Friday, May 15, 2015

GlusterFS 3.7.0 has been released, introducing many new features and improvements

Yesterday morning in Barcelona, the day after the Gluster Summit, GlusterFS 3.7.0 got released. Close to 600 bug reports, a mix of feature requests, enhancements and problems have been addressed in a little over 1220 patches since July last year.

This release marks the end of life of the maintenance of GlusterFS 3.4.x, users of the 3.4 stable version should consider upgrading.

The upcoming Fedora 23 (currently Rawhide) will be the first release that comes with 3.7.x by default, older Fedora releases will be kept on their stable Gluster versions. Packages for different distributions and versions will become available shortly on download.gluster.org. Once packages are available, announcements will be made on the Gluster Users list.

These Release Notes below have been put together with help from many engineers that worked on features and testing. You can also find the release notes in the repositories along with the rest of the code.

Major Changes and Features

Documentation about major changes and features is included in the doc/features/ directory of GlusterFS repository.

Bitrot Detection

Bitrot detection is a technique used to identify an “insidious” type of disk error where data is silently corrupted with no indication from the disk to the storage software layer that an error has occurred. When bitrot detection is enabled on a volume, gluster performs signing of all files/objects in the volume and scrubs data periodically for signature verification. All anomalies observed will be noted in log files.
For more information, refer here.

Multi threaded epoll for performance improvements

Gluster 3.7 introduces multiple threads to dequeue and process more requests from epoll queues. This improves performance by processing more I/O requests. Workloads that involve read/write operations on a lot of small files can benefit from this enhancement.
For more information refer here.

Volume Tiering [Experimental]

Policy based tiering for placement of files. This feature will serve as a foundational piece for building support for data classification.
For more information refer here.
Volume Tiering is marked as an experimental feature for this release. It is expected to be fully supported in a 3.7.x minor release.

Trashcan

This feature will enable administrators to temporarily store deleted files from Gluster volumes for a specified time period.
For more information refer here.

Efficient Object Count and Inode Quota Support

This improvement enables an easy mechanism to retrieve the number of objects per directory or volume. Count of objects/files within a directory hierarchy is stored as an extended attribute of a directory. The extended attribute can be queried to retrieve the count.
For more information refer here.
This feature has been utilized to add support for inode quotas.
For more details about inode quotas, refer here.

Pro-active Self healing for Erasure Coding

Gluster 3.7 adds pro-active self healing support for erasure coded volumes.

Exports and Netgroups Authentication for NFS

This feature adds Linux-style exports & netgroups authentication to the native NFS server. This enables administrators to restrict access to specific clients & netgroups for volume/sub-directory NFSv3 exports.
For more information refer here.

GlusterFind

GlusterFind is a new tool that provides a mechanism to monitor data events within a volume. Detection of events like modified files is made easier without having to traverse the entire volume.
For more information refer here.

Rebalance Performance Improvements

Rebalance and remove brick operations in Gluster get a performance boost by speeding up identification of files needing movement and a multi-threaded mechanism to move all such files.
For more information refer here.

NFSv4 and pNFS support

Gluster 3.7 supports export of volumes through NFSv4, NFSv4.1 and pNFS. This support is enabled via NFS Ganesha. Infrastructure changes done in Gluster 3.7 to support this feature include:
  • Addition of upcall infrastructure for cache invalidation.
  • Support for lease locks and delegations.
  • Support for enabling Ganesha through Gluster CLI.
  • Corosync and pacemaker based implementation providing resource monitoring and failover to accomplish NFS HA.
For more information refer the below links:
pNFS support for Gluster volumes and NFSv4 delegations are in beta for this release. Infrastructure changes to support Lease locks and NFSv4 delegations are targeted for a 3.7.x minor release.

Snapshot Scheduling

With this enhancement, administrators can schedule volume snapshots.
For more information, see here.

Snapshot Cloning

Volume snapshots can now be cloned to create a new writeable volume.
For more information, see here.

Sharding [Experimental]

Sharding addresses the problem of fragmentation of space within a volume. This feature adds support for files that are larger than the size of an individual brick. Sharding works by chunking files to blobs of a configurabe size.
For more information, see here.
Sharding is an experimental feature for this release. It is expected to be fully supported in a 3.7.x minor release.

RCU in glusterd

Thread synchronization and critical section access has been improved by introducing userspace RCU in glusterd

Arbiter Volumes

Arbiter volumes are 3 way replicated volumes where the 3rd brick of the replica is automatically configured as an arbiter. The 3rd brick contains only metadata which provides network partition tolerance and prevents split-brains from happening.
For more information, see here.

Better split-brain resolution

split-brain resolutions can now be also driven by users without administrative intervention.
For more information, see the 'Resolution of split-brain from the mount point' section here.

Geo-replication improvements

There have been several improvements in geo-replication for stability and performance. For more details, see here.

Minor Improvements

  • Message ID based logging has been added for several translators.
  • Quorum support for reads.
  • Snapshot names contain timestamps by default.Subsequent access to the snapshots should be done by the name listed in gluster snapshot list
  • Support for gluster volume get <volname> added.
  • libgfapi has added handle based functions to get/set POSIX ACLs based on common libacl structures.

Known Issues

  • Enabling Bitrot on volumes with more than 2 bricks on a node is known to cause problems.
  • Addition of bricks dynamically to cold or hot tiers in a tiered volume is not supported.
  • The following configuration changes are necessary for qemu and samba integration with libgfapi to work seamlessly:
      # gluster volume set <volname> server.allow-insecure on
    
    Edit /etc/glusterfs/glusterd.vol to contain this line: option rpc-auth-allow-insecure on
    Post 1, restarting the volume would be necessary:
      # gluster volume stop <volname> 
      # gluster volume start <volname>
    
    Post 2, restarting glusterd would be necessary:
      # service glusterd restart
    
    or
      # systemctl restart glusterd
    

Upgrading to 3.7.0

Instructions for upgrading from previous versions of GlusterFS are maintained on this wiki page.

Wednesday, May 6, 2015

1st beta of GlusterFS 3.5.4 is available for testing

You can download the packages (tarball and RPMs) for glusterfs-3.5.4beta1 from download.gluster.org.

23 bugs have been fixed. Once the fixes have been confirmed, we will do a real 3.5.4 (non-beta) release and update the packages in Fedora 20/21.

Release Notes for GlusterFS 3.5.4beta1

This is a bugfix release. The Release Notes for 3.5.0, 3.5.1, 3.5.2 and 3.5.3 contain a listing of all the new features that were added and bugs fixed in the GlusterFS 3.5 stable release.

Bugs Fixed:

  • 1092037: Issues reported by Cppcheck static analysis tool
  • 1101138: meta-data split-brain prevents entry/data self-heal of dir/file respectively
  • 1115197: Directory quota does not apply on it's sub-directories
  • 1159968: glusterfs.spec.in: deprecate *.logrotate files in dist-git in favor of the upstream logrotate files
  • 1160711: libgfapi: use versioned symbols in libgfapi.so for compatibility
  • 1161102: self heal info logs are filled up with messages reporting split-brain
  • 1162150: AFR gives EROFS when fop fails on all subvolumes when client-quorum is enabled
  • 1162226: bulk remove xattr should not fail if removexattr fails with ENOATTR/ENODATA
  • 1162230: quota xattrs are exposed in lookup and getxattr
  • 1162767: DHT: Rebalance- Rebalance process crash after remove-brick
  • 1166275: Directory fd leaks in index translator
  • 1168173: Regression tests fail in quota-anon-fs-nfs.t
  • 1173515: [HC] - mount.glusterfs fails to check return of mount command.
  • 1174250: Glusterfs outputs a lot of warnings and errors when quota is enabled
  • 1177339: entry self-heal in 3.5 and 3.6 are not compatible
  • 1177928: Directories not visible anymore after add-brick, new brick dirs not part of old bricks
  • 1184528: Some newly created folders have root ownership although created by unprivileged user
  • 1186121: tar on a gluster directory gives message "file changed as we read it" even though no updates to file in progress
  • 1190633: self-heal-algorithm with option "full" doesn't heal sparse files correctly
  • 1191006: Building argp-standalone breaks nightly builds on Fedora Rawhide
  • 1192832: log files get flooded when removexattr() can't find a specified key or value
  • 1200764: [AFR] Core dump and crash observed during disk replacement case
  • 1202675: Perf: readdirp in replicated volumes causes performance degrade

Known Issues:

  • The following configuration changes are necessary for 'qemu' and 'samba vfs plugin' integration with libgfapi to work seamlessly:
    1. gluster volume set <volname> server.allow-insecure on
    2. restarting the volume is necessary
      gluster volume stop <volname>
      gluster volume start <volname>
    3. Edit /etc/glusterfs/glusterd.vol to contain this line:
      option rpc-auth-allow-insecure on
    4. restarting glusterd is necessary
      service glusterd restart
    More details are also documented in the Gluster Wiki on the Libgfapi with qemu libvirt page.
  • For Block Device translator based volumes open-behind translator at the client side needs to be disabled.
    gluster volume set <volname> performance.open-behind disabled
  • libgfapi clients calling glfs_fini before a successful glfs_init will cause the client to hang as reported here. The workaround is NOT to call glfs_fini for error cases encountered before a successful glfs_init. This is being tracked in Bug 1134050 for glusterfs-3.5 and Bug 1093594 for mainline.
  • If the /var/run/gluster directory does not exist enabling quota will likely fail (Bug 1117888).

Sunday, March 1, 2015

Automatically subscribe RHEL systems for receiving updates and installing more packages

While fixing bugs and testing patches, I often use virtual machines running RHEL. These systems are short living, and normally do not survive a day or two. For most tests and development tries, I have little need to install additional packages or updates. An installation from the DVD contains all that is needed. Mostly...

To install additional packages or updates, it is needed to register the system to the Red Hat Customer Portal. The subscription-manager tool that is installed on all current RHEL systems can be used for that. For simple usage of the utility, a username and password is sufficient. Automating the subscribing process would require saving those credentials in a kickstart or ansible configuration, that's not what I want. Manually subscribing the VM when I need to was the annoying workaround.

A few weeks ago, I finally took the time to setup my automated RHEL installations to use subscription-manager for registering at the Red Hat Customer Portal. The Customer Portal offers the possibility to configure Activation Keys. The subscription-manager tool can use this key with a command like this:

# subscription-manager register \
        --org 123456 \
        --activationkey my-rhel-example-key
# subscription-manager attach --auto

The --org option seems required for me. I am not sure everyone needs that. The number (or name) can be found on an installed and registered system but executing:

# subscription-mananger identity

After subscribing like the above, it may well be that many repositories/channels get enabled. If you know which repositories you need, you can disable all repositories, and enable a select few:

# subscription-manager repos \
        --disable '*'
# subscription-manager repos \
        --enable rhel-7-server-rpms \
        --enable rhel-7-server-optional-rpms

At the moment, I have this done in the %post section of my kickstart configuration. I would prefer to set this up with the redhat_subscription ansible module, but the --org option is not available there (yet?).

Sunday, February 22, 2015

Add-to-agenda: Google+ Hangout about BitRot detection in GlusterFS

The day after tomorrow, on Tuesday 24 February at 13:30 UTC, Gluster developer Venky Shankar is hosting a Google+ Hangout BitRot Detection in GlusterFS: Overview, implementation details & demo. BitRot detection is one of the major features that will come with the next GlusterFS release, version 3.7.

I am looking very much forward to hear all the details from Venky!

Sunday, December 7, 2014

Configured Zabbix to keep my server cool

Recently I got myself an APC NetShelterCX mini. It is a 12U rack, with integrated fans for cooling. At the moment it is populated with some ARM boards (not rack mounted), their PDUs, a switch and (for now) one 2U server.

Surprisingly, the fans of the NetShelter are louder than the server (the rest does not have fans at all, except for the switch). But, it is not an option to keep the fans turned off all the time. When the server is idle, its CPUs temperature stays somewhere between 40-50 Celsius. However, when starting several virtual machines for testing some Gluster changes, the temperature rises steadily. To prevent overheating, the fans of the cabinet need to be turned on.

Of course, turning on the fans manually is possible, but it requires me to plug in the power cable. This is not very convenient when the cabinet is normally closed to reduce the noise. With the PDUs and fence_netio from the fence-agents-netio package, the fans inside the cabinet can be controlled remotely. That was a great step already!

Well, things can be even better. I don't want to monitor the temperature of my server and then decide when to turn on the fans. After spending some time looking for and comparing different monitoring solutions, I settled to try Zabbix. Packages for Zabbix are available for Fedora and EPEL, which makes trying it out pretty simple.

In a couple of hours playing with the installation and configuration, I was able to monitor the basics of my server. With a little manual configuration, the Zabbix Agent on the server can send the temperatures of the CPUs. All I had to do was setup a UserParameter in /etc/zabbix_agentd.conf:

UserParameter=cpu.temp.0,sensors | sed -n -r '/Physical id 0/s/^.*:[[:space:]]+\+([[:digit:]]+\.[[:digit:]]+).*$/\1/p'
UserParameter=cpu.temp.1,sensors | sed -n -r '/Physical id 1/s/^.*:[[:space:]]+\+([[:digit:]]+\.[[:digit:]]+).*$/\1/p'

The above configuration snippet tells the Zabbix Agent on the server to execute sensors (from the lm_sensors package), and filter the output through a sed command. The result is captured and sent to the Zabbix Server.

With the new cpu.temp.0 and .1 keys, the Zabbix webui can use these temperature items to setup a trigger which then invokes an action when the temperature rises above 55 Celsius. When the trigger enters the PROBLEM state, the action calls fence_netio and turns on the port that has the cable for the fans connected. When the trigger returns back to normal (checked every 5 minutes, now moved to 10), the port is disabled again.

This is the first time that I actually have setup monitoring with some custom actions. It was quite fun, and I'm certainly happy with the result.

Monday, November 17, 2014

Testing GlusterFS with very fast disks on Fedora 20

In the past I used to test with RAM-disks, provided by /dev/ram*. Gluster uses extended attributes on the filesystem, that makes is not possible to use tmpfs. While thinking about improving some of the GlusterFS regression tests, I noticed that Fedora 20 (and possibly earlier versions too) does not provide the /dev/ram* devices anymore. I could not find the needed kernel module quickly, so I decided to look into the newer zram module.

Getting zram working seems to be pretty simple. By default one /dev/zram0 is made available after loading the module. But, if needed, the module offers a parameter num_devices to create more devices. After loading the module with modprobe zram, you can do the following to create your high-performance volatile storage:

# SIZE_2GB=$(expr 1024 \* 1024 \* 1024 \* 2)
# echo ${SIZE_2GB} > /sys/class/block/zram0/disksize
# mkfs -t xfs /dev/zram0
# mkdir /bricks/fast
# mount /dev/zram0 /bricks/fast

With this mountpoint it is now possible to create a Gluster volume:

# gluster volume create fast ${HOSTNAME}:/bricks/fast/data
# gluster volume start fast

Once done with testing, stop and delete the Gluster volume, and free the zram like this:

# umount /bricks/fast
# echo 1 > /sys/class/block/zram0/reset

Of course, unloading the module with rmmod zram would free the resources too.

It is getting more important for Gluster to be prepared for very fast disks. Hardware like Fusion-io Flash drives and in future Persistent Memory/NVM will get more available in storage clouds, and of course we would like to see Gluster staying part of that!

Wednesday, November 5, 2014

Installing GlusterFS 3.4.x, 3.5.x or 3.6.0 on RHEL or CentOS 6.6

With the release of RHEL-6.6 and CentOS-6.6, there are now glusterfs packages in the standard channels/repositories. Unfortunately, these are only the client-side packages (like glusterfs-fuse and glusterfs-api). Users that want to run a Gluster Server on a current RHEL or CentOS now have difficulties installing any of todays current version of the Gluster Community packages.

The most prominent issue is that the glusterfs package from RHEL has a version of 3.6.0.28, and that is higher than the last week released version of 3.6.0. RHEL is shipping a pre-release that was created while the Gluster Community was still developing 3.6. An unfortunate packaging decision added a .28 to the version, where most other pre-releases would fall-back to a (rpm-)version like 3.6.0-0.1.something.bla.el6. The difference might look minor, but the result is a major disruption in the much anticipated 3.6 community release.

For the immediate need to fix this in a most easy way for our community users, we have decided to release version 3.6.1 later this week (maybe on Thursday November 6). This version is higher than the version in RHEL/CentOS, and therefore yum will prefer the package from the community repository over the one available in RHEL/CentOS. This is also the main reason why there have been no 3.6.0 packages provided on the download server.

Installing an older stable release (like 3.4 or 3.5) on RHEL/CentOS 6.6 requires a different approach. At the moment we can offer two solutions that can be used. We are still working on making this easier, until that is finalized, some manual actions are required.

Lets assume you want to verify if todays announced glusterfs-3.5.3beta2 packages indeed fix that bug you reported. (These steps apply to the other versions as well, this just happens to be what I have been testing.)

Option A: use exclude in the yum repository files for RHEL/CentOS

  1. download the glusterfs-353beta2-epel.repo file and save it under /etc/yum.repos.d/

  2. edit /etc/yum.repos.d/redhat.repo or /etc/yum.repos.d/CentOS-Base.repo and under each repository that you find, add the following line

    exclude=glusterfs*

This prevents yum from installing the glusterfs* packages from the standard RHEL/CentOS repositories, but allows those packages from others. The Red Hat Customer Portal has an article about this configuration too.

Option B: install and configure yum-plugin-priorities

Using yum-plugin-priorities is probably a more stable solution. This does not require changes to the standard RHEL/CentOS repositories. However, an additional package needs to get installed.

  1. enable the optional repository when on RHEL, CentOS users can skip this step

    # subscription-manager repos --list | grep optional-rpms
        # subscription-manager repos --enable=*optional-rpms

  2. install the yum-plugin-priorities package:

    # yum install yum-plugin-priorities

  3. download the glusterfs-353beta2-epel.repo file and save it under /etc/yum.repos.d/

  4. edit the /etc/yum.repos.d/glusterfs-353beta2-epel.repo file and add the following option to each repository definition:

    priority=50

The default priority for repositories is 99. The repositories with the lowest number have the highest priority. As long as the RHEL/CentOS repositories do not have the priority option set, the packages from the glusterfs-353beta2-epel.repo will get preferred by yum.

When using the yum-plugin-priorities approach, we highly recommend that you check if all your repositories have a suitable (or missing) priority option. In case some repositories have the option set, but yum-plugin-priorities was not installed yet, the order of the repositories might have changed. Because of this, we do not want to force using yum-plugin-priorities on all the Gluster Community users that run on RHEL/CentOS.

In case users still have issues installing the Gluster Community packages on RHEL or CentOS, we recommend getting in touch with us on the Gluster Users mailinglist (archive) or in the #gluster IRC channel on Freenode.