All in all, it was definitely  worth attending this event. Given my role at HP, I regularly participate in similar events and I have found it well executed. Not all sessions were equal both in their content and delivery, but the informations I have gathered in Boston will definitely help me better understanding and managing Red Hat and JBoss on HP servers.

The slides of all presentations are now available. Enjoy !

The second presentation was made by the technical director of the JBoss project at Red Hat about tuning of JBoss.

I then attended a presentation about SElinux that was unfortunately not technical enough and did not meet my expectations.

However, the two last presentations (actually, one presentation filling two time slots) were absolutely fantastic. One kernel developer and a guy from Red Hat’s performance team have made a presentation about tuning Red Hat. It could not go further, as far as the technical level is concerned, and the questions from the audience were also interesting.

Finally, the day ended with a party in the Faneuil hall in Boston and in the surrounding pubs.

The first day started with a good keynote speech from Red Hat’s CEO, Jim Whitehurst, who talked about the shift that software companies will have to make in the future towards more modularity and more openness. He also mentioned how customers could get their money trapped in IT projects for several reasons: complexity, project failures, etc.  It was not a revolutionary speech, but rather a well executed introduction to Red Hat’s vision for the future.

Then came Paul Cormier, Red Hat’s VP for products and technologies who did the dirty job and first explained how Oracle wants to lock customers in their proprietary model with Solaris, Weblogic and the Oracle database.
Wait a second, isn’t that exactly what IBM is trying to do with AIX, Websphere and DB2 ? Of course yes, but IBM being the top sponsor of the event, Mr. Cormier preferred to target Oracle. Fair enough. Then Mr. Cormier tried to show us how Microsoft, VMware and Novell were offering closed proprietary solutions. Even though Red Hat remains a model in terms of a company making money out of open-source software, I do not buy what Mr. Cormier said about Novell and VMware. Novell remains a strong Linux vendor that is committed to open-source software and VMware, even though it is based on a proprietary and costly model, does an interesting job from a standards perspective: they have backed the OVF standard for virtual machines and have acquired SpringSource, a company selling free software.

Finally, a Cisco VP came to tell us how simple the world is with UCS. The more Cisco presentations I am hearing, the more I find their architecture complicated and their rip-and-replace-and-buy-new-switches approach brutal. But I have never seen their systems in action so far.

Following the keynote, I chose the 2 hours lab on Red Hat Enterprise Virtualization (RHEV). Even though the management interface is not as polished as VMware Vcenter (it lacks, for instance, the storage view) and even though storage online migration is not available, it will, in my humble opinion, cover the needs of 85% of the customers for a much lower price.

I then attended a session about achieving best I/O performance, both storage and networking, in a KVM virtualized environment., followed by a great presentation about how to setup a clustered environment with Red Hat 6. Even though RHEL6 is still in beta, the cluster management tool looks really promising.

Finally, I attended two sessions: one about SOA architectures in the cloud and RESTful applications with JBoss SOA platform which were fine.

To conclude the day, Red Hat organized an on-site party with a barbecue. It was interesting to see Jim Whitehurst, Red Hat’s CEO, be there and discuss very simply with the attendees.

All in all, it was a very interesting day and I am looking forward to seeing tomorrow’s presentations!

HP-UX is one of the few remainig commercial Unices out there. Today, only 3 play a significant role on the enterprise OS market : IBM AIX, running on Power processors, SUN Solaris, running on SPARC and x86 processors and HP-UX, running today on Intel Itanium-based servers. The latter used to run on HP’s own processors PA-RISC (for Precision Architecture) but after HP consolidated its portfolio on the only Itanium processors, it can only be purchased on IA64 processors.
HP-UX’s last version is HP-UX 11i v3, and this version has been enhanced and extended by three following updates.

File Hierarchy Standard (FHS)

HP-UX is a System V Unix. Let’s take a look at the root directory structure:

- /dev
This directory contains all device files, representing all devices attached to the system. Raw devices can be found under /dev/rdisk, block devices are in /dev/dsk and logical volumes are also placed in /dev

- /etc
As usual on Unix systems, all configuration files are stored in this directory.
All files controlling system startup and shutdown and daemons are located in /etc/rc.config.d. For instance, to configure your network interfaces for each boot and setup routes, you will edit the file /etc/rc.config.d/netconf and insert

INTERFACE_NAME[0]=lan0
IP_ADDRESS[0]=xx.xx.xx.xx
SUBNET_MASK[0]=xx.xx.xx.xx
BROADCAST_ADDRESS[0]=”"
INTERFACE_STATE[0]=”"
DHCP_ENABLE[0]=0
INTERFACE_MODULES[0]=”"

- /home
The usual home directory

- /sbin
The files located here are needed at boot time. The rc script that controls the the system startup is located in this directory as well as all runlevel dependent links to scripts respectively in /sbin/rc0.d/ , /sbin/rc1.d/ , etc. All available startup scripts are located in /sbin/init.d/

- /stand
The HP-UX kernel is kept in this directory. The kernel file is /stand/vmunix and the configuration file is /stand/system.

- /net
Reserved for remote file system mount points

- /opt
This directory is used to install applications on HP-UX. Each application has its own directory hierarchy. For the Apache Webserver, this will look like the following : /opt/hpws/ (for HP web services, the root directory, where Apache, Tomcat, and others are located) : /opt/hpws/apache/bin for executable binaries, /opt/hpws/apache/etc for the config files, and so on…

- /tmp
The temporary directory where anybody can create or delete files. Unlike Solaris, this directory is not flushed after reboot.

- /usr
Most HP-UX files are located here, especially user-relates commands, libraries, doc & manual pages and contributed software.
/usr/bin : user commands, applications and utilities.
/usr/include : include and header files for programming
/usr/lib : libraries for programming
/usr/sbin : many administration commands are located here
/usr/share/man : man pages

- /var
/var contains log files distributed across different directories:
/var/adm : system administration related programs (for instance /var/adm/sw for software installation/removal, /var/cron for cron log files, etc.)
/var/opt : application-related logs
/var/spool : print services related files

Virtualization

On the virtualization side, there are one feature shipped with any HP-UX instance : the Secure Resource Partitions (SRP) which are somehow similar to the Solaris zones.
You can assign particular resources to these partitions such as CPU shares, entire CPU’s (psets), memory as well as disk bandwith. Moreover, the security features of the SRP allow you to assign disk and network access to a particular partition. On the top of that, the SRP will make sur that your files (sockets, links, whatever…) cannot be accessed from outside the partition, unless a rule has been expressly defined in the config file. The SRP’s also use the capabilities of the role-based access control (RBAC) that allow a user to gain privileges to execute a single operation (e.g. a user can get privileges to mount a disk without having to be root). This allows the admin to define precise roles and increases the accountability of each user.

Beside the Base Operating Environment (BOE), which is shipped with any HP-UX, the user can also choose between the High Availibility OE and the Virtual Server Environment OE (or the Datacenter OE that contains everything).

With the VSE-OE, you have the choice between different virtualization types :
- nPars
nPars (for Node Partitions) are only available on mid-range and high-end cell-based servers. These cells consist of a kind of a plate on which CPUs and memory reside. Npars are just grouped cells that act as a single server.
- vPars
vPars (for Virtual Partitions) are more fine-grained: you can assign here resources on a per CPU basis. They run on HP-UX 11i v1, v2 and v3 only.
- Integrity Virtual Machines
The Integrity Virtual Machines are some kind of VMware-like virtual machines. They run on the top of a HP-UX guest that emulates the hardware and can paravirtualize the I/O drivers to accelerate the I/O throughput. Sysadmins create then virtual switches that can be internal or related to a physical interface to let the IVM communicate with the outside world. For network intensive workloads, you can assign an interface to a dedicated VM over a virtual switch to enhance performance. The last feature brought by the third update of HP-UX 11i v3 is the capability for IVM to migrate from one physical host to another (just like VMotion or Citrix XenMotion do).

Software and patch management

HP-UX’s software and patch management is based on Software Depot (SD-UX). Software packages are divided in different structures:

- Filesets: a collection of files
- Subproducts: group of filesets
- Products: examples are X11, Oracle, etc.
- Bundles: group of products
- Software depot: repository where software packages are made available for download and install on clients

Software depot can be used with an X or text-only interface, or for some tasks with the command line.

You can list the currently installed software on your system by entering the “swlist” command.
To install a new package, find out first on which server server has software depots. Then enter
# swlist -l depot @ hostname
to list all available depots on this server. Since you want to install a single package, enter
# swlist -d @ hostname:/path/to/the/depot
in order to list all applications and their description.
# swinstall -s hostname:/path/to/the/depot packagename
will then install the required software, resolving automatically the software dependencies.

Patches are managed in the same way and are also installed with the swinstall command.

Applications

HP-UX and the Integrity servers are designed for mission critical applications. Running an Apache server is possible, but it is not the kind of applications you will expect to see on such machines. Databases and Java application servers are rather the sweet spots of this OS.

A lot of commercial applications are available (to name a few : Oracle, Websphere, Weblogic,…) but MySQL, Tomcat, Apache and other Free Software are also ported and maintained on HP-UX.

The default shell is the Bourne shell, set in vi mode. Coming from a Linux/Bash world, it took me a little adaptation time.

Network

The network is managed with the tools lanscan (which will give information about the interfaces) and ifconfig (that will help configuring them). As we have seen it, the interfaces can be configured automatically at boot time in the file /etc/rc.config.d/netconf.

The domain names are resolved with the following files: /etc/hosts,  /etc/nsswitch.conf

Managing disks / SAN / Filesystems

The disks files are placed in the /dev/disks/ directory, the raw devices in /dev/rdisks/ . Note that you will only have one file to manage for a presented SAN LUN. The multipathing and failover are handled at the kernel level.

The most commonly used file system is onlineJFS, which is in fact a rebranded Veritas VxFS, and offers such features as online shrinking and extending of the filesystem and online refragmenting, which are much more reliable than, for instance, for ext3.

Resources : http://www.amazon.com/HP-Certified-HP-UX-System-Administration/dp/0130183741

SUN’s stock has increased of 78% yesterday, meaning that Wall Street is taking the rumors pretty seriously.

This move would create an industry giant and would reinforce IBM on the server as well as on the software and services market. Synergy effects resulting from this acquisition would especially help IBM compete against HP on the server and storage market. SUN’s latest financial results were poor and even put in question the viability of the company itself. SUN’s customers can now be sure that one strong company will support them. On the hardware side, IBM’s and SUN’s blade market shares account respectively for 25% and 5% of the total. A consolidation of these portfolios could help IBM compete against HP’s unrivaled leadership on this fast-growing market. IBM and SUN already work together (Lotus Symphony, IBM’s collaboration suite, already integrates SUN’s open-source OpenOffice.org tools suite) and these two companies are known to be strong open-source advocates. However, this acquisition still raises lots of questions.

SUN develops (with Fujitsu) its own SPARC processor line for high-end, business-critical servers and IBM’s POWER processors are their direct competitors. Processor development is an extremely expensive process and I cannot imagine IBM maintaining two different processor lines within the same company and I assume IBM would eventually drop SPARC. So what about customers, especially in the financial and communication industries which have a big SPARC install base ?

AIX, IBM’s Unix, competes directly versus SUN’s operating system Solaris on SPARC, moreover, Solaris is a direct competitor to Linux on the mass market x86. SUN’s strategy toward Linux has always been from hesitant to hostile whereas IBM has always been a great Linux supporter. Big Blue will have to deal with a portfolio problem, as well as with internal cultural problems. Solaris’ users and admins are very loyal to it and SUN has put Solaris at the heart of its datacenter strategy. If IBM was to drop Solaris for AIX, this would mean a lot of frustration for the Solaris community.

Finally, on the database level, IBM’s DB2 and Informix compete versus MySQL and PostgreSQL (the latter not directly owned but pushed by SUN) and SUN’s open source application server Glassfish is a direct threat to IBM’s middleware technology Websphere.

A merger between SUN and Dell would have made more sense from a pure portfolio perspective :
- Dell has a strong blade and x86 portfolio
- it owns no business-critical line of servers
- it could have pushed Solaris on x86
- Dell has no footprint neither on the application, nor on the middleware, nor on the application layer.
Perhaps has IBM moved before Dell did, fearing an emerging third strong player on the datacenter market.

Even though SUN’s acquisition by IBM is not official yet, I think the two companies will have really a hard time until they are completely integrated. The two companies have different technology and business cultures and SUN’s products mostly directly compete with IBM’s. Furthermore, today’s economic climate will make things even harder, since Big Blue could loose a lot of energy for the integration instead of focusing on customers. Of course, both companies together can leverage their huge portfolio to create synergy effects, however, I doubt this would happen soon.