
Ran Fix Permissions on public_html folder in account’s File Manager within Cyber Panel
PHP error log is showing PHP Notice for undefined offsets and indexes but no Fatal Errors.
Any ideas on what’s causing the 500 error, or what else I can do to troubleshoot this? Also be nice to know why it keeps happening to random sites.
any system error? The website loads fine and in seconds, you know, when it actually loads… df -ih? centos or ubuntu?
Thanks for checking - sounds like it loaded for you? I can’t seem to get it to load at all. I’m now getting ERR_HTTP2_PROTOCOL_ERROR when I try to load it.
Getting this error in the litespeed error log now:
2022-08-02 02:52:40.061828 [NOTICE]  [184.108.40.206:50982:HTTP2-3#realhomesofgpab.ca] No request delivery notification has been received from LSAPI application, possible dead lock.
Running Ubuntu 20.04
Nothing odd I can see in df -ih:
# df -ih
Filesystem Inodes IUsed IFree IUse% Mounted on
udev 4.0M 466 4.0M 1% /dev
tmpfs 4.0M 798 4.0M 1% /run
/dev/vda1 157M 6.7M 150M 5% /
tmpfs 4.0M 10 4.0M 1% /dev/shm
tmpfs 4.0M 3 4.0M 1% /run/lock
tmpfs 4.0M 18 4.0M 1% /sys/fs/cgroup
/dev/loop0 11K 11K 0 100% /snap/core18/2409
/dev/loop2 11K 11K 0 100% /snap/core18/2538
/dev/loop3 1.6K 1.6K 0 100% /snap/lxd/21029
/dev/loop5 486 486 0 100% /snap/snapd/16292
/dev/vdb1 63M 621 63M 1% /mnt/blockstorage
/dev/loop7 486 486 0 100% /snap/snapd/16010
/dev/loop6 802 802 0 100% /snap/lxd/22753
/dev/loop8 12K 12K 0 100% /snap/core20/1581
tmpfs 4.0M 23 4.0M 1% /run/user/0
/dev/loop1 12K 12K 0 100% /snap/core20/1587
I believe I’ve found the issue with a slow php/mysql script causing the loading issue. I disabled it and restarted php and the site is loading again. Thanks for the help!
yeah look like a deadlocking script, and with high timeouts, it cause this. Good luck
i have same issue , how you fixed? thanks you
Explain a bit more about your issue?
hello shoaibkk can you check this please
Being polite is when someone that spent the time reading your log telling you that obviously wp-rocket is misconfigured, there’s no cache and thus it’s never going to work, you don’t just dismiss it, you remove it, or configure it correctly ( you can’t under free OLS ). It’s just something that we know, wp-rocket, out of the box, unless you pay for litespeed, never gonna happen, gonna fail all stress test. And then you tell us that lscache generate the exact same invalid configuration…
there i posted more info with logs