CHG-2024-07-26 yum update
Status
2024-07-24 05:30 UTC
Creating first draft of this change ticket
Change Info
Scheduled Time
This change will take place on 2024-07-26 16:00 UTC
= 2024-07-26 11:00 Kansas City, US = 2024-07-26 11:00 Guayaquil, EC
https://www.timeanddate.com/worldclock/converter.html?iso=20240727T160000&p1=405&p2=1440&p3=93
Points of Contact
Change being performed by: Michael Altfield
Service owners: Catarina Mota & Marcin Jakubowski
Purpose
This CHG will update all the yum-installed packages on the server.
The last system-wide `yum update` appears to have been preformed on 2020-05-04, so we're well over-due for an update.
Note that security-critical updates are already being updated automatically via `yum-cron`.
Time Length
This whole process is expected to take 1-3 hours.
Some systems could be impacted for days, if issues are encountered.
Systems Impacted
The production server will be restarted after the update to apply kernel changes, and this will necessarily bring all of OSE's sites offline temporarily.
Staging Test
TODO
Pre-state Proof
The below command will output a ton of packages requiring updates before this CHG
yum list updates
Change Steps
# open screen screen -S CHG-2024-07-26_yum_update # become root sudo su - # confirm that backups have finished uploading to B2 less /var/log/backups/backup.log sudo -u b2user /home/b2user/virtualenv/bin/b2 ls ose-server-backups | grep `date "+%Y%m%d"` date -u sudo -u b2user /home/b2user/virtualenv/bin/b2 ls ose-server-backups # create dir for logging the change tmpDir="/var/tmp/CHG-2024-07-26_yum_update" mkdir -p $tmpDir # begin to gracefully shutdown nginx in the background time nice /sbin/nginx -s quit # first record the set of packages currently installed time nice rpm -qa &> "${tmpDir}/before.log" # update packages time nice yum -y update-to TODO &> "${tmpDir}/update.log" # log the post-state packages and versions time nice rpm -qa &> "${tmpDir}/after.log" # check to see what changes require a reboot, if any time nice needs-restarting &> "${tmpDir}/needs-restarting.log" time nice needs-restarting -r &> "${tmpDir}/needs-reboot.log" # reboot the system to apply kernel changes reboot # after reboot, initiate a new backup time sudo /bin/nice /root/backups/backup.sh &>> /var/log/backups/backup.log
Post-state Proof
The below command will output zero or nearly-zero packages after this CHG is successfully complete
yum list updates
Validation Steps
- Access the following URLs; make sure redirects and the destination pages look sane
- http://wiki.opensourceecology.org/
- http://opensourceecology.org/
- http://openbuildinginstitute.org/
- http://store.opensourceecology.org/
- http://oswh.opensourceecology.org/
- http://fef.opensourceecology.org/
- http://microfactory.opensourceecology.org/
- http://forum.opensourceecology.org/
- http://phplist.opensourceecology.org/lists/
- Access and login to the following private sites too to make sure everything is sane
- Login and attempt to make a trivial change on the following sites
Revert Steps
All the packages can be reverted to their previous versions using the following command
yum update-to TODO