T4O 4.2.6 Release Notes

Release Versions

Packages

Deliverables against TVO-4.2.6

Following packages changed/added in the current release

Containers and Gitbranch

Changelog

• Verification of Jira issues targeted for 4.2.6 release

• Cohesity NFS/S3 storage backend support

Fixed Bugs and issues

1. Mounting a snapshot fails if the first disk of the File Recovery Manager VM is a CDROM

2. TVM HA configuration fails when one of the chosen Controller's hostname is part of any other TVM's hostname

3. Validation of keystone and s3 endpoint is stuck on the TVM UI

4. Temp volume creation taking too long

5. The httpd service is in a failed state on a freshly deployed T4O

6. The sshd option UseDNS should be set to "no" to avoid issues

Known issues

[RHOSP 16.1.8 and RHOSP 16.2.4] Trilio Horizon container in reboot loop

Observation : Post upgrade from a previous release/hotfix or on fresh deployment, the Trilio Horizon container is in a reboot loop

Workaround:

Either of the below workarounds should be performed on the controller where the issue occurs for the horizon pod.

option-1: Restart the memcached service on controller using systemctl (command: systemctl restart tripleo_memcached.service).

option-2: Restart the memcached pod (command: podman restart memcached).

[Snapshot mount] Unable to see correct content after mounting snapshot

Observation : Snapshot mount using RHEL8 File recovery manager image is not showing any mounted device

Workaround:

Use RHEL7 image for File recovery manager vm instead RHEL8.

[Backup failure for HPE nimble storage volumes with timeout receiving packet error

Observation : Snapshot is failing for VM's that have volumes backed by a Nimble iSCSI SAN with a "timeout receiving packet" error

Workaround: add uxsock_timeout parameter

log into respective datamover container and add uxsock_timeout with value as 60000 (i.e. 60 sec) in /etc/multipath.conf. Restart datamover container

[Input/Output error while writing to Cohesity NFS share]

Observation : Input/Output error during qemu-img convert operation while writing to Cohesity NFS share

Workaround: For cohesity NFS use below nfs options during Trilio configuration and datamover deployment and if issue still persists then increase timeo and retrans parameter values in nfs options

nfs_options = nolock,soft,timeo=600,intr,lookupcache=none,nfsvers=3,retrans=10