| Summary: | Failed to run in-place upgrade from 5.5 to 5.7 | ||
|---|---|---|---|
| Product: | Red Hat CloudForms Management Engine | Reporter: | luke couzens <lcouzens> |
| Component: | Red Hat Access | Assignee: | Satoe Imaishi <simaishi> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | luke couzens <lcouzens> |
| Severity: | high | Docs Contact: | |
| Priority: | high | ||
| Version: | 5.7.0 | CC: | abellott, dajohnso, ihands, jhardy, lphiri, ncarboni, obarenbo, simaishi |
| Target Milestone: | GA | Keywords: | TestOnly |
| Target Release: | 5.7.0 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | black:upgrade:migration | ||
| Fixed In Version: | 5.7.0.3 (rh-ruby23-rubygem-redhat_access_cfme-1.1.0-1, rh-ruby23-rubygem-redhat_access_lib-0.1.0-1) | Doc Type: | If docs needed, set a value |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2017-01-11 20:20:20 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
|
Description
luke couzens
2016-09-16 18:30:15 UTC
Lindani, can you take a look at this? We moved to ruby 2.3 in CFME 5.7 and the RPM name changed, so it's no longer considered as an upgrade. @Satoe We need to add an "obsoletes" directive to the the ruby 2.3 spec file. Okay, so it sounds like this is an issue with the RPM itself. If you're going to work on that can you take this BZ Lindani? @Nick, I am taking this one. Thanks Lindani! Updated Access RPMs with Obsolete directive. Will also require Obsolete directives for cfme rpm It looks like just using the new Access Insight RPMs is enough to get rid of that error. The new RPMs will be included in the next 5.7 build. There were some warning messages during upgrade. I will take a look at that separately as that's not related to access insight and that doesn't cause upgrade to fail. Since we're going from ruby 2.2 to ruby 2.3 as well as postgresql 9.4 to 9.5, the upgrade instruction will need to be updated. NickC will look into that. Nick, can you please provide the updated instruction when ready? Will do, will likely be waiting on https://github.com/ManageIQ/manageiq-appliance/pull/92 Hey Nick, Any update on the instructions to upgrade? Cheers, Luke I'm still waiting on https://github.com/ManageIQ/manageiq-appliance/pull/92 to be merged, but when that happens, I got it to work with the following steps: To put it simply it's https://access.redhat.com/articles/2297391 without the rubyrep stuff and with a postgres upgrade. I'll outline the steps I took here for clarity. First, set up the upgrade repos on all the appliances On all appliances: 1. `systemctl stop evmserverd` 2. `yum update` 3. `bin/rake db:migrate` 4. `bin/rake evm:automate:reset` On database server appliances: 1. `miq_postgres_upgrade.sh` (See PR referenced in comment 11) 2. Make required edits to postgresql.conf (a patch should be printed by the previous script) 3. `systemctl enable rh-postgresql95-postgresql && systemctl start rh-postgresql95-postgresql` On all appliances: `systemctl start evmserverd` Rebuild replication if rubyrep was being used - This upgrade will remove all the rubyrep tables and triggers as well as all the remote region data from the global database. After everything looks good you can `yum remove rh-postgresql94-*` on the database server appliances. I haven't actually upgraded a replicated environment though so this is just my best guess for that (we'll need to check both pglogical -> pglogical and rubyrep -> pglogical for this upgrade). You may also need to log out and log back in after the `yum update` and before the `bin/rake db:migrate` to properly load the new ruby 2.3 SCL. Verified yum update works from 5.5 to 5.7.0.5 |