glibc update on CentOS 7.4 results in Apache failure
All sites went down after today's 66.0.19 update.
WHM and cPanel were fine on those domains.
In WHM it said "You must reboot the server to apply software updates."
and
Apache status reported: "Failed to receive status information from Apache.
Unable to connect to local httpd server."
All services reported up (including httpd).
Apparently an automated Apache restart fixed the httpd issue.
I manually rebooted to apply updates.
Note: according to the update log the update took about 15 minutes to complete. Web services dropped almost immediately when it started, then 15 minutes after it completed (30 minutes in) Apache restarted and services came back. But I didn't manually restart Apache. Maybe cPanel monitoring is set for 15 minutes?
-
Hello, Internal case CPANEL-15830 is open to address an issue where updates to the glibc package in CentOS 7.4 result in Apache failing on systems using EasyApache 4 and Mod_Security. Here's an example of the output you will see in /usr/local/apache/error_log when this happens: httpd: Syntax error on line 27 of /etc/apache2/conf/httpd.conf: Syntax error on line 14 of /etc/apache2/conf.modules.d/800-mod_security2.conf: Cannot load modules/mod_security2.so into server: /lib64/libm.so.6: symbol __strtold_nan, version GLIBC_PRIVATE not defined in file libc.so.6 with link time reference
This is automatically fixed on systems with monitoring enabled for "Apache Web Server" in "WHM >> Service Manager". Otherwise, manually restarting Apache using "WHM Home " Restart Services " HTTP Server (Apache)" or with the following command will correct this issue:/scripts/restartsrv_httpd
I'll monitor the status of case CPANEL-15830 and update this thread with more information as it becomes available. Thank you.0 -
Yes, it was exactly as you described, as well as the solution. Question about WHM >> Service Manager, though. I verified that the config for it is set to monitor every 5 minutes, but it clearly took 15 minutes. Could this indicate an issue somewhere? Also my r1soft backup began failing after this update. At first I thought all of this was due to the 66.0.19 cPanel update, but the kernel update coincided with that and seems to be the cause. FYI I was not yet on 7.4 when all this happened, I was still on 7.3. Aany further details would be welcome. 0 -
Hello @PeteS, I moved those questions into their own separate threads: New Thread - Service Manager Question New Thread - R1Soft Backups failing Thank you. 0 -
Hello, To update, the following case is planned for inclusion in an EasyApache 4 maintenance release on 09-20-2017 to ensure Apache is restarted whenever glibc is updated through YUM: EA-6808: restart apache when glibc is updated Thank you. 0
Please sign in to leave a comment.
Comments
4 comments