Enable log collection for telemetry data through license servers.

Feb 13, 2025
Checksum:a46c1e09c58201616c747c7d3f2d169c5de0c7df8dea079f9b515a2123589edb

These updates contain fixes for the following issues:

• If there are connectivity issues between the pool coordinator and a host in the pool, a VM on the host that is rebooted during or after the connectivity issue might become unresponsive or might fail to start with the error: “an emulator required to run this VM failed to start”.
• In pools with a large number of hosts, changing the pool coordinator can fail and leave the proposed new coordinator in a broken state. After this failure, XenCenter cannot connect to either the old or the new pool coordinator. To recover any pools that had previously experienced this issue run the commands on the original pool master: xe pool-emergency-transition-to-master and xe pool-recover-slaves.
• Cross-pool migration can fail for halted VMs.
• Storage operations on one host can get stuck waiting for a response from an external process while holding an exclusive lock, which prevented other hosts from performing required operations.
• In a pool with a GFS2 SR, sometimes a failure in the distributed lock manager can cause a host to fail and require a reboot.
• A soft lockup and potential crash might occur when using an XFS SR with a PERC12 RAID controller.
• Sometimes, the forget SR operation can fail.
• When unpausing after a snapshot, read caching can fail to activate.
• On hosts with user-supplied certificates installed, sometimes the toolstack can fail to start.
• The toolstack fails to start when the host’s PEM does not have its private key followed by a certificate.
• Sometimes, XenServer can’t initialize Workload Balancing with a CA-signed certificate.
• When a VM is created in XenCenter, the Console tab for the VM can show the DVD option as not available.
• When the number of hosts in a pool is large, sometimes high availability (HA) can’t process heartbeats in time. As a result, the host whose heartbeats can’t be processed is marked as offline and self-fences.
• The pool coordinator is unable to join a clustered pool after applying updates.
• After a host successfully joins a pool after an offline update, XenCenter can show the incorrect state for this host.
• Sometimes, the synchronization of updates on a pool never completes.
• The rrd2csv command prints out a list of comma-separated metrics names, but it doesn’t accept this list as command-line arguments.
• RRDs for “Gauge” and absolute data sources change for those constant values.
• Some RRD-derived metrics can be incorrect.

These updates contain upstream code changes for the following issues:

• CVE-2024-12084
• CVE-2024-12085
• CVE-2024-12086
• CVE-2024-12087
• CVE-2024-12088
• CVE-2024-12747

These updates contain the following improvements:

• Update the Intel microcode to the IPU 2025.1 drop.
• Update the Intel ice driver to version 1.15.5.
• Improvements to logging.
• Enable log collection for telemetry data through license servers. For more information, see Data governance.