Solaris

I've got quite a number of tweets and mails with the question "But is zfs scrub" not something like fsck. And the answer is "Well ... no".

• It's zpool scrub, not zfs scrub. So it can't be a fsck. Perhaps it's a pock, but not fsck
  (Okay, that's a really lame argument. )
• Scrub is not part of the ZFS Posix layer, the entity that let you access a zpool with the semantics of a filesystem. It's done much deeper in ZFS and while it's following a tree there, this tree has nothing to do with your filesystem structure.
  (Ah, not much better reasoning.)
• You don't run zfs scrub on a unimportable pool, you run run zpool clear -F, which starts the txg rollback to make the pool importable again.
  (Don't know if this one is more convincing)
• A scrub works as well on a ZFS Emulated Volume.It can't do a filesystem check on it. Obviously, as Solaris has no idea for example how to check a ext2 file system that you are writing via iSCSI into a zvol. However the scrub has still to work.
  (Perhaps a little better.)
• I've got one tweet saying that it finds silent corruption in the data, thus it's a filesystem check. What I find interesting about this, is the point that this definition would exclude other fsck from being fsck, as they just check the validity of the metadata and not the validity of the data. zpool scrub is a data validity checker, not a filesystem checker. It checks if you are reading the data from a location that you have once written to this location. Whatever this data is. As far as i understand the source, it even doesn't understand the concept of "filesystems" at all.
• What is zpool scrub? It is not much more than reading everything. However the repair mechanism is the same as the one when you read a block and the checksum on-disk doesn't match the checksum computed. The scrub has no own repair code. In the case an error is detected , for example for RAID1 the known good copy is written in the place of the bad copy. Whereas the repair while normal reading is more or less accidental (you've stumbled over an error and reading at one location doesn't mean that other locations are correct, you have to check all locations to be sure). Scrub is the forced search for such incorrectnesses on all copies of a block stored in a pool. With a scrub you ensure that all redundancies are correct. At end scrub checks if the data on-disk is still the data you have written once by other means (ZPL or ZVOL) on it. It simply doesn't care about the structure, as it doesn't know about the structure. It's like memory scrubbing or RAID scrubbing. When your trigger a scrubbing in your RAID array the array gives nothing what's on it.

So, no zfs scrub is no fsck, it has a different job. It might look like a fsck, but it isn't. Forget everything how you expect filesystems internals from your pre-ZFS knowledge. I can give just one important advice in regard of understanding ZFS. An advice that was already used on the first presentations about ZFS: Free your mind!

Kristina Tripp and Isaac Rozenfeld wrote a nice article about the steps needed to enable an existing Solaris 10 jumpstart server to install Solaris 11 automatically as well: How to Use an Existing Oracle Solaris 10 JumpStart Server to Provision Oracle Solaris 11 11/11"

Friday was a day that i called once 10k day. More 10.000 visitors to my blog in one day. Saturday was similar. This surge was create by an link on news.ycombinator.com article i wrote roughly four years ago about ZFS: No, ZFS really doesn't need a fsck.

Just wanted to express that four years later and a lot more experience with ZFS later, 12 years after ZFS saw the light of the word, i'm more of the opinion that ZFS doesn't need a fsck than ever.

One of the features introduced with 11.1 is the Address Space Layout Randomization (ASLR) . And when you work with 11.1 you are already using it. So it's a less known, but frequently used feature: less known in the point that it exists, less known in the point of the methods to control it, frequently used as it's activated per default for selected binaries (and many were selected)

With the announcement of SPARC T5 and SPARC M5 a lot of benchmarks were mentioned. I won't repeat all of them here, however i would like to point you to the BestPerf blog, which has a long lists of benchmarks related to the announcement.

T5

• SPARC T5-8 Produces TPC-C Benchmark Single-System World Record Performance
• SPARC T5-8 Delivers SPECjEnterprise2010 Benchmark World Record Performance
• SPARC T5-2 Achieves SPECjbb2013 Benchmark World Record Result
• SPARC T5-8 Realizes SAP SD Two-Tier Benchmark World Record for 8 Chip Systems
• SPARC T5 Systems Deliver SPEC CPU2006 Rate Benchmark Multiple World Records
• SPARC T5-2 Achieves JD Edwards EnterpriseOne Benchmark World Records
• SPARC T5-2 Scores Siebel CRM Benchmark World Record
• SPARC T5 Systems Produce Oracle TimesTen Benchmark World Record
• SPARC T5-8 Delivers Oracle OLAP World Record Performance
• SPARC T5-2 Achieves ZFS File System Encryption Benchmark World Record
• SPARC T5-2 Obtains Oracle Internet Directory Benchmark World Record Performance
• SPARC T5-2 Scores Oracle FLEXCUBE Universal Banking Benchmark World Record Performance
• SPARC T5-2 Performance Running Oracle Fusion Middleware SOA
• SPARC T5-1B Performance Running Oracle Communications ASAP

M5

• SPARC M5-32 Produces SAP SD Two-Tier Benchmark World Record for SAP Enhancement Package 5 for SAP ERP 6.0

The "Oracle Processor Core Factor Table" has been updated in order to include SPARC T5 and M5: It's 0.5 for both.

Glenn Faden strikes again: In the article "Oracle Solaris Extended Policy and MySQL" he describes, how to use the Extended Policy feature to lock down the mysql service.

Glenn Faden wrote a really great article about the sandboxing of applications with privileges: Application Containment via Sandboxing. Worth a read.

After my blog entry about the both events in Frankfurt and Munich i was asked "And what's about other cities? in Germany. Thus i would like to point you to similar events in other cities in Germany. As this events are events in Germany, i will proceed in German.

Zu einen findet ein Event mit dem Titel "Ankündigung: Neue SPARC Server". In Berlin am 18. April statt. Ort ist das Konrad-Zuse-Zentrum für Informationstechnik Berlin. Anmelden könnt ihr euch hier. Die Agenda kann dort ebenfalls schon eingesehen werden.

Auch interessant ist der Oracle Innovation Truck. Dieser kommt am 15.4. nach Hamburg (hier werde ich mich unter anderem in den LKW stellen), am 17.4. nach Düsseldorf und am 18.4. nach Stuttgart.. Ort ist jeweils eine Lokation in der Innenstadt, genaueres wird da noch bekannt gegeben. Bitte meldet euch vorher über den zum gewünschten Ort gehörigen Ort an, dort findet ihr auch eine genaue beschreibung des Events. Bei diesen geht es auch um die neuen Server, aber nicht ausschliesslich. Es gibt aber auch andere Demohardware zu betrachten: T-Server, Oracle Database Appliance sowie die Storage Library SL150.

I used this feature in the HA-loadbalancing tutorial already. However the future too useful to stay just a "by-word" in a different article. It is DLMP. Or by its full name:"Data Link Multipathing".

The events announced in this blog entry take place in Germany, thus this entry will be in german language.

Ich habe ja gestern schon auf den Webcast "Announcing New SPARC Servers with the World’s Fastest Microprocessor"am 26.März hingewiesen. Ich möchte aber aber nun ebenfalls Eure Aufmerksamkeit auf die dazu passenden etwas späteren stattfindenden Live-Events lenken:

• 9. April 2013 in Frankfurt
• 10. April 2013 in München

Sprechen auf den Events wird unter anderem Rick Hetherington, Vice President Hardware Development Oracle. Eine mehr als berufene Stimme also. Ihr könnt euch mit den Links anmelden, die mit den Datumsangaben verknüpft sind. Um zahlreiches Erscheinen wird gebeten. Es verspricht sehr interessant zu werden!

I would like to point you to an upcoming webcast: "Announcing New SPARC Servers with the World’s Fastest Microprocessor"

The breakthrough speed and power of Oracle SPARC can transform your enterprise computing, igniting innovation and driving results. Join this live Webcast featuring Oracle’s Larry Ellison and John Fowler as they unveil the new standard for extreme performance.

Don't miss it. The webcast takes place on Tuesday, March 26, 2013 between 20:00 and 21:30 GMT. You can register here.

Preview. Will change in the next evenings for typo correction, inclusion of pictures and so on. It's a draft, i just put it here this early, because a customer asked me for such a tutorial and i didn't wanted to reformat it for readability as a Word or LaTeX document . The commandlines are final, the language in between will be proof-read soon. That said, all suggestions are really welcome

As you may know, Solaris contains an integrated load balancer. It's really easy to configure. Not that well known is the point, that you can make it higly-available in an easy way as well. The following tutorial will give you an overview on the configuration of this feature.

In the recent day, a video about a Solaris 9 machine up and running for 3737 days made news in blogs or at Slashdot. I know the machine must have been downrev as hell and perhaps not that loaded (they wrote that it was idling for a year before the shutdown). But never the less quite some time. But i wanted to add something else: I immediately thought, that it was unnecessary to stop the machine. There are ways to keep the system running while moving the system as readers of this blog know, as i linked to a video of some guys moving a running system through Hamburg with public transport some time ago.

Brendan Gregg has written an interesting piece about finding performance problems: "The USE method addresses shortcomings in other commonly used methodologies".

It's a good paper, however ... well let's say, I don't understand why so many people find it especially cool or especially good, because at the end it isn't something really new. Don't understand me wrong: It's good. But not extraordinarily good. Like many methodologies it's basically just codified common sense with a personal spin. So I would prefer to say "My-personal-way-of-doing-stuff" instead of calling it methodology. There is nothing new in it. Just a lot of common sense.

I really think that performance analysis is not so much about a "methodology" you can simply follow that will lead you magically to a result. It's about a mindset how to tackle problems, it's about being structured in the approach, it's about "being prepared", it's a lot about knowing stuff.

As I do performance analysis quite frequently, I have created my own "methodology", or to be more correct ... my own mindset of doing such stuff. I don't call it method or methodologies. Perhaps it's useful for some or the other ... so i write it down here.