Skip to content

add a scsb connection checker on the recap_hosts#6438

Draft
kayiwa wants to merge 4375 commits intomainfrom
i2_recap_connection
Draft

add a scsb connection checker on the recap_hosts#6438
kayiwa wants to merge 4375 commits intomainfrom
i2_recap_connection

Conversation

@kayiwa
Copy link
Member

@kayiwa kayiwa commented Aug 14, 2025

this installs a shell script that is a yank and paste of https://github.com/pulibrary/orangelight/blob/3eb7ff5cd64cd5da7c7e9d3d59f5af9b64021b93/app/models/concerns/requests/scsb.rb#L30 converted into a shell
we will need the api_key for this to work
related to https://gitlab.lib.princeton.edu/recap/recap/-/issues/2

Beck-Davis and others added 30 commits May 9, 2025 15:30
Add nomad-host-staging VMs to nomad project inventory
[CircleCI] Build a docker image for CircleCI Runners, deploy it to Nomad.
* use become for the tasks.
We've wiped Figgy Staging to test tigerdata.
* exclude warn messages at match

this excludes WARN messages from going to datadog
related #6143

* search for warn

search for warn in the logs

* regex for warn with space
Added bitcurator machines to inventory/all_projects/bitcurator under
[bitcurator_production]. Added bitcurator_production to
inventory/by_environment/production
removed amando user
Co-authored-by: Angel Ruiz <aruiz1789@users.noreply.github.com>
Co-authored-by: Angel Ruiz <aruiz1789@users.noreply.github.com>
…raefik (#6170)

Co-authored-by: Hector Correa <hectorcorrea@users.noreply.github.com>
Co-authored-by: Trey Pendragon <tpendragon@users.noreply.github.com>
[Datacommons] Finish bot wall implementation
* create new solr9cloud role

we create a new separate solr9cloud role. We have re-written solrcloud
but these commits make it difficult to work with our current running
solr8cloud. To simplify movement forward we elect to separate these
roles
This role is capable of installing solr of any version in a greenfield
ensemble

* add solr smb credentials

* create correct package

* restart daemon on servers and not CI

* make the role idempotent

* add verification

* add a bind-mount for backup

Solr 9 needs to be able to write under the solr data home and cannot
write to the mounted path.

this PR pulibrary/pul_solr#472 will be changing
where rake task looks at. A bind-mount is a true mount point that will
allow solr8 to continue working with the new path.
We're getting rid of Cantaloupe soon, and so switching this out during
the migration will mean we don't have to pay the Kakadu license.
Add reservoir-staging{1,2}.lib.princeton.edu to princeton_ansible
inventory
* Add pulmonitor-prod VMs to inventory
Add pulmonitor-prod{1,2} to ansible inventory

* Update checkmk

Corrected from .princeton.edu to .lib.princeton.edu
Corrects the FQDN for reservoir-staging{1,2} vms from #6178
This points Figgy to Tigerdata, but falls back to our Isilon whenever a
tigerdata file's not found.

Work towards pulibrary/figgy#6700
tpendragon and others added 29 commits August 5, 2025 15:14
Made small changes to aknowledge the role is now tested for more than just rails
Did not rename all the variables, becuase I did not want to change everyone's config files
Rely on the app config created by the ruby app role
[DPUL-C/Traefik] Add Traefik bot wall to DPUL-Collections
A kernel upgrade removed RabbitMQ, this fixes it and makes it
installable again. We had to totally delete RabbitMQ for this..
[LIB-JOBS] Remove the lib-jobs specific app config sfx_wroldcat
…riable

With the duplicate quotes, only the first username in the array actually gets added to the `ORANGELIGHT_ADMIN_NETIDS` environment variable, and everybody else does not get access to the flipflop features.
[orangelight] Remove unnecessary duplicate quotes from environment variable
…/libsvn (#6421)

Closes #6420

Co-authored-by: Christina Chortaria <christinach@users.noreply.github.com>
Co-authored-by: Jane Sandberg <sandbergja@users.noreply.github.com>
Put dpulc-prod behind the load bot wall.
The lib-svn- infrastructure has been retired. Cleans up inventory, removes the lib_svn and saxon roles.
[Lib_Jobs] Using active updates directory
Tidy inventory, get rid of playbook warnings
Our common role is setup assuming our virtual machines have checkmk
we add a task that checks for the checkmk binary. If it is present the
logrotate file is added. Otherwise it skips
This will let us queue things that are below "default", but above "low",
so that bulk ingests won't prevent other things.
* Try building a CI deployer with podman installed.

* Install podman in podman dependencies.

* Add privileged CI container.

* No privileged, set shm_size.

* Okay can't set shm_size.

* I give up.

---------

Co-authored-by: Jaymee Hyppolite <54775395+JaymeeH@users.noreply.github.com>
* remove ACME cert generation

our relayhost has a different domain name than the upstream relay.
we do not actually send mail but route to another relay

closes #6322

* remove host that was sunset

* many of our steps run on server only

this allows our CI to pass

* postfix back in 🤦
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.