2021-05-28 - Sys Ops & Management SIG Agenda and Notes

Owned by Ingolf Kuss
Last updated: May 28, 2021
2 min read

Date

Attendees

Discussion items

TimeItemWhoNotes

Find a note takerPhilip Robinson

Review Action Items


Technical Design DecisionAll

Teams must document deployment requirements in a clear & consistent way

With all the changes in Elasticsearch, data import, etc. with the Iris R1 release, it was more difficult to figure out how to deploy FOLIO. Edge modules became more complex (ex OCLC). The heavy use of Spring, while a good decision, was challenging to work with for some developers, and it is difficult to find people knowledgeable enough about it (and who would have sufficient spare time) to document it.

jroot started a running list of all of the back-end modules, their env vars and ports to work, and if they needed a db connection or not. It will be useful to carry forward this list to have it as a central place of reference.



R1 release aftermathall

Ingolf's notes:

New module mod-aes. Exporting SRS data. It's really super slow. It monitors all of Okapi traffic
and then reroutes the traffic to the reporting system. When you turn it off, migration gets faster.
AES: Audit data. Migration through the APIs is exteremely slow, compared to what it was before. Libraries reported, response in R1 is real slow compared to Q3. TAMU took mod-aes from the tenants. It really sped up a lot.

(Wayne after the meetimg : mod-aes has been taken out platform-complete and possibly only slipped in there by mistake).
SSO modules had to be installed in the Honeysuckle  mod-login-saml in Honeysuckle version.
That module is scheduled to be removed from the platform-complete list. If you don't have a robust Kafka install, it spams Kafka and really slows it down. TAMU had never did Kafka go down, but they had seen some Kafka slowness which they had not seen in Q3.
mod-aes sips the incoming and outcoming traffic; this is what an Okapi filter does.
Tod: AES = asynchronous event service. This was introduced along with the idea of a data lake. Get data out of Folio for reporting purposes. But it has undesired effects.
John Ballestro of A&M about issues that Cornell has been adding with data migration:
https://folio-org.atlassian.net/browse/MODDICORE-136
https://folio-org.atlassian.net/browse/MODDATAIMP-440 Data Import job is creating duplicate records when there is a background activity
Going through Kafka directly and not going via mod-pubsub
Configuring the new data import system:
Data import module parameters squeezing: https://wiki.folio.org/pages/viewpage.action?spaceKey=FOLIJET&title=Checklist+for+Data-Import+application+setup+on+a+new+environment%3A+Iris


Documentation GlossaryIngolf

Folio Documentation 

Technical terms in the Glossary

==> Deferred to June 4th session.


R1 2021 Iris Hotfix #1 

The hotfix has been scheduled for May 28th – today !

R1 2021 Iris Hot Fix #1 - FOLIO Issue Tracker .

==> Hotfix #1 is not yet out by the time of this meeting.

Meeting Notes

Cornell goes live July 1.

Chicago goes live with ERM only this summer.

Meeting references:

Action items


  •