Email from Jon →
“just created a redhat directory on astrobackup2, and modified /etc/auto.home on astronomy to point the automount (/home/redhat) to it, the did a
cd /var/yp make
to push this to NIS for all other machines.”
Zach will start updating machines on a monthly basis on the last day of the month. After kernel updates, reboots will be required in order for new kernel to take effect (best schedule for reboots?).
== Last Time 'Yum Update' Was Run on the Machine?
Best practice - Run 'yum update' on one machine first to get an idea of what's being installed/updated. Then once that completes, you can run the doit scripts for CentOS 6 and 7 machines. For example:
[user@astronomy]$ /home/local/etc/doit-centos7 yum update -y
It is perhaps a good idea, after the doit script runs, to check the history to make sure no errors were noted:
[user@astronomy]$ /home/local/etc/doit-centos7 yum history
And then list any oddities in the below table –
NOTE - At 9:00am on 7-31-2020, Zach ran the doit script for CentOS7 machines. (this note is for me to track how long this process takes for a 'long' update. Updates finished at ~3:30pm).
Ran DO-IT centos7 script across all turned on machines on June 21st
| Machine Name | Last Updated? | Notes | |||
| Achernar | 6-1-2021 | ||||
| Acrux | 6-1-2022 | Zach's Office | |||
| Aldebaran | 7-2-2021 | ||||
| Alphaboo | 6-1-2022 | ||||
| Altair | 6-1-2022 | ||||
| Antares | 12-1-2020 | exclude kernel | |||
| Arcturus | 7-6-2021 | ||||
| Astrogpu | 7-2-2021 | ||||
| Babylon | 7-2-2021 | update & shutdown | |||
| Bellatrix | 7-2-2021 | update & shutdown | |||
| Betelgeuse | 7-2-2021 | ||||
| Boomerang | 6-1-2022 | ||||
| Canopus | 6-1-2021 | (unplugged 6/30) | |||
| Capella | 7-2-2021 | ||||
| Castor | 7-6-2021 | ||||
| Denebola | 6-1-2022 | ||||
| Fomalhaut | 7-2-2021 | ||||
| Gacrux | 11-25-2020 | CentOS 6 | |||
| Hadar | 6-1-2022 | ISSUE (see below) | |||
| Hyades | 7-2-2021 | ||||
| Kermodei | 6-30-2021 | ||||
| Kochab | 6-1-2022 | ||||
| Matrix | 6-1-2022 | ||||
| Milkyway | 6-30-2021 | Exclude Kernel | |||
| Mimosa | 6-1-2022 | ||||
| Mintaka | 7-7-2021 | ||||
| Oort | 10-25-2020 | Exclude Kernel (Shut Down) | |||
| Orion | 6-30-2022 | Upgraded to OS7 | |||
| Phoebe | 6-1-2021 | ||||
| Polaris | 6-1-2022 | ||||
| Pollux | 6-1-2022 | ||||
| Praesepe | 7-2-2021 | Exclude Kernel | |||
| Regulus | 7-6-2021 | Update & Shutdown | |||
| Rigel | 6-1-2022 | ||||
| Solarstorm | 6-29-2020 | ||||
| Spica | 7-2-2021 | ||||
| Tequila | 6-1-2022 | ||||
| Vega | 6-30-2021 | ||||
| Virgo | 6-30-2021 | CHECK | |||
| Wasp | 6-1-2022 | ||||
| Zenith | 6-29-2021 | Turned on - 3/22/21 |
NOTE - On Sept. 30th - Zach Noticed that Acrux, Mintake, Spica, and Zenith were shut down. They were shut down by means OTHER than Zach or Jon. Actions - Zach will go into the building Next week to inspect these machines, power them on and access them. Zach went into the department on 10-5-2020. Machines were okay. No foul play detected.
| Machine Name | Last Updated? | Notes | |||
| Sol | 5-7-2021 | Excluded HP packages, reboot | |||
| AstroData | 7-7-2021 | Updated apps, need to update firmware | |||
| Helix | 6-1-2022 | Updated applications Security risks? | |||
| AstroBackup3 | 7-7-2021 | No updates needed at this time | |||
| Cosmology1 | 7-7-2021 | Firmware and applications update | |||
| Cosmology4 | Hardware failure - data moved to hyades / Cos5 | ||||
| Cosmology5 | 7-7-2021 | Firmware and applications update | |||
| SSOC | 7-7-2021 | yum update –skip-brokeny |
NOTE For AstroBackUp, AstroBackUp3, Cosmology1,4,5 → you need to log in through the astronomy.nmsu.edu subnet and follow the GUI to update these machines.
Some machines will not be able to update the kernel due to small partition sizes in /boot file system.
In order to clear up space:
First, use yum to check the kernel installed:
[root@machine ~]# yum list kernel
Then, figure out which kernel is the active one:
[root@machine ~]# uname -a
Then, use yum to remove kernels that is 'NOT' the active kernal:
[root@machine ~]# yum remove kernel-old_version_number
Also, an additional work around is to just exclude the kernel in the update:
[root@machine ~]# yum update --exclude=kernel*
On 6-30-2021 Zach did an update to Hadar - In middle of the update, Hadar black screened and rebooted. This affected the display mainly (but other packages might have been affected). zach had to reinstalled package gdm (yum reinstall gdm).