Add DR with active-passive and external storage #3679
Conversation
github-actions
bot
added
Needs tech review
|
The PR preview for 9aba517 is available at theforeman-foreman-documentation-preview-pr-3679.surge.sh The following output files are affected by this PR: |
|
@ehelms Can you please review another DR scenario? I am once again making a lot of barely-educated guesses and I added several questions as comments so please keep that in mind and read carefully :) Also, I'm not sure what level of detail we are aiming for with this scenario: Do we want to keep things high-level like the virtualization scenario or can we/should we provide more detailed instructions? |
...dules/con_disaster-recovery-with-active-and-passive-project-server-and-external-storage.adoc
Outdated
Show resolved
Hide resolved
|
|
||
| In this scenario, you maintain two {ProjectServer}s and you store critical data externally on a network-attached storage (NAS). | ||
| The primary server is active while the secondary server remains passive. | ||
| If the primary server fails, you attach the NAS to your secondary server and turn it into your new primary server. |
There was a problem hiding this comment.
@evgeni They could attach it to both, and then flip it to write for the passive server when it needs to become the primary? Or is this a better /safer method?
...ring-for-disaster-recovery-with-active-and-passive-project-server-with-external-storage.adoc
Show resolved
Hide resolved
...ring-for-disaster-recovery-with-active-and-passive-project-server-with-external-storage.adoc
Outdated
Show resolved
Hide resolved
| . Stop the active server to mimic failure. | ||
| // How? foreman-maintain service stop? | ||
| . Start the passive server. | ||
| // How? foreman-maintain service start? |
There was a problem hiding this comment.
I can see two scenarios:
- A passive server that is running all the time but inaccessible. In this case the user would simply ensure storage is attached properly and flip DNS.
- A passive server that is something like an image, that gets turned up, storage attached and then update is run.
- A machine that is spun up, restore is run with storage attached.
@evgeni Vet me here
Co-authored-by: Eric Helms <[email protected]>
What changes are you introducing?
Adding a new disaster recovery scenario: Active and passive server with external storage.
Why are you introducing these changes? (Explanation, links to references, issues, etc.)
A follow-up to #3558
The new scenario follows the structure established in #3615 for consistency
Anything else to add? (Considerations, potential downsides, alternative solutions you have explored, etc.)
Now that I'm working on the second scenario, I can already see opportunities for reuse or snippets. However, I think it will be better to deal with this after all the planned scenarios are in place.
Checklists
Please cherry-pick my commits into: