Search Results

Search found 89 results on 4 pages for 'dtrace'.

Page 2/4 | < Previous Page | 1 2 3 4  | Next Page >

• Brendan Gregg's "Systems Performance: Enterprise and the Cloud"

  - by user12608550

  Long ago, the prerequisite UNIX performance book was Adrian Cockcroft's 1994 classic, Sun Performance and Tuning: Sparc & Solaris, later updated in 1998 as Java and the Internet. As Solaris evolved to include the invaluable DTrace observability features, new essential performance references have been published, such as Solaris Performance and Tools: DTrace and MDB Techniques for Solaris 10 and OpenSolaris (2006)  by McDougal, Mauro, and Gregg, and DTrace: Dynamic Tracing in Oracle Solaris, Mac OS X and FreeBSD (2011), also by Mauro and Gregg. Much has occurred in Solaris Land since those books appeared, notably Oracle's acquisition of Sun Microsystems in 2010 and the demise of the OpenSolaris community. But operating system technologies have continued to improve markedly in recent years, driven by stunning advances in multicore processor architecture, virtualization, and the massive scalability requirements of cloud computing. A new performance reference was needed, and I eagerly waited for something that thoroughly covered modern, distributed computing performance issues from the ground up. Well, there's a new classic now, authored yet again by Brendan Gregg, former Solaris kernel engineer at Sun and now Lead Performance Engineer at Joyent. Systems Performance: Enterprise and the Cloud is a modern, very comprehensive guide to general system performance principles and practices, as well as a highly detailed reference for specific UNIX and Linux observability tools used to examine and diagnose operating system behaviour.  It provides thorough definitions of terms, explains performance diagnostic Best Practices and "Worst Practices" (called "anti-methods"), and covers key observability tools including DTrace, SystemTap, and all the traditional UNIX utilities like vmstat, ps, iostat, and many others. The book focuses on operating system performance principles and expands on these with respect to Linux (Ubuntu, Fedora, and CentOS are cited), and to Solaris and its derivatives [1]; it is not directed at any one OS so it is extremely useful as a broad performance reference. The author goes beyond the intricacies of performance analysis and shows how to interpret and visualize statistical information gathered from the observability tools.  It's often difficult to extract understanding from voluminous rows of text output, and techniques are provided to assist with summarizing, visualizing, and interpreting the performance data. Gregg includes myriad useful references from the system performance literature, including a "Who's Who" of contributors to this great body of diagnostic tools and methods. This outstanding book should be required reading for UNIX and Linux system administrators as well as anyone charged with diagnosing OS performance issues.  Moreover, the book can easily serve as a textbook for a graduate level course in operating systems [2]. [1] Solaris 11, of course, and Joyent's SmartOS (developed from OpenSolaris) [2] Gregg has taught system performance seminars for many years; I have also taught such courses...this book would be perfect for the OS component of an advanced CS curriculum.

  Read the article

• Linux to Solaris @ Morgan Stanley

  - by mgerdts

  I came across this blog entry and the accompanying presentation by Robert Milkoski about his experience switching from Linux to Oracle Solaris 11 for a distributed OpenAFS file serving environment at Morgan Stanley. If you are an IT manager, the presentation will show you: Running Solaris with a support contract can cost less than running Linux (even without a support contract) because of technical advantages of Solaris. IT departments can benefit from hiring computer scientists into Systems Programmer or similar roles.  Their computer science background should be nurtured so that they can continue to deliver value (savings and opportunity) to the business as technology advances. If you are a sysadmin, developer, or somewhere in between, the presentation will show you: A presentation that explains your technical analysis can be very influential. Learning and using the non-default options of an OS can make all the difference as to whether one OS is better suited than another.  For example, see the graphs on slides 3 - 5.  The ZFS default is to not use compression. When trying to convince those that hold the purse strings that your technical direction should be taken, the financial impact can be the part that closes the deal.  See slides 6, 9, and 10.  Sometimes reducing rack space requirements can be the biggest impact because it may stave off or completely eliminate the need for facilities growth. DTrace can be used to shine light on performance problems that may be suspected but not diagnosed.  It is quite likely that these problems have existed in OpenAFS for a decade or more.  DTrace made diagnosis possible. DTrace can be used to create performance analysis tools without modifying the source of software that is under analysis.  See slides 29 - 32. Microstate accounting, visible in the prstat output on slide 37 can be used to quickly draw focus to problem areas that affect CPU saturation.  Note that prstat without -m gives a time-decayed moving average that is not nearly as useful. Instruction level probes (slides 33 - 34) are a super-easy way to identify which part of a function is hot.

  Read the article

• Hurry! See the uncensored OOW videos before they get edited!

  - by rickramsey

  source Uploaded so far: Which Oracle Solaris 11 Technologies Have Sysadmins Been Using Most? Director's Cut - Uncensored - Markus Flierl, VP Solaris Core Engineering, describes how Oracle Solaris 11 customers are taking advantage of the Image Packaging System and the snapshot capability of ZFS to run more frequent updates of not only the OS, but also the applications (agile development, anyone?), and how they're using the network virtualization capabilities in Oracle Solaris 11 to isolate applications and manage workloads on the cloud. Watch How Hybrid Columnar Compression Saves Storage Space Director's Cut - Uncensored - Art Licht shows how hyprid columnar compression (HCC) compresses data 30x without slowing down other queries that the database is performing. First he shows what happens when he runs database queries without HCC, then he shows what happens when he runs the queries with HCC. Security Capabilities and Design in Oracle Solaris 11 Director's Cut - Uncensored - Compliance reporting. Extended policy. Immutable zones. Three of the best minds in Oracle Solaris security explain what they are, what customers are doing with them, and how they were engineered. Filmed at Oracle Open World 2012. Why DTrace and Ksplice Have Made Oracle Linux 6 Popular with Sysadmins Use the DTrace scripts you wrote for Oracle Solaris on Oracle Linux without modification. Wim Coekaerts, VP of Engineering for Oracle Linux, explains how this capability of DTrace, the zero downtime updates enabled by KSplice, and other performance and stability enhancements have made Oracle Linux 6 popular with sysadmins. Why Solaris 11 Is Being Adopted Faster Than Solaris 10 Sneak Preview - Uncut Version - Lynn Rohrer, Director of Oracle Solaris Product Management explains why customers are adopting Oracle Solaris 11 at a faster rate than Oracle Solaris 10, and proves why you should never challenge a Montana woman to a test of strength. What Forsythe Corp Is Helping Its Customers Do With Oracle Solaris 11 Director's Cut - Unedited - Lee Diamante, Solutions Architect for Forsythe Corp, an Oracle Solaris Partner, explains why Forsythe has been recommending Oracle Solaris to its customers, and what those customers have been doing with it. Lots more to come ... - Rick Website Newsletter Facebook Twitter

  Read the article

• Morgan Stanley chooses Solaris 11 to run cloud file services

  - by Frederic Pariente

  At the EAKC2012 Conference last week in Edinburg, Robert Milkowski, Unix engineer at Morgan Stanley, presented on deploying OpenAFS on Solaris 11. It makes a great proofpoint on how ZFS and DTrace gives a definite advantage to Solaris over Linux to run AFS distributed file system services, the "cloud file system" as it calls it in his blog. Mike used ZFS to achieve a 2-3x compression ratio on data and greatly lower the TCA and TCO of the storage subsystem, and DTrace to root-cause scalability bottlenecks and improve performance. As future ideas, Mike is looking at leveraging more Solaris features like Zones, ZFS Dedup, SSD for ZFS, etc.

  Read the article

• MySQL getting stuck, eating up disk i/o

  - by bonez05

  Hi all, Using mySQL 5.0.51 on Solaris . At intermittent times it looks like MySQL is getting 'stuck' . The disk usage on the server spikes to 98% busy from reads. I used dtrace (specifically DTrace toolkit - iosnoop) to track down what processes was using all the reads. Mysql was calling tablename.TDM hundreds of times per second. There was no more than average load on the webserver that could account for this. There were no cronjobs running, and no other utilities like mysqldump or anything. It is a master / slave replication setup. As a jury-rigged fix, I altered the mysql table from 'tablename' to 'tablename2' and then back to 'tablename' This fixed the problem temporarily, and "unsticks" mysql. The disk usage goes back down and dtrace is no longer showing hundreds of reads to 'tablename.TDM' / second. A couple ideas I had are: 1. MySQL version bug 2. Infinite loop somewhere in my application (which i'm not sure how likely this is) 3. ?? Has anybody seen this before or have any insight? Thanks

  Read the article

• PENGUIN IS GETTING READY FOR ORACLE OPENWORLD 2012

  - by Zeynep Koch

  Are you looking for reasons to attend Oracle Openworld, how about below Oracle Linux sessions and hands-on-labs.  1. General Session: Oracle Linux Strategy and Roadmap  In this session, Oracle executives will discuss Linux strategy; the roadmap; contributions to the Linux mainline kernel; and what's in store for upcoming releases of Oracle Linux and the Unbreakable Enterprise Kernel. Don’t miss this session. 2. New Features in Oracle Linux- A Technical Deep Dive Collaborating with the Linux community, Oracle engineers contribute to advancing Linux for mission-critical deployments. In this technical session, attendees will learn about the recent developments in Oracle Linux and the Unbreakable Enterprise Kernel 3. Why Switch to Oracle Linux?  Oracle is the only company that provides a complete Linux solution from applications to disk, fully optimized for Oracle hardware and software, with one-stop support. In this session you will hear from two customers that have successfully implemented Oracle Linux and saved 50 to 90 percent on Linux support costs as well as the reasons to switch to Oracle Linux. 4. Debugging and Configuration Best Practices for Oracle Linux This is one of our best attended sessions and most informative. In this best practices session, learn how to save time and money while preventing headaches and hassles. Discover expert secrets to get your Linux systems up and running (and keep them running), avoid common pitfalls, prevent problems, and circumvent known issues. 5. Top Technical Tips for Automatic and Secure Oracle Linux Deployments In this session, attendees will learn about how to easily deploy and install Oracle Linux systems using various technologies like Kickstart, Oracle Enterprise Manager OpsCenter, and Oracle VM Templates for applications on Linux. Additionally, the session will share useful Linux security tips and introduce utilities to help with hardening and securely operating an Oracle Linux system. We also have a great session in Oracle Develop track: 6. DTrace for Oracle Linux Initially announced at last year's Oracle Openworld, DTrace for Oracle Linux is now available for the Unbreakable Enterprise Kernel R.2. In this session held by one of the engineers working on the DTrace for Linux port, you will learn how you can use this powerful and flexible framework in your development environment. If you prefer to really have practical experience, don’t miss our two Hands-on-Labs where we will cover: HOL-1 : Oracle Linux Package Management: Configuring and Enabling Services In this session you will be Installing and configuring Oracle VM VirtualBox, importing the Oracle Linux virtual appliance. You will then use the package management on Oracle Linux using RPM and yum. You will also be able to review Ksplice, zero downtime kernel updates that enable you to apply security updates, patches and critical bug fixes without rebooting. HOL-2: Oracle Linux Storage Management with LVM and Device Mapper In this session you will learn about storage management with LVM2, the Linux Logical Volume Manager, Btrfs, preparing block devices, creating physical and logical volumes, creating file systems on top of logical volumes, and resizing file systems dynamically. You will also practice setting up software RAID devices, configuring encrypted block devices. You will also see Oracle Linux and Kpslice in the three demopods we will feature at Exhibition demogrounds. One in MySQL Connect and two in Oracle Openworld. What more do you need to come to San Francisco? Oh, I forgot to mention we also have great weather in fall.. Check out the Content Catalog and register to attend Oracle Linux sessions.

  Read the article

• Solaris TCP/IP performance tuning

  - by Andy Faibishenko

  I am trying to tune a high message traffic system running on Solaris. The architecture is a large number (600) of clients which connect via TCP to a big Solaris server and then send/receive relatively small messages (.5 to 1K payload) at high rates. The goal is to minimize the latency of each message processed. I suspect that the TCP stack of the server is getting overwhelmed by all the traffic. What are some commands/metrics that I can use to confirm this, and in case this is true, what is the best way to alleviate this bottleneck? PS I posted this on StackOverflow originally. One person suggested snoop and dtrace. dtrace seems pretty general - are there any additional pointers on how to use it to diagnose TCP issues?

  Read the article

• Announcing Unbreakable Enterprise Kernel Release 3 for Oracle Linux

  - by Lenz Grimmer

  We are excited to announce the general availability of the Unbreakable Enterprise Kernel Release 3 for Oracle Linux 6. The Unbreakable Enterprise Kernel Release 3 (UEK R3) is Oracle's third major supported release of its heavily tested and optimized Linux kernel for Oracle Linux 6 on the x86_64 architecture. UEK R3 is based on mainline Linux version 3.8.13. Some notable highlights of this release include: Inclusion of DTrace for Linux into the kernel (no longer a separate kernel image). DTrace for Linux now supports probes for user-space statically defined tracing (USDT) in programs that have been modified to include embedded static probe points Production support for Linux containers (LXC) which were previously released as a technology preview Btrfs file system improvements (subvolume-aware quota groups, cross-subvolume reflinks, btrfs send/receive to transfer file system snapshots or incremental differences, file hole punching, hot-replacing of failed disk devices, device statistics) Improved support for Control Groups (cgroups)  The ext4 file system can now store the content of a small file inside the inode (inline_data) TCP fast open (TFO) can speed up the opening of successive TCP connections between two endpoints FUSE file system performance improvements on NUMA systems Support for the Intel Ivy Bridge (IVB) processor family Integration of the OpenFabrics Enterprise Distribution (OFED) 2.0 stack, supporting a wide range of Infinband protocols including updates to Oracle's Reliable Datagram Sockets (RDS) Numerous driver updates in close coordination with our hardware partners UEK R3 uses the same versioning model as the mainline Linux kernel version. Unlike in UEK R2 (which identifies itself as version "2.6.39", even though it is based on mainline Linux 3.0.x), "uname" returns the actual version number (3.8.13). For further details on the new features, changes and any known issues, please consult the Release Notes. The Unbreakable Enterprise Kernel Release 3 and related packages can be installed using the yum package management tool on Oracle Linux 6 Update 4 or newer, both from the Unbreakable Linux Network (ULN) and our public yum server. Please follow the installation instructions in the Release Notes for a detailed description of the steps involved. The kernel source tree will also available via the git source code revision control system from https://oss.oracle.com/git/?p=linux-uek3-3.8.git If you would like to discuss your experiences with Oracle Linux and UEK R3, we look forward to your feedback on our public Oracle Linux Forum.

  Read the article

• Oracle Solaris Zones Physical to virtual (P2V)

  - by user939057

  IntroductionThis document describes the process of creating and installing a Solaris 10 image build from physical system and migrate it into a virtualized operating system environment using the Oracle Solaris 10 Zones Physical-to-Virtual (P2V) capability.Using an example and various scenarios, this paper describes how to take advantage of theOracle Solaris 10 Zones Physical-to-Virtual (P2V) capability with other Oracle Solaris features to optimize performance using the Solaris 10 resource management advanced storage management using Solaris ZFS plus improving operating system visibility with Solaris DTrace. The most common use for this tool is when performing consolidation of existing systems onto virtualization enabled platforms, in addition to that we can use the Physical-to-Virtual (P2V) capability  for other tasks for example backup your physical system and move them into virtualized operating system environment hosted on the Disaster Recovery (DR) site another option can be building an Oracle Solaris 10 image repository with various configuration and a different software packages in order to reduce provisioning time.Oracle Solaris ZonesOracle Solaris Zones is a virtualization and partitioning technology supported on Oracle Sun servers powered by SPARC and Intel processors.This technology provides an isolated and secure environment for running applications. A zone is a virtualized operating system environment created within a single instance of the Solaris 10 Operating System.Each virtual system is called a zone and runs a unique and distinct copy of the Solaris 10 operating system.Oracle Solaris Zones Physical-to-Virtual (P2V)A new feature for Solaris 10 9/10.This feature provides the ability to build a Solaris 10 images from physical system and migrate it into a virtualized operating system environmentThere are three main steps using this tool1. Image creation on the source system, this image includes the operating system and optionally the software in which we want to include within the image. 2. Preparing the target system by configuring a new zone that will host the new image.3. Image installation on the target system using the image we created on step 1. The host, where the image is built, is referred to as the source system and the host, where theimage is installed, is referred to as the target system. Benefits of Oracle Solaris Zones Physical-to-Virtual (P2V)Here are some benefits of this new feature:  Simple- easy build process using Oracle Solaris 10 built-in commands.  Robust- based on Oracle Solaris Zones a robust and well known virtualization technology.  Flexible- support migration between V series servers into T or -M-series systems.For the latest server information, refer to the Sun Servers web page. PrerequisitesThe target Oracle Solaris system should be running the latest version of the patching patch cluster. and the minimum Solaris version on the target system should be Solaris 10 9/10.Refer to the latest Administration Guide for Oracle Solaris for a complete procedure on how todownload and install Oracle Solaris. NOTE: If the source system that used to build the image is an older version then the targetsystem, then during the process, the operating system will be upgraded to Solaris 10 9/10(update on attach).Creating the Image Used to distribute the software.We will create an image on the source machine. We can create the image on the local file system and then transfer it to the target machine, or build it into a NFS shared storage andmount the NFS file system from the target machine.Optional  before creating the image we need to complete the software installation that we want to include with the Solaris 10 image.An image is created by using the flarcreate command:Source # flarcreate -S -n s10-system -L cpio /var/tmp/solaris_10_up9.flarThe command does the following:  -S specifies that we skip the disk space check and do not write archive size data to the archive (faster).  -n specifies the image name.  -L specifies the archive format (i.e cpio). Optionally, we can add descriptions to the archive identification section, which can help to identify the archive later.Source # flarcreate -S -n s10-system -e "Oracle Solaris with Oracle DB10.2.0.4" -a "oracle" -L cpio /var/tmp/solaris_10_up9.flarYou can see example of the archive identification section in Appendix A: archive identification section.We can compress the flar image using the gzip command or adding the -c option to the flarcreate commandSource # gzip /var/tmp/solaris_10_up9.flarAn md5 checksum can be created for the image in order to ensure no data tamperingSource # digest -v -a md5 /var/tmp/solaris_10_up9.flar Moving the image into the target system.If we created the image on the local file system, we need to transfer the flar archive from the source machine to the target machine.Source # scp /var/tmp/solaris_10_up9.flar target:/var/tmpConfiguring the Zone on the target systemAfter copying the software to the target machine, we need to configure a new zone in order to host the new image on that zone.To install the new zone on the target machine, first we need to configure the zone (for the full zone creation options see the following link: http://docs.oracle.com/cd/E18752_01/html/817-1592/index.html  )ZFS integrationA flash archive can be created on a system that is running a UFS or a ZFS root file system.NOTE: If you create a Solaris Flash archive of a Solaris 10 system that has a ZFS root, then bydefault, the flar will actually be a ZFS send stream, which can be used to recreate the root pool.This image cannot be used to install a zone. You must create the flar with an explicit cpio or paxarchive when the system has a ZFS root.Use the flarcreate command with the -L archiver option, specifying cpio or pax as themethod to archive the files. (For example, see Step 1 in the previous section).Optionally, on the target system you can create the zone root folder on a ZFS file system inorder to benefit from the ZFS features (clones, snapshots, etc...).Target # zpool create zones c2t2d0 Create the zone root folder:Target # chmod 700 /zones Target # zonecfg -z solaris10-up9-zonesolaris10-up9-zone: No such zone configuredUse 'create' to begin configuring a new zone.zonecfg:solaris10-up9-zone> createzonecfg:solaris10-up9-zone> set zonepath=/zoneszonecfg:solaris10-up9-zone> set autoboot=truezonecfg:solaris10-up9-zone> add netzonecfg:solaris10-up9-zone:net> set address=192.168.0.1zonecfg:solaris10-up9-zone:net> set physical=nxge0zonecfg:solaris10-up9-zone:net> endzonecfg:solaris10-up9-zone> verifyzonecfg:solaris10-up9-zone> commitzonecfg:solaris10-up9-zone> exit Installing the Zone on the target system using the imageInstall the configured zone solaris10-up9-zone by using the zoneadm command with the install -a option and the path to the archive.The following example shows how to create an Image and sys-unconfig the zone.Target # zoneadm -z solaris10-up9-zone install -u -a/var/tmp/solaris_10_up9.flarLog File: /var/tmp/solaris10-up9-zone.install_log.AJaGveInstalling: This may take several minutes...The following example shows how we can preserve system identity.Target # zoneadm -z solaris10-up9-zone install -p -a /var/tmp/solaris_10_up9.flar Resource management Some applications are sensitive to the number of CPUs on the target Zone. You need tomatch the number of CPUs on the Zone using the zonecfg command:zonecfg:solaris10-up9-zone>add dedicated-cpuzonecfg:solaris10-up9-zone> set ncpus=16DTrace integrationSome applications might need to be analyzing using DTrace on the target zone, you canadd DTrace support on the zone using the zonecfg command:zonecfg:solaris10-up9-zone>setlimitpriv="default,dtrace_proc,dtrace_user" Exclusive IP stack An Oracle Solaris Container running in Oracle Solaris 10 can have a shared IP stack with the global zone, or it can have an exclusive IP stack (which was released in Oracle Solaris 10 8/07). An exclusive IP stack provides a complete, tunable, manageable and independent networking stack to each zone. A zone with an exclusive IP stack can configure Scalable TCP (STCP), IP routing, IP multipathing, or IPsec. For an example of how to configure an Oracle Solaris zone with an exclusive IP stack, see the following example zonecfg:solaris10-up9-zone set ip-type=exclusivezonecfg:solaris10-up9-zone> add netzonecfg:solaris10-up9-zone> set physical=nxge0 When the installation completes, use the zoneadm list -i -v options to list the installedzones and verify the status.Target # zoneadm list -i -vSee that the new Zone status is installedID NAME STATUS PATH BRAND IP0 global running / native shared- solaris10-up9-zone installed /zones native sharedNow boot the ZoneTarget # zoneadm -z solaris10-up9-zone bootWe need to login into the Zone order to complete the zone set up or insert a sysidcfg file beforebooting the zone for the first time see example for sysidcfg file in Appendix B: sysidcfg filesectionTarget # zlogin -C solaris10-up9-zoneTroubleshootingIf an installation fails, review the log file. On success, the log file is in /var/log inside the zone. Onfailure, the log file is in /var/tmp in the global zone.If a zone installation is interrupted or fails, the zone is left in the incomplete state. Use uninstall -F to reset the zone to the configured state.Target # zoneadm -z solaris10-up9-zone uninstall -FTarget # zonecfg -z solaris10-up9-zone delete -FConclusionOracle Solaris Zones P2V tool provides the flexibility to build pre-configuredimages with different software configuration for faster deployment and server consolidation.In this document, I demonstrated how to build and install images and to integrate the images with other Oracle Solaris features like ZFS and DTrace.Appendix A: archive identification sectionWe can use the head -n 20 /var/tmp/solaris_10_up9.flar command in order to access theidentification section that contains the detailed description.Target # head -n 20 /var/tmp/solaris_10_up9.flarFlAsH-aRcHiVe-2.0section_begin=identificationarchive_id=e4469ee97c3f30699d608b20a36011befiles_archived_method=cpiocreation_date=20100901160827creation_master=mdet5140-1content_name=s10-systemcreation_node=mdet5140-1creation_hardware_class=sun4vcreation_platform=SUNW,T5140creation_processor=sparccreation_release=5.10creation_os_name=SunOScreation_os_version=Generic_142909-16files_compressed_method=nonecontent_architectures=sun4vtype=FULLsection_end=identificationsection_begin=predeploymentbegin 755 predeployment.cpio.ZAppendix B: sysidcfg file sectionTarget # cat sysidcfgsystem_locale=Ctimezone=US/Pacificterminal=xtermssecurity_policy=NONEroot_password=HsABA7Dt/0sXXtimeserver=localhostname_service=NONEnetwork_interface=primary {hostname= solaris10-up9-zonenetmask=255.255.255.0protocol_ipv6=nodefault_route=192.168.0.1}name_service=NONEnfs4_domain=dynamicWe need to copy this file before booting the zoneTarget # cp sysidcfg /zones/solaris10-up9-zone/root/etc/

  Read the article

• How to Set Up a MongoDB NoSQL Cluster Using Oracle Solaris Zones

  - by Orgad Kimchi

  This article starts with a brief overview of MongoDB and follows with an example of setting up a MongoDB three nodes cluster using Oracle Solaris Zones. The following are benefits of using Oracle Solaris for a MongoDB cluster: • You can add new MongoDB hosts to the cluster in minutes instead of hours using the zone cloning feature. Using Oracle Solaris Zones, you can easily scale out your MongoDB cluster. • In case there is a user error or software error, the Service Management Facility ensures the high availability of each cluster member and ensures that MongoDB replication failover will occur only as a last resort. • You can discover performance issues in minutes versus days by using DTrace, which provides increased operating system observability. DTrace provides a holistic performance overview of the operating system and allows deep performance analysis through cooperation with the built-in MongoDB tools. • ZFS built-in compression provides optimized disk I/O utilization for better I/O performance. In the example presented in this article, all the MongoDB cluster building blocks will be installed using the Oracle Solaris Zones, Service Management Facility, ZFS, and network virtualization technologies. Figure 1 shows the architecture:

  Read the article

• Who keeps removing that file?

  - by mgerdts

  Over the years, I've had many times when some file gets removed and there's no obvious culprit.  With dtrace, it is somewhat easy to figure out:  #! /usr/sbin/dtrace -wqs syscall::unlinkat:entry /cleanpath(copyinstr(arg1)) == "/dev/null"/ {         stop();         printf("%s[%d] stopped before removing /dev/null\n", execname, pid);         system("ptree %d; pstack %d", pid, pid); } That script will stop the process trying to remove /dev/null before it does it.  You can allow it to continue by restarting (unstopping?) the command with prun(1) or killing it with kill -9.  If you want the command to continue automatically after getting the ptree and pstack output, you can add "; prun %d" and another pid argument to the system() call.

  Read the article

• How to track dealloc of an abstract object class (NSMutable...)

  - by Thomas Tempelmann

  I have a bug in my code caused by a premature freeing of a ObjC object. I seem not to be able to find it by just looking at my code. There's usually a simple trick to track the dealloc to any class: Implement dealloc, and set a breakpoint. This even usually works with standard objects, by subclassing them and making sure I instantiate the subclass instead of the standard class. However, this does not seem to work with NSMutableArray, and probably neither with similar NSMutable... classes. Some explanations for this can be found here: link text Any other ideas how to track the dealloc invocation of a particular class or object so that I can see the call stack? It's probably possible with DTrace. Any pointers without having to read the entire dtrace docs first?

  Read the article

• Interesting articles and blogs on SPARC T4

  - by mv

  Interesting articles and blogs on SPARC T4 processor   I have consolidated all the interesting information I could get on SPARC T4 processor and its hardware cryptographic capabilities.  Hope its useful. 1. Advantages of SPARC T4 processor  Most important points in this T4 announcement are : "The SPARC T4 processor was designed from the ground up for high speed security and has a cryptographic stream processing unit (SPU) integrated directly into each processor core. These accelerators support 16 industry standard security ciphers and enable high speed encryption at rates 3 to 5 times that of competing processors. By integrating encryption capabilities directly inside the instruction pipeline, the SPARC T4 processor eliminates the performance and cost barriers typically associated with secure computing and makes it possible to deliver high security levels without impacting the user experience." Data Sheet has more details on these  : "New on-chip Encryption Instruction Accelerators with direct non-privileged support for 16 industry-standard cryptographic algorithms plus random number generation in each of the eight cores: AES, Camellia, CRC32c, DES, 3DES, DH, DSA, ECC, Kasumi, MD5, RSA, SHA-1, SHA-224, SHA-256, SHA-384, SHA-512" I ran "isainfo -v" command on Solaris 11 Sparc T4-1 system. It shows the new instructions as expected  : $ isainfo -v 64-bit sparcv9 applications crc32c cbcond pause mont mpmul sha512 sha256 sha1 md5 camellia kasumi des aes ima hpc vis3 fmaf asi_blk_init vis2 vis popc 32-bit sparc applications crc32c cbcond pause mont mpmul sha512 sha256 sha1 md5 camellia kasumi des aes ima hpc vis3 fmaf asi_blk_init vis2 vis popc v8plus div32 mul32  2.  Dan Anderson's Blog have some interesting points about how these can be used : "New T4 crypto instructions include: aes_kexpand0, aes_kexpand1, aes_kexpand2,         aes_eround01, aes_eround23, aes_eround01_l, aes_eround_23_l, aes_dround01, aes_dround23, aes_dround01_l, aes_dround_23_l.       Having SPARC T4 hardware crypto instructions is all well and good, but how do we access it ?      The software is available with Solaris 11 and is used automatically if you are running Solaris a SPARC T4.  It is used internally in the kernel through kernel crypto modules.  It is available in user space through the PKCS#11 library." 3.   Dans' Blog on Where's the Crypto Libraries? Although this was written in 2009 but still is very useful  "Here's a brief tour of the major crypto libraries shown in the digraph:   The libpkcs11 library contains the PKCS#11 API (C_\*() functions, such as C_Initialize()). That in turn calls library pkcs11_softtoken or pkcs11_kernel, for userland or kernel crypto providers. The latter is used mostly for hardware-assisted cryptography (such as n2cp for Niagara2 SPARC processors), as that is performed more efficiently in kernel space with the "kCF" module (Kernel Crypto Framework). Additionally, for Solaris 10, strong crypto algorithms were split off in separate libraries, pkcs11_softtoken_extra libcryptoutil contains low-level utility functions to help implement cryptography. libsoftcrypto (OpenSolaris and Solaris Nevada only) implements several symmetric-key crypto algorithms in software, such as AES, RC4, and DES3, and the bignum library (used for RSA). libmd implements MD5, SHA, and SHA2 message digest algorithms" 4. Difference in T3 and T4 Diagram in this blog is good and self explanatory. Jeff's blog also highlights the differences  "The T4 servers have improved crypto acceleration, described at https://blogs.oracle.com/DanX/entry/sparc_t4_openssl_engine. It is "just built in" so administrators no longer have to assign crypto accelerator units to domains - it "just happens". Every physical or virtual CPU on a SPARC-T4 has full access to hardware based crypto acceleration at all times. .... For completeness sake, it's worth noting that the T4 adds more crypto algorithms, and accelerates Camelia, CRC32c, and more SHA-x." 5. About performance counters In this blog, performance counters are explained : "Note that unlike T3 and before, T4 crypto doesn't require kernel modules like ncp or n2cp, there is no visibility of crypto hardware with kstats or cryptoadm. T4 does provide hardware counters for crypto operations.  You can see these using cpustat: cpustat -c pic0=Instr_FGU_crypto 5 You can check the general crypto support of the hardware and OS with the command "isainfo -v". Since T4 crypto's implementation now allows direct userland access, there are no "crypto units" visible to cryptoadm.  " For more details refer Martin's blog as well. 6. How to turn off  SPARC T4 or Intel AES-NI crypto acceleration  I found this interesting blog from Darren about how to turn off  SPARC T4 or Intel AES-NI crypto acceleration. "One of the new Solaris 11 features of the linker/loader is the ability to have a single ELF object that has multiple different implementations of the same functions that are selected at runtime based on the capabilities of the machine.   The alternate to this is having the application coded to call getisax(2) system call and make the choice itself.  We use this functionality of the linker/loader when we build the userland libraries for the Solaris Cryptographic Framework (specifically libmd.so and libsoftcrypto.so) The Solaris linker/loader allows control of a lot of its functionality via environment variables, we can use that to control the version of the cryptographic functions we run.  To do this we simply export the LD_HWCAP environment variable with values that tell ld.so.1 to not select the HWCAP section matching certain features even if isainfo says they are present.  This will work for consumers of the Solaris Cryptographic Framework that use the Solaris PKCS#11 libraries or use libmd.so interfaces directly.  For SPARC T4 : export LD_HWCAP="-aes -des -md5 -sha256 -sha512 -mont -mpul" .. For Intel systems with AES-NI support: export LD_HWCAP="-aes"" Note that LD_HWCAP is explained in  http://docs.oracle.com/cd/E23823_01/html/816-5165/ld.so.1-1.html "LD_HWCAP, LD_HWCAP_32, and LD_HWCAP_64 -  Identifies an alternative hardware capabilities value... A “-” prefix results in the capabilities that follow being removed from the alternative capabilities." 7. Whitepaper on SPARC T4 Servers—Optimized for End-to-End Data Center Computing This Whitepaper on SPARC T4 Servers—Optimized for End-to-End Data Center Computing explains more details.  It has DTrace scripts which may come in handy : "To ensure the hardware-assisted cryptographic acceleration is configured to use and working with the security scenarios, it is recommended to use the following Solaris DTrace script. #!/usr/sbin/dtrace -s pid$1:libsoftcrypto:yf*:entry, pid$target:libsoftcrypto:rsa*:entry, pid$1:libmd:yf*:entry { @[probefunc] = count(); } tick-1sec { printa(@ops); trunc(@ops); }" Note that I have slightly modified the D Script to have RSA "libsoftcrypto:rsa*:entry" as well as per recommendations from Chi-Chang Lin. 8. References http://www.oracle.com/us/corporate/features/sparc-t4-announcement-494846.html http://www.oracle.com/us/products/servers-storage/servers/sparc-enterprise/t-series/sparc-t4-1-ds-487858.pdf https://blogs.oracle.com/DanX/entry/sparc_t4_openssl_engine https://blogs.oracle.com/DanX/entry/where_s_the_crypto_libraries https://blogs.oracle.com/darren/entry/howto_turn_off_sparc_t4 http://docs.oracle.com/cd/E23823_01/html/816-5165/ld.so.1-1.html   https://blogs.oracle.com/hardware/entry/unleash_the_power_of_cryptography https://blogs.oracle.com/cmt/entry/t4_crypto_cheat_sheet https://blogs.oracle.com/martinm/entry/t4_performance_counters_explained  https://blogs.oracle.com/jsavit/entry/no_mau_required_on_a http://www.oracle.com/us/products/servers-storage/servers/sparc-enterprise/t-series/sparc-t4-business-wp-524472.pdf

  Read the article

• Configuring Oracle iPlanet WebServer / Oracle Traffic Director to use crypto accelerators on T4-1 servers

  - by mv

  Configuring Oracle iPlanet Web Server / Oracle Traffic Director to use crypto accelerators on T4-1 servers Jyri had written a technical article on Configuring Solaris Cryptographic Framework and Sun Java System Web Server 7 on Systems With UltraSPARC T1 Processors. I tried to find out what has changed since then in T4. I have used a T4-1 SPARC system with Solaris 10. Results slightly vary for Solaris 11.  For Solaris 11, the T4 optimization was implemented in libsoftcrypto.so while it was in pkcs11_softtoken_extra.so for Solaris 10. Overview of T4 processors is here in this blog. Many thanx to Chi-Chang Lin and Julien for their help. 1. Install Oracle iPlanet Web Server / Oracle Traffic Director.  Go to instance/config directory.  # cd /opt/oracle/webserver7/https-hostname.fqdn/config 2. List default PKCS#11 Modules # ../../bin/modutil -dbdir . -listListing of PKCS #11 Modules-----------------------------------------------------------1. NSS Internal PKCS #11 Moduleslots: 2 slots attachedstatus: loadedslot: NSS Internal Cryptographic Servicestoken: NSS Generic Crypto Servicesslot: NSS User Private Key and Certificate Servicestoken: NSS Certificate DB2. Root Certslibrary name: libnssckbi.soslots: 1 slot attachedstatus: loadedslot: NSS Builtin Objectstoken: Builtin Object Token----------------------------------------------------------- 3. Initialize the soft token data store in the $HOME/.sunw/pkcs11_softtoken/ directory # pktool setpin keystore=pkcs11Enter token passphrase: olderpasswordCreate new passphrase: passwordRe-enter new passphrase: passwordPassphrase changed. 4. Offload crypto operations to Solaris Crypto Framework on T4 $ ../../bin/modutil -dbdir . -nocertdb -add SCF -libfile /usr/lib/libpkcs11.so -mechanisms RSA:AES:SHA1:MD5 Module "SCF" added to database. Note that -nocertdb means modutil won't try to open the NSS softoken key database. It doesn't even have to be present. PKCS#11 library used is /usr/lib/libpkcs11.so. If the server is running in 64 bit mode, we have to use /usr/lib/64/libpkcs11.so Unlike T1 and T2, in T4 we do not have to disable mechanisms in softtoken provider using cryptoadm. 5. List again to check that a new module SCF is added # ../../bin/modutil -dbdir . -list Listing of PKCS #11 Modules-----------------------------------------------------------1. NSS Internal PKCS #11 Moduleslots: 2 slots attachedstatus: loadedslot: NSS Internal Cryptographic Servicestoken: NSS Generic Crypto Servicesslot: NSS User Private Key and Certificate Servicestoken: NSS Certificate DB2. SCFlibrary name: /usr/lib/libpkcs11.soslots: 2 slots attachedstatus: loadedslot: Sun Metaslottoken: Sun Metaslotslot: n2rng/0 SUNW_N2_Random_Number_Generator token: n2rng/0 SUNW_N2_RNG 3. Root Certs library name: libnssckbi.so slots: 1 slot attached status: loaded slot: NSS Builtin Objects token: Builtin Object Token----------------------------------------------------------- 6.  Create certificate in “Sun Metaslot” : I have used certutil, but you must use Admin Server CLI / GUI # ../../bin/certutil -S -x -n "Server-Cert" -t "CT,CT,CT" -s "CN=*.fqdn" -d . -h "Sun Metaslot"Enter Password or Pin for "Sun Metaslot": password 7. Verify that the certificate is created properly in “Sun Metslaot” # ../../bin/certutil -L -d . -h "Sun Metaslot"Certificate Nickname Trust AttributesSSL,S/MIME,JAR/XPIEnter Password or Pin for "Sun Metaslot": passwordSun Metaslot:Server-Cert CTu,Cu,Cu# 8. Associate this newly created certificate to http listener using Admin CLI/GUI. After that server.xml should have <http-listener> ...    <ssl>        <server-cert-nickname>Sun Metaslot:Server-Cert</server-cert-nicknamer>    </ssl> Note the prefix "Sun Metaslot" 9. Disable PKCS#11 bypass To use the accelerated AES algorithm, turn off PKCS#11 bypass, and configure modutil to have the AES mechanism go to the Metaslot. After you disable PKCS#11 bypasss using Admin GUI/CLI,  check that server.xml should have <server> ....    <pkcs11>         <enabled>1</enabled>         <allow-bypass>0</allow-bypass>     </pkcs11> With PKCS#11 bypass enabled, Oracle iPlanet Web Server will only use the RSA capability of the T4, provided certificate and key are stored in the T4 slot (Metaslot). Actually, the RSA op is never bypassed in NSS, it's always done with PKCS#11 calls. So the bypass settings won't affect the behavior of the probes for RSA at all. The only thing that matters if where the RSA key and certificate live, ie. which PKCS#11 token, and thus which PKCS#11 module gets called to do the work. If your certificate/key are in the NSS certificate/key db, you will see libsoftokn3/libfreebl libraries doing the RSA work. If they are in the Sun Metaslot, it should be the Solaris code. 10. Start the server instance # ../bin/startserv Oracle iPlanet Web Server 7.0.16 B09/14/2012 03:33Please enter the PIN for the "Sun Metaslot" token: password...info: HTTP3072: http-listener-1: https://hostname.fqdn:80 ready to accept requestsinfo: CORE3274: successful server startup 11. Figure out which process to run this DTrace script on # ps -eaf | grep webservd | grep -v dogwebservd 18224 18223 0 13:17:25 ? 0:07 webservd -d /opt/oracle/webserver7/https-hostname.fqdn/config -r /opt/root 18225 18224 0 13:17:25 ? 0:00 webservd -d /opt/oracle/webserver7/https-hostname.fqdn/config -r /opt/ (For Oracle Traffic Director look for process named "trafficd") We see that the child process id is “18225” 12. Clients for testing : You can use any browser. I used NSS tool tstclnt for testing $cat > req.txtGET /index.html HTTP/1.0 For checking both RSA and AES, I used cipher “:0035” which is TLS_RSA_WITH_AES_256_CBC_SHA $./tstclnt -h hostname -p 80 -d . -T -f -o -v -c “:0035” < req.txt 13. How do I make sure that crypto accelerator is being used 13.1 Create DTrace script The following D script should be able to uncover whether T4-specific crypto routine are being called or not. It also displays stats per second. # cat > t4crypto.d#!/usr/sbin/dtrace -spid$target::*rsa*:entry,pid$target::*yf*:entry{    @ops[probemod, probefunc] = count();}tick-1sec{    printa(@ops);    trunc(@ops);} Invoke with './t4crypto.d -p <pid> ' 13.2 EXPECTED PROBES FOR Solaris 10 : If offloading to T4 HW are correctly set up, the expected DTrace output would have these probes and libraries library Operations PROBES pkcs11_softtoken_extra.so RSA soft_decrypt_rsa_pkcs_decode, soft_encrypt_rsa_pkcs_encode soft_rsa_crypt_init_common soft_rsa_decrypt, soft_rsa_encrypt soft_rsa_decrypt_common, soft_rsa_encrypt_common AES yf_aes_instructions_present yf_aes_expand256, yf_aes256_cbc_decrypt, yf_aes256_cbc_encrypt, yf_aes256_load_keys_for_decrypt, yf_aes256_load_keys_for_encrypt, Note that these are for 256, same for 128, 192... these are for cbc, same for ecb, ctr, cfb128... DES yf_des_expand, yf_des_instructions_present yf_des_encrypt libmd_psr.so MD5 yf_md5_multiblock, yf_md5_instruction_present SHA1 yf_sha1_instruction_present, yf_sha1_multibloc 13.3 SAMPLE OUTPUT FOR CIPHER TLS_RSA_WITH_AES_256_CBC_SHA (0x0035) ON T4 SPARC SOLARIS 10 WITHOUT PKCS#11 BYPASS # ./t4crypto.d -p 18225 pkcs11_softtoken_extra.so.1   soft_decrypt_rsa_pkcs_decode    1 pkcs11_softtoken_extra.so.1   soft_rsa_crypt_init_common      1 pkcs11_softtoken_extra.so.1   soft_rsa_decrypt                1 pkcs11_softtoken_extra.so.1   big_mp_mul_yf                   2 pkcs11_softtoken_extra.so.1   mpm_yf_mpmul                    2 pkcs11_softtoken_extra.so.1   mpmul_arr_yf                    2 pkcs11_softtoken_extra.so.1   rijndael_key_setup_enc_yf       2 pkcs11_softtoken_extra.so.1   soft_rsa_decrypt_common         2 pkcs11_softtoken_extra.so.1   yf_aes_expand256                2 pkcs11_softtoken_extra.so.1   yf_aes256_cbc_decrypt           3 pkcs11_softtoken_extra.so.1   yf_aes256_load_keys_for_decrypt 3 pkcs11_softtoken_extra.so.1   big_mont_mul_yf                 6 pkcs11_softtoken_extra.so.1   mm_yf_montmul                   6 pkcs11_softtoken_extra.so.1   yf_des_instructions_present     6 pkcs11_softtoken_extra.so.1   yf_aes256_cbc_encrypt           8 pkcs11_softtoken_extra.so.1   yf_aes256_load_keys_for_encrypt 8 pkcs11_softtoken_extra.so.1   yf_mpmul_present                8 pkcs11_softtoken_extra.so.1   yf_aes_instructions_present    13 pkcs11_softtoken_extra.so.1   yf_des_encrypt                 18 libmd_psr.so.1                yf_md5_multiblock              41 libmd_psr.so.1                yf_md5_instruction_present     72 libmd_psr.so.1                yf_sha1_instruction_present    82 libmd_psr.so.1                yf_sha1_multiblock             82 This indicates that both RSA and AES ops are done in Solaris Crypto Framework. 13.4 SAMPLE OUTPUT FOR CIPHER TLS_RSA_WITH_AES_256_CBC_SHA (0x0035) ON T4 SPARC SOLARIS 10 WITH PKCS#11 BYPASS # ./t4crypto.d -p 18225 pkcs11_softtoken_extra.so.1   soft_decrypt_rsa_pkcs_decode 1 pkcs11_softtoken_extra.so.1   soft_rsa_crypt_init_common   1 pkcs11_softtoken_extra.so.1   soft_rsa_decrypt             1 pkcs11_softtoken_extra.so.1   soft_rsa_decrypt_common      1 pkcs11_softtoken_extra.so.1   big_mp_mul_yf                2 pkcs11_softtoken_extra.so.1   mpm_yf_mpmul                 2 pkcs11_softtoken_extra.so.1   mpmul_arr_yf                 2 pkcs11_softtoken_extra.so.1   big_mont_mul_yf              6 pkcs11_softtoken_extra.so.1   mm_yf_montmul                6 pkcs11_softtoken_extra.so.1   yf_mpmul_present             8 For this cipher, when I enable PKCS#11 bypass, Only RSA probes are being hit AES probes are not being hit. 13.5 ustack() for RSA operations / probefunc == "soft_rsa_decrypt" / Shows that libnss3.so is calling C_* functions of libpkcs11.so which is calling functions of pkcs11_softtoken_extra.so for both cases with and without bypass. When PKCS#11 bypass is disabled (allow-bypass is 0) pkcs11_softtoken_extra.so.1`soft_rsa_decrypt pkcs11_softtoken_extra.so.1`soft_rsa_decrypt_common+0x94 pkcs11_softtoken_extra.so.1`soft_unwrapkey+0x258 pkcs11_softtoken_extra.so.1`C_UnwrapKey+0x1ec libpkcs11.so.1`meta_unwrap_key+0x17c libpkcs11.so.1`meta_UnwrapKey+0xc4 libpkcs11.so.1`C_UnwrapKey+0xfc libnss3.so`pk11_AnyUnwrapKey+0x6b8 libnss3.so`PK11_PubUnwrapSymKey+0x8c libssl3.so`ssl3_HandleRSAClientKeyExchange+0x1a0 libssl3.so`ssl3_HandleClientKeyExchange+0x154 libssl3.so`ssl3_HandleHandshakeMessage+0x440 libssl3.so`ssl3_HandleHandshake+0x11c libssl3.so`ssl3_HandleRecord+0x5e8 libssl3.so`ssl3_GatherCompleteHandshake+0x5c libssl3.so`ssl_GatherRecord1stHandshake+0x30 libssl3.so`ssl_Do1stHandshake+0xec libssl3.so`ssl_SecureRecv+0x1c8 libssl3.so`ssl_Recv+0x9c libns-httpd40.so`__1cNDaemonSessionDrun6M_v_+0x2dc When PKCS#11 bypass is enabled (allow-bypass is 1) pkcs11_softtoken_extra.so.1`soft_rsa_decrypt pkcs11_softtoken_extra.so.1`soft_rsa_decrypt_common+0x94 pkcs11_softtoken_extra.so.1`C_Decrypt+0x164 libpkcs11.so.1`meta_do_operation+0x27c libpkcs11.so.1`meta_Decrypt+0x4c libpkcs11.so.1`C_Decrypt+0xcc libnss3.so`PK11_PrivDecryptPKCS1+0x1ac libssl3.so`ssl3_HandleRSAClientKeyExchange+0xe4 libssl3.so`ssl3_HandleClientKeyExchange+0x154 libssl3.so`ssl3_HandleHandshakeMessage+0x440 libssl3.so`ssl3_HandleHandshake+0x11c libssl3.so`ssl3_HandleRecord+0x5e8 libssl3.so`ssl3_GatherCompleteHandshake+0x5c libssl3.so`ssl_GatherRecord1stHandshake+0x30 libssl3.so`ssl_Do1stHandshake+0xec libssl3.so`ssl_SecureRecv+0x1c8 libssl3.so`ssl_Recv+0x9c libns-httpd40.so`__1cNDaemonSessionDrun6M_v_+0x2dc libnsprwrap.so`ThreadMain+0x1c libnspr4.so`_pt_root+0xe8 13.6 ustack() FOR AES operations / probefunc == "yf_aes256_cbc_encrypt" / When PKCS#11 bypass is disabled (allow-bypass is 0) pkcs11_softtoken_extra.so.1`yf_aes256_cbc_encrypt pkcs11_softtoken_extra.so.1`aes_block_process_contiguous_whole_blocks+0xb4 pkcs11_softtoken_extra.so.1`aes_crypt_contiguous_blocks+0x1cc pkcs11_softtoken_extra.so.1`soft_aes_encrypt_common+0x22c pkcs11_softtoken_extra.so.1`C_EncryptUpdate+0x10c libpkcs11.so.1`meta_do_operation+0x1fc libpkcs11.so.1`meta_EncryptUpdate+0x4c libpkcs11.so.1`C_EncryptUpdate+0xcc libnss3.so`PK11_CipherOp+0x1a0 libssl3.so`ssl3_CompressMACEncryptRecord+0x264 libssl3.so`ssl3_SendRecord+0x300 libssl3.so`ssl3_FlushHandshake+0x54 libssl3.so`ssl3_SendFinished+0x1fc libssl3.so`ssl3_HandleFinished+0x314 libssl3.so`ssl3_HandleHandshakeMessage+0x4ac libssl3.so`ssl3_HandleHandshake+0x11c libssl3.so`ssl3_HandleRecord+0x5e8 libssl3.so`ssl3_GatherCompleteHandshake+0x5c libssl3.so`ssl_GatherRecord1stHandshake+0x30 libssl3.so`ssl_Do1stHandshake+0xec Shows that libnss3.so is calling C_* functions of libpkcs11.so which is calling functions of pkcs11_softtoken_extra.so However when PKCS#11 bypass is disabled (allow-bypass is 1) this stack isn't getting called. 14. LIST OF ALL THE PROBES MATCHED BY D SCRIPT FOR REFERENCE # ./t4crypto.d -p 18225 -l ID PROVIDER MODULE FUNCTION NAME ... 55720 pid18225 libmd_psr.so.1 yf_md5_instruction_present entry 55721 pid18225 libmd_psr.so.1 yf_sha256_instruction_present entry 55722 pid18225 libmd_psr.so.1 yf_sha512_instruction_present entry 55723 pid18225 libmd_psr.so.1 yf_sha1_instruction_present entry 55724 pid18225 libmd_psr.so.1 yf_sha256 entry 55725 pid18225 libmd_psr.so.1 yf_sha256_multiblock entry 55726 pid18225 libmd_psr.so.1 yf_sha512 entry 55727 pid18225 libmd_psr.so.1 yf_sha512_multiblock entry 55728 pid18225 libmd_psr.so.1 yf_sha1 entry 55729 pid18225 libmd_psr.so.1 yf_sha1_multiblock entry 55730 pid18225 libmd_psr.so.1 yf_md5 entry 55731 pid18225 libmd_psr.so.1 yf_md5_multiblock entry 55732 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_instructions_present entry 55733 pid18225 pkcs11_softtoken_extra.so.1 rijndael_key_setup_enc_yf entry 55734 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_expand128 entry 55735 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_encrypt128 entry 55736 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_decrypt128 entry 55737 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_expand192 entry 55738 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_encrypt192 entry 55739 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_decrypt192 entry 55740 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_expand256 entry 55741 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_encrypt256 entry 55742 pid18225 pkcs11_softtoken_extra.so.1 yf_aes_decrypt256 entry 55743 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_load_keys_for_encrypt entry 55744 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_load_keys_for_encrypt entry 55745 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_load_keys_for_encrypt entry 55746 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_ecb_encrypt entry 55747 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_ecb_encrypt entry 55748 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_ecb_encrypt entry 55749 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_cbc_encrypt entry 55750 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_cbc_encrypt entry 55751 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_cbc_encrypt entry 55752 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_ctr_crypt entry 55753 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_ctr_crypt entry 55754 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_ctr_crypt entry 55755 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_cfb128_encrypt entry 55756 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_cfb128_encrypt entry 55757 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_cfb128_encrypt entry 55758 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_load_keys_for_decrypt entry 55759 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_load_keys_for_decrypt entry 55760 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_load_keys_for_decrypt entry 55761 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_ecb_decrypt entry 55762 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_ecb_decrypt entry 55763 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_ecb_decrypt entry 55764 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_cbc_decrypt entry 55765 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_cbc_decrypt entry 55766 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_cbc_decrypt entry 55767 pid18225 pkcs11_softtoken_extra.so.1 yf_aes128_cfb128_decrypt entry 55768 pid18225 pkcs11_softtoken_extra.so.1 yf_aes192_cfb128_decrypt entry 55769 pid18225 pkcs11_softtoken_extra.so.1 yf_aes256_cfb128_decrypt entry 55771 pid18225 pkcs11_softtoken_extra.so.1 yf_des_instructions_present entry 55772 pid18225 pkcs11_softtoken_extra.so.1 yf_des_expand entry 55773 pid18225 pkcs11_softtoken_extra.so.1 yf_des_encrypt entry 55774 pid18225 pkcs11_softtoken_extra.so.1 yf_mpmul_present entry 55775 pid18225 pkcs11_softtoken_extra.so.1 yf_montmul_present entry 55776 pid18225 pkcs11_softtoken_extra.so.1 mm_yf_montmul entry 55777 pid18225 pkcs11_softtoken_extra.so.1 mm_yf_montsqr entry 55778 pid18225 pkcs11_softtoken_extra.so.1 mm_yf_restore_func entry 55779 pid18225 pkcs11_softtoken_extra.so.1 mm_yf_ret_from_mont_func entry 55780 pid18225 pkcs11_softtoken_extra.so.1 mm_yf_execute_slp entry 55781 pid18225 pkcs11_softtoken_extra.so.1 big_modexp_ncp_yf entry 55782 pid18225 pkcs11_softtoken_extra.so.1 big_mont_mul_yf entry 55783 pid18225 pkcs11_softtoken_extra.so.1 mpmul_arr_yf entry 55784 pid18225 pkcs11_softtoken_extra.so.1 big_mp_mul_yf entry 55785 pid18225 pkcs11_softtoken_extra.so.1 mpm_yf_mpmul entry 55786 pid18225 libns-httpd40.so nsapi_rsa_set_priv_fn entry ... 55795 pid18225 libnss3.so prepare_rsa_priv_key_export_for_asn1 entry 55796 pid18225 libresolv.so.2 sunw_dst_rsaref_init entry 55797 pid18225 libnssutil3.so NSS_Get_SEC_UniversalStringTemplate entry ... 55813 pid18225 libsoftokn3.so prepare_low_rsa_priv_key_for_asn1 entry 55814 pid18225 libsoftokn3.so rsa_FormatOneBlock entry 55815 pid18225 libsoftokn3.so rsa_FormatBlock entry 55816 pid18225 libnssdbm3.so lg_prepare_low_rsa_priv_key_for_asn1 entry 55817 pid18225 libfreebl_32fpu_3.so rsa_build_from_primes entry 55818 pid18225 libfreebl_32fpu_3.so rsa_is_prime entry 55819 pid18225 libfreebl_32fpu_3.so rsa_get_primes_from_exponents entry 55820 pid18225 libfreebl_32fpu_3.so rsa_PrivateKeyOpNoCRT entry 55821 pid18225 libfreebl_32fpu_3.so rsa_PrivateKeyOpCRTNoCheck entry 55822 pid18225 libfreebl_32fpu_3.so rsa_PrivateKeyOpCRTCheckedPubKey entry 55823 pid18225 pkcs11_kernel.so.1 key_gen_rsa_by_value entry 55824 pid18225 pkcs11_kernel.so.1 get_rsa_private_key entry 55825 pid18225 pkcs11_kernel.so.1 get_rsa_public_key entry 55826 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_encrypt entry 55827 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_decrypt entry 55828 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_crypt_init_common entry 55829 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_encrypt_common entry 55830 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_decrypt_common entry 55831 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_sign_verify_init_common entry 55832 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_sign_common entry 55833 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_verify_common entry 55834 pid18225 pkcs11_softtoken_extra.so.1 generate_rsa_key entry 55835 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_genkey_pair entry 55836 pid18225 pkcs11_softtoken_extra.so.1 get_rsa_sha1_prefix entry 55837 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_digest_sign_common entry 55838 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_digest_verify_common entry 55839 pid18225 pkcs11_softtoken_extra.so.1 soft_rsa_verify_recover entry 55840 pid18225 pkcs11_softtoken_extra.so.1 rsa_pri_to_asn1 entry 55841 pid18225 pkcs11_softtoken_extra.so.1 asn1_to_rsa_pri entry 55842 pid18225 pkcs11_softtoken_extra.so.1 soft_encrypt_rsa_pkcs_encode entry 55843 pid18225 pkcs11_softtoken_extra.so.1 soft_decrypt_rsa_pkcs_decode entry 55844 pid18225 pkcs11_softtoken_extra.so.1 soft_sign_rsa_pkcs_encode entry 55845 pid18225 pkcs11_softtoken_extra.so.1 soft_verify_rsa_pkcs_decode entry 55770 profile tick-1sec

  Read the article

• New: ZFS Storage Appliance Videos

  - by Roxana Babiciu

  Check out part one of a new video series for ZFS Storage Appliance. In video #1, you’ll learn about the advantages built into Oracle’s ZS3 Storage Appliance that come from the unique position that Oracle holds in the market. In video #2, you’ll learn how best to monitor large ZS3 installations as well as the use of Enterprise Manager as a complement to dtrace analytics at the ZFS Storage Appliance device level.

  Read the article

• Oracle 12c: First (and best!) on Solaris

  - by mgerdts

  Oracle 12c is now available for download.  Notice that support for Solaris SPARC and x86-64 are among the operating systems supported on the first day of availability. New database features that relate to Solaris include: I/O outlier support.  This is made possible through the database's use of DTrace, and as such Solaris has a clear edge here. Oracle ACFS Replication and Tagging for Solaris Integration with Solaris resource pools As has been the case for some time, Oracle databases are supported in zones.

  Read the article

• Down to the Wire - Yet More Solaris Things to See at OpenWorld (and JavaOne!)

  - by Larry Wake

  San Francisco is bracing for the annual invasion. The airport's jammed, the tweets are flying, and the numbers are crazy: more than 50,000 attendees and 2,500+ sessions, taking over Moscone Convention Center, two streets, Union Square, and seemingly every hotel in town (98,000 hotel room nights). So yeah, it's busy. And it's not just OpenWorld--we've also got JavaOne, MySQL Connect, and four other sub-events going on as well. Speaking of JavaOne, you can find Solaris-related activity there, too -- I've highlighted one hands-on lab below. Here's a last pre-event roundup of activities for consideration; enjoy the show(s)! (Remember, Schedule Builder is your friend; use it with the session numbers below to register.) Monday, October 1st: 3:15 PM - General Session: Accelerate Your Business with the Oracle Hardware Advantage(GEN9691, Moscone North Hall D) John Fowler, head of Oracle's Systems organization, will talk about Oracle hardware technology and how it's co-engineered with other key technologies, including Oracle Solaris. Tuesday, October 2nd: 10:15 AM - Building an IaaS Platform with SPARC, Oracle Solaris 11, and Oracle VM Server for SPARC(CON4431, Moscone South 270)Get the birds-eye lowdown (whatever that means) on how U.S. Cellular  built its Infrastructure as a Service (IaaS) cloud delivery platform with Oracle’s SPARC T4 servers, Oracle Solaris 11, Oracle Solaris Cluster 4, and Oracle VM Server for SPARC. The session covers the high-level design, business case made, implementation details, and lessons learned. 11:45 AM - Oracle Solaris 11 Panel: Insights and Directions from Oracle Solaris Core Engineering(CON8790, Moscone South 252) This has been one of the livelier Solaris-related sessions in years past (and I'm not saying that just because I get to moderate it this year). A panel of core engineers responsible for a wide range of key Solaris technologies will talk about some of the interesting work they've been doing -- but mostly we keep time open for the panel to take questions from attendees, because that's the fun part. Wednesday, October 3rd: 10:00 AM - Tracing Your Java Application Tuning on Oracle Solaris with DTrace(HOL10214, Hilton San Francisco, Franciscan A/B/C/D) This JavaOne hands-on lab will show how to use the DTrace framework to dynamically trace your Java applications on Oracle Solaris and uncover new tuning opportunities. Thursday, October 4th: 12:45 PM - Oracle Solaris 11: Optimized for Oracle Database, Oracle WebLogic Server, and Java(CON8800, Moscone South 252) Explore how Oracle Solaris 11 has been built to be the best platform for the cloud and enterprise applications, with built-in optimizations to improve performance and deliver unique functionality with Oracle Database, Oracle WebLogic Server, and Java.

  Read the article

• ????????????????? Oracle Solaris ??? - Solaris 11 ????

  - by kazun

  ???????????????? OS ????????????????????? 20 ??????????????????? Solaris?????????????????????? OS????????????????????????Solaris ???????????????????? ??????Solaris ?????????????? 4 ????????????????????? Oracle Solaris 11 ??????????????? [????????????]???? ?(????????????? ???)???? ?(??????????????)???? ?(????????)???? ?(?????????????????)(50??) Oracle Solaris 11 ??????? ???:Oracle Solaris 11??????? Solaris ??????????????????????????????????????????????Oracle Solaris 10 ?????????????????Oracle Solaris 11 ???Oracle Solaris ???????????????????????????????????????????????????????????????????????????????????Oracle Solaris ??????????????????????????Oracle Solaris ????????????????????????????????????????? ??:??????????????????????????????????????????????????????????????????????????????????Oracle Solaris 11 ????????????????????????????????????????????????????????????????????????????????????????????????Oracle Solaris 11 ????????????? ???:Oracle Solaris 11 ????????????????????????????????????????????????????????????Oracle Solaris ??????????????????????Solaris ????????????????????????Oracle Solaris 11 ?????????????????????????????????????????????????????????????????????? Oracle Solaris 11 ??????????????? ??:???????????????????ZFS?????????????Oracle Solaris ???????????????????????????????????????IPS???????????????????????? ??:?????????????????????? OS ??????????????????????????????????????????????????????????????????? OS ????? Oracle Solaris 11 ??? ??:????????????????????????????????????????????????? Oracle Solaris 11 ???????????????????????????????????????????????????????? ??:????????????????????Solaris Trusted Extensions ? Solaris ????????????Privilege ????????????????????????????????Solaris 10 ???????????????????? ??:UNIX ?????????????????????????????????????Oracle Solaris 11 ??????????????????????????????????????????????????ZFS ??????????????????????????????????????????????????Oracle Solaris ?????????????????????????????????? ???:Oracle Solaris ????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????Oracle Solaris 11 ??????????????????????????????????????????????????? ?????????????Oracle Solaris 11 ??:Oracle Solaris ?????????????????????????????????????????????????????????????????? ZFS ?????????????????????????????????????????????? ??:?????????????????????????????????????????????????????????????????????Oracle Solaris ??????????????????????? ??:DTrace ????????????????????????????????????????????????????????????? ??:???????????????????????????????????????????????????????????????????????????????????????? ???:?????????????????????????????????????????????????????????????????????????Solaris ???????????????????????????????????????????????????????????????????????????? ??:?????????????????? Solaris 10 ?????????????????????????????????????? ???:ISV ??????????????????????????????? ????????????????? T4 ??????????????? T ?????? 5 ???????????????????????????????????? CPU ?????????????????? OS ? Solaris ??????????????????????????????????????????????? ??:???????????????? OS ???? Solaris ???????????????????????????????????????????????????? Oracle Solaris ????????? ??:???????????????????????????????????????????????????????????????????????????????????????? ZFS ?????????????????????????ZFS ??????????????????????????????????????? ???:Linux ??????????????????????????Solaris 11 ?????????????????????? GNU ????????????????????????????Solaris ?????????(Oracle Solaris Studio)????????????? Oracle Solaris Studio ?Linux??????????????Linux ??????????? Solaris ????????????????Linux ? Oracle Solaris Studio ?????????????????????????? Solaris ??????????????????? ??:??????????????????????????????????????????????????????????????????????????????????????????????????????????????????BCP(??????)??????????????????? ??:???????? Solaris 11 ?????? IA Server ?????????????????????????????????????????????…???????????Windows ????????????????????? Oracle Virtual Box ????????? Oracle Solaris 11 ??????????? ??:???Oracle Solaris ????????(?)??????????????ZFS???DTrace?????????????????????????????????????Solaris ?????????????????????????????????????????????????????????????????????????????Solaris ????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????? ??:??? 3 ??????????????????????????????????????Oracle Solaris 11 ???????????? ??:???????????????????? OS ??????????????????????????Oracle Solaris 11 ???????????????????????????????????????????????????????????????? ???:???????????? Solaris ???????????????Solaris 10 ???? 6 ??????????????????????????????????Oracle Solaris ?????????????????????????????? ?????? Oracle Solaris 10 Oracle Solaris 11

  Read the article

• How can I change how OS X's 'say' command pronounces a word?

  - by jwhitlock

  OS X's say command is useful for some tasks (such as Skype's 'notify me when a contact comes online), but it is pronouncing some names incorrectly. Is there a way to teach say to pronounce a word differently? For example, try: say "Hi, Joel Spolsky" The 'ol' sounds like 'ball' rather than 'old'. I'd like to add an exception that say "Pronounce Spolsky like this", rather than try to teach new linguistic rules. I bet there is a way since it can pronounce "iphone" as Apple wants. Update - After some research, here's what I've learned: Text-to-speech is split between turning the text to phonemes, and then the phonemes are turned into audio using a voice. Changing the voice doesn't effect the phonemes. The Speech Synthesis Manager has some functions for turning text to phonemes, and a method for registering a speech dictionary that will add new text-phoneme maps. However, Apple's speech dictionary must be in a binary form - I didn't find any plist XML. Using dtrace while running say, I found some interesting files opened in /System/Library/PrivateFrameworks/SpeechDictionary.framework/Resources. This is probably the speech dictionary, but they are all binary, except for Homophones, which is XML. Adding entries to Homophones does nothing - it is probably used in speech-to-text. They are also code signed by Apple - changing them may prevent some programs from working. PrefixDictionary CartNames CartLite SymbolDictionary Homophones There are ways to add text versions of application interface elements so VoiceOver works, a lot of which a developer gets for free, but there are tricky bits. The standard here appears to be to use a phonetic spelling as needed. My guesses are: say is a light layer of code on top of the Speech Synthesis Manager. It would be easy for the Apple devs to add a command line option to take the path to a speech dictionary plist for alternate phoneme mapping, but they didn't. It may be a useful open-source project to write a better say. Skype probably uses Speech Synthesis Manager directly, leaving no hooks to change the way my friend's names are pronounced, other than spelling them phonetically, which is silly. The easiest way to make a command line version of say is how JRobert suggested. Here's my quick implementation, using Doug Harris's spelling suggestion: #!/bin/sh echo $@ | tr '[A-Z]' '[a-z]' | sed "s/spolsky/spowlsky/g" | /usr/bin/say Finally, some fun command line stuff: # Apple is weird sqlite3 /System/Library/PrivateFrameworks/SpeechDictionary.framework/Resources/Tuples .dump # Get too much information about what files are being opened sudo dtrace -n 'syscall::open*:entry { printf("%s %s",execname,copyinstr(arg0)); }' # Just fun say -v bad "Joel Spolsky Spolsky Spolsky Spolsky Spolsky, Joel Spolsky Spolsky Spolsky Spolsky Spolsky" echo "scale=1000; 4*a(1)" | bc -l | say

  Read the article

• Oracle Solaris 11 Developer Webinar Series

  - by Larry Wake

  This coming Tuesday, a new series of webcasts (not to be confused with a series of tubes) is kicking off, aimed at developers. Register today Next week's session covers IPS and related topics: What: Modern Software Packaging for Enterprise Developers When: Tuesday, March 27, 9 AM Pacific Who: Eric Reid, Oracle Systems ISV Engineering We've got several more queued up -- here's the full schedule, with registration links for each one. Or, see the series overview, which includes a link to a "teaser" preview of all the sessions. Topic Date (all sessions 9 AM Pacific) Speaker Modern Software Packaging for Enterprise Developers March 27th Eric Reid (Principal Software Engineer) Simplify Your Development Environment with Zones, ZFS & More April 10th Eric Reid (Principal Software Engineer)Stefan Schneider (Chief Technologist, ISV Engineering) Managing Application Services – Using SMF Manifests in Solaris 11 April 24th Matthew Hosanee (Principal Software Engineer) Optimize Your Applications on Oracle Solaris 11: The DTrace Advantage May 8th Angelo Rajadurai (Principal Software Engineer) Maximize Application Performance and Reliability on Oracle Solaris 11 May 22nd Ikroop Dhillon (Principal Product Manager) Writing Oracle Solaris 11 Device Drivers June 6th Bill Knoche (Principal Software Engineer)

  Read the article

• Webcast: New Features of Solaris 11.1 and Solaris Cluster 4.1

  - by Jeff Victor

  If you missed last week's webcast of the new features in Oracle Solaris 11.1 you can view the recording. The speakers discuss changes that improve performance and scalability, particularly for Oracle DB, and many other enhancements. New features include Optimized Shared Memory (improves DB startup time), accelerated kernel locks (improves Oracle RAC performance and scalability), virtual memory improvements, a DTrace data collecter in the DB, Zones installed on Shared Storage (simplifies migration), Data Center Bridging, and Edge Virtual Bridging. To view the archived webcast, you must register and use the URL that you receive in e-mail.

  Read the article

• Oracle OpenWord 2012 - Managing Storage in the Cloud

  - by jwalker

  At Oracle OpenWorld this year attendees will get experience using the Sun ZFS Storage Appliance during the Managing Storage in the Cloud Hands-On-Lab. Using Sun ZFS Storage, we will be provisioning Oracle Enterprise Linux Virtual Machines and filesystem shares that can be used with Oracle Database. We will also be using Oracle DTrace Analytics to analyze I/O workloads and drill down to see how the storage is really being used. Hope you can join us! Session ID: HOL10034 Session Title: Managing Storage in the Cloud Speakers: Brian Haskins, Nagendran J, Paul Johnson, Karlheinz Vogel and Jim Walker Venue and Room: Marriott Marquis - Salon 14/15 Date and Times: Monday October 1 - 3:15-4:15PM, Tuesday October 2 - 5:00-6:00PM Oracle OpenWorld Storage Sessions

  Read the article

• Oracle Solaris 11 Developer Webinar Series

  - by nospam(at)example.com (Joerg Moellenkamp)

  Starting March 27th a number of events called "Oracle Solaris 11 Developer Webinar Series" (that said, i think they are not only interesting for developers, but for admins as well that support admins in their job) will take place by WebEx/Phone: March 27th: Modern Software Packaging for Enterprise Developers April 10th: Simplify Your Development Environment with Zones, ZFS & More April 24th: Managing Application Services ? Using SMF Manifests in Solaris 11 May 8th: Optimize Your Applications on Oracle Solaris 11: The DTrace Advantage May 22nd: Maximize Application Performance and Reliability on Oracle Solaris 11 June 5th: Writing Oracle Solaris 11 Device Drivers All events start at 9am Pacific Daylight Time (Should be 18:00 in Germany, 17:00 in UK, 01:00 in Tokyo). Of course the events are for free (before someone asks) and available for anyone who signs up. The chat is staffed with Solaris 11 Engineering Experts sitting in to answer questions. And most important: Mostly technical content and the events will be recorded.

  Read the article

• Performance triage

  - by Dave

  Folks often ask me how to approach a suspected performance issue. My personal strategy is informed by the fact that I work on concurrency issues. (When you have a hammer everything looks like a nail, but I'll try to keep this general). A good starting point is to ask yourself if the observed performance matches your expectations. Expectations might be derived from known system performance limits, prototypes, and other software or environments that are comparable to your particular system-under-test. Some simple comparisons and microbenchmarks can be useful at this stage. It's also useful to write some very simple programs to validate some of the reported or expected system limits. Can that disk controller really tolerate and sustain 500 reads per second? To reduce the number of confounding factors it's better to try to answer that question with a very simple targeted program. And finally, nothing beats having familiarity with the technologies that underlying your particular layer. On the topic of confounding factors, as our technology stacks become deeper and less transparent, we often find our own technology working against us in some unexpected way to choke performance rather than simply running into some fundamental system limit. A good example is the warm-up time needed by just-in-time compilers in Java Virtual Machines. I won't delve too far into that particular hole except to say that it's rare to find good benchmarks and methodology for java code. Another example is power management on x86. Power management is great, but it can take a while for the CPUs to throttle up from low(er) frequencies to full throttle. And while I love "turbo" mode, it makes benchmarking applications with multiple threads a chore as you have to remember to turn it off and then back on otherwise short single-threaded runs may look abnormally fast compared to runs with higher thread counts. In general for performance characterization I disable turbo mode and fix the power governor at "performance" state. Another source of complexity is the scheduler, which I've discussed in prior blog entries. Lets say I have a running application and I want to better understand its behavior and performance. We'll presume it's warmed up, is under load, and is an execution mode representative of what we think the norm would be. It should be in steady-state, if a steady-state mode even exists. On Solaris the very first thing I'll do is take a set of "pstack" samples. Pstack briefly stops the process and walks each of the stacks, reporting symbolic information (if available) for each frame. For Java, pstack has been augmented to understand java frames, and even report inlining. A few pstack samples can provide powerful insight into what's actually going on inside the program. You'll be able to see calling patterns, which threads are blocked on what system calls or synchronization constructs, memory allocation, etc. If your code is CPU-bound then you'll get a good sense where the cycles are being spent. (I should caution that normal C/C++ inlining can diffuse an otherwise "hot" method into other methods. This is a rare instance where pstack sampling might not immediately point to the key problem). At this point you'll need to reconcile what you're seeing with pstack and your mental model of what you think the program should be doing. They're often rather different. And generally if there's a key performance issue, you'll spot it with a moderate number of samples. I'll also use OS-level observability tools to lock for the existence of bottlenecks where threads contend for locks; other situations where threads are blocked; and the distribution of threads over the system. On Solaris some good tools are mpstat and too a lesser degree, vmstat. Try running "mpstat -a 5" in one window while the application program runs concurrently. One key measure is the voluntary context switch rate "vctx" or "csw" which reflects threads descheduling themselves. It's also good to look at the user; system; and idle CPU percentages. This can give a broad but useful understanding if your threads are mostly parked or mostly running. For instance if your program makes heavy use of malloc/free, then it might be the case you're contending on the central malloc lock in the default allocator. In that case you'd see malloc calling lock in the stack traces, observe a high csw/vctx rate as threads block for the malloc lock, and your "usr" time would be less than expected. Solaris dtrace is a wonderful and invaluable performance tool as well, but in a sense you have to frame and articulate a meaningful and specific question to get a useful answer, so I tend not to use it for first-order screening of problems. It's also most effective for OS and software-level performance issues as opposed to HW-level issues. For that reason I recommend mpstat & pstack as my the 1st step in performance triage. If some other OS-level issue is evident then it's good to switch to dtrace to drill more deeply into the problem. Only after I've ruled out OS-level issues do I switch to using hardware performance counters to look for architectural impediments.

  Read the article

• A Patent for Workload Management Based on Service Level Objectives

  - by jsavit

  I'm very pleased to announce that after a tiny :-) wait of about 5 years, my patent application for a workload manager was finally approved. Background Many operating systems have a resource manager which lets you control machine resources. For example, Solaris provides controls for CPU with several options: shares for proportional CPU allocation. If you have twice as many shares as me, and we are competing for CPU, you'll get about twice as many CPU cycles), dedicated CPU allocation in which a number of CPUs are exclusively dedicated to an application's use. You can say that a zone or project "owns" 8 CPUs on a 32 CPU machine, for example. And, capped CPU in which you specify the upper bound, or cap, of how much CPU an application gets. For example, you can throttle an application to 0.125 of a CPU. (This isn't meant to be an exhaustive list of Solaris RM controls.) Workload management Useful as that is (and tragic that some other operating systems have little resource management and isolation, and frighten people into running only 1 app per OS instance - and wastefully size every server for the peak workload it might experience) that's not really workload management. With resource management one controls the resources, and hope that's enough to meet application service objectives. In fact, we hold resource distribution constant, see if that was good enough, and adjust resource distribution if that didn't meet service level objectives. Here's an example of what happens today: Let's try 30% dedicated CPU. Not enough? Let's try 80% Oh, that's too much, and we're achieving much better response time than the objective, but other workloads are starving. Let's back that off and try again. It's not the process I object to - it's that we to often do this manually. Worse, we sometimes identify and adjust the wrong resource and fiddle with that to no useful result. Back in my days as a customer managing large systems, one of my users would call me up to beg for a "CPU boost": Me: "it won't make any difference - there's plenty of spare CPU to be had, and your application is completely I/O bound." User: "Please do it anyway." Me: "oh, all right, but it won't do you any good." (I did, because he was a friend, but it didn't help.) Prior art There are some operating environments that take a stab about workload management (rather than resource management) but I find them lacking. I know of one that uses synthetic "service units" composed of the sum of CPU, I/O and memory allocations multiplied by weighting factors. A workload is set to make a target rate of service units consumed per second. But this seems to be missing a key point: what is the relationship between artificial 'service units' and actually meeting a throughput or response time objective? What if I get plenty of one of the components (so am getting enough service units), but not enough of the resource whose needed to remove the bottleneck? Actual workload management That's not really the answer either. What is needed is to specify a workload's service levels in terms of externally visible metrics that are meaningful to a business, such as response times or transactions per second, and have the workload manager figure out which resources are not being adequately provided, and then adjust it as needed. If an application is not meeting its service level objectives and the reason is that it's not getting enough CPU cycles, adjust its CPU resource accordingly. If the reason is that the application isn't getting enough RAM to keep its working set in memory, then adjust its RAM assignment appropriately so it stops swapping. Simple idea, but that's a task we keep dumping on system administrators. In other words - don't hold the number of CPU shares constant and watch the achievement of service level vary. Instead, hold the service level constant, and dynamically adjust the number of CPU shares (or amount of other resources like RAM or I/O bandwidth) in order to meet the objective. Instrumenting non-instrumented applications There's one little problem here: how do I measure application performance in a way relating to a service level. I don't want to do it based on internal resources like number of CPU seconds it received per minute - We need to make resource decisions based on externally visible and meaningful measures of performance, not synthetic items or internal resource counters. If I have a way of marking the beginning and end of a transaction, I can then measure whether or not the application is meeting an objective based on it. If I can observe the delay factors for an application, I can see which resource shortages are slowing an application enough to keep it from meeting its objectives. I can then adjust resource allocations to relieve those shortages. Fortunately, Solaris provides facilities for both marking application progress and determining what factors cause application latency. The Solaris DTrace facility let's me introspect on application behavior: in particular I can see events like "receive a web hit" and "respond to that web hit" so I can get transaction rate and response time. DTrace (and tools like prstat) let me see where latency is being added to an application, so I know which resource to adjust. Summary After a delay of a mere few years, I am the proud creator of a patent (advice to anyone interested in going through the process: don't hold your breath!). The fundamental idea is fairly simple: instead of holding resource constant and suffering variable levels of success meeting service level objectives, properly characterise the service level objective in meaningful terms, instrument the application to see if it's meeting the objective, and then have a workload manager change resource allocations to remove delays preventing service level attainment. I've done it by hand for a long time - I think that's what a computer should do for me.

  Read the article

< Previous Page | 1 2 3 4  | Next Page >