Synopsis: New Parallels Cloud Server 6.0 update introduces new features and provides stability fixes.
Issue date: 06-14-2013
Product: Parallels Cloud Server 6.0
Keywords: "stability" "usability"
1. What's Included in This Update
The new packages for Parallels Cloud Server 6.0 introduce new features and provide fixes for managing virtual machines, Containers, and Parallels Cloud Storage.
2. Problem Description
The new packages introduce the following features:
High Availability for Parallels Cloud Storage clusters (#PSBM-13291). High Availability keeps virtual machines and Containers operational even if the Hardware Node they are hosted on fails. In such cases, the affected virtual environments continue working on other, healthy Hardware Nodes in the cluster. High Availability is ensured by:
- Replication of metadata servers. For a Parallels Cloud Storage cluster to function, not all, but just the majority of MDS servers must be up. By setting up multiple MDS servers in the cluster, you will make sure that if an MDS server fails, other MDS servers will continue controlling the cluster.
- Replication of data. All data chunks are replicated, and the replicas are kept on different chunk servers. If a chunk server fails, other chunk servers will continue providing the data chunks that were stored on the failed server.
- Monitoring of Hardware Nodes' health.
Basic Dynamic Resource Scheduling (#PSBM-19150). Dynamic Resource Scheduling (DRS) provides load balancing for the High Availability feature in the Parallels Cloud Storage cluster and ensures that:
- Virtual environments that were moved from a failed Hardware Node to a healthy one are returned to the original Hardware Node once it is up again.
- Running virtual environments are spread across Hardware Nodes according to Nodes' license capacity. For example, if there are more virtual environments running on a Hardware Node than its license allows, the "extra" virtual environments will be moved to the Node that is the least occupied in terms of license capacity.
Online conversion of VZFS4-based Containers to the Container-in-an-image-file (ploop) layout (#PSBM-18696). Using the vzmigrate --convert-vzfs --online command, you can now convert VZFS4-based Containers to the Container-in-an-image-file (ploop) layout without restarting them. Source servers running the following Parallels products are supported:
- Parallels Virtuozzo Containers for Linux 4.6 (TU-4.6.0-254 (http://kb.parallels.com/en/116413))
- Parallels Virtuozzo Containers for Linux 4.7 (TU-4.7.0-310 (http://kb.parallels.com/en/116400))
- Parallels Server Bare Metal 5 Update 9 (http://kb.parallels.com/en/116417)
- Parallels Cloud Server 6.0 Update 3
The new packages contain fixes for the following issues:
Cannot migrate a Container online or configure IPtables in that Container if the "connlimit" rule is set in the IPtables configuration file (#PSBM-17553)
SSD/HDD read balancing is required inside CS (#PSBM-17791)
Resuming a suspended Container fails on vz restart (#PSBM-17995)
Parallels Cloud Server documentation only available after product installation (#PSBM-18949)
DNS resolver does not use all available methods to obtain the default domain names (#PSBM-19540)
Virtual machines relocated from a failed Hardware Node are reported as "running" by the shaman utility despite being, in fact, unregistered (#PSBM-19076)
Two sockets in /var/run/vzevents causing tools to hang (#PSBM-19765)
File leases stuck after MDS master change prevent virtual machines and Containers from being started, migrated, or backed up (#PSBM-19814)
Container registration issues after migration between Hardware Nodes in a Parallels Cloud Storage cluster (#PSBM-19855)
P2V migration fails due to segmentation fault (#PSBM-19931)
Restarting the master MDS causes virtual machines to stop (#PSBM-19978)
Sporadic replication going on for hours after CS reboot (#PSBM-20007)
A number of other minor issues were fixed (#PSBM-16846, #PSBM-16880,
PSBM-16988, #PSBM-18267, #PSBM-18511, #PSBM-18581, #PSBM-18636, #PSBM-18695,
PSBM-18737, #PSBM-18988, #PSBM-19103, #PSBM-19229, #PSBM-19255, #PSBM-19280,
PSBM-19373, #PSBM-19395, #PSBM-19399, #PSBM-19400, #PSBM-19401, #PSBM-19423,
PSBM-19426, #PSBM-19461, #PSBM-19463, #PSBM-19473, #PSBM-19495, #PSBM-19505,
PSBM-19506, #PSBM-19583, #PSBM-19620, #PSBM-19656, #PSBM-19681, #PSBM-19767,
PSBM-19768, #PSBM-19808, #PSBM-19813, #PSBM-19833, #PSBM-19838, #PSBM-19864,
PSBM-19915, #PSBM-19969, #PSBM-19973, #PSBM-20210, #PSBM-20171, #PSBM-20248)
The following known issue remains:
- Under certain conditions, virtual machines may restart during vzreboot operation (#PSBM-18708)
All Parallels Cloud Server 6.0 users are advised to update their tools packages.
3. Obtaining Update
You can download and install this tools update using the "yum" utility included in the Parallels Cloud Server 6.0 distribution.
Copyright (c) 1999-2013 Parallels Holdings, Ltd. and its affiliates. All rights reserved.