[Ocfs2-test-devel] [PATCH 0/5] Ocfs2-test: Add indexed-dirs tests for ocfs2-test V2.

[tristan.ye]

>From v1 to v2:

Use public common shell file to avoid redundancy.

Tristan


On Wed, 2009-02-25 at 11:22 +0800, tristan.ye wrote:
> Following patches set is aiming to add both single-node and multi-nodes
> indexed-dirs tests for ocfs2-test.
> 
> The reference URL for this tests as follows, where all designing doc,
> testing tools and testing status/report kept.
> http://oss.oracle.com/osswiki/OCFS2/DxdirsTest   
> I therefore will not take room for details in this letter.
> 
> 
> Mark,
> I'm glad to say that my single-node tests survived well even under a
> heavy workload with your latest dx_dirs patches. while multi-nodes tests
> still ongoing...
> 
> 
> Regards,
> Tristan
> 
> 
> 
> On Wed, 2009-02-04 at 19:49 +0800, tristan.ye wrote:
> > Hi mark,
> > 
> > Glad to see directory indexing become possible in ocfs2 by your
> > patches:)
> > 
> > I'm going to test this out in coming days.
> > 
> > Following is something rough I've drafted for the testplan and
> > scenarios. 
> > 
> > ==========Testing scenario for ocfs2 indexed directory===========
> > 1. Feature/unit tests.
> >    1) Test all directory
> > manipulations(readdir,telldir,unlink,create,lookup) with a fixed
> > workload(vary from small,medium and large).
> > 
> >    2) Random tests,run tests with a random directory dirent
> > length,dirent numbers or directory depth, also perform dirent
> > deletion/rename from a random logical offset. also fill up empty dirent
> > blocks randomly.
> > 
> >   3) Do verifications after each test to see if dirents we're
> > manipulating are missing/duplicated or not.
> > 
> >   4) Multi-nodes tests, to perform a cross-node rename,deletion to race
> > the tests.
> >   
> >  Besides,we're going to run as much more blocksize and clustersize
> > combination as we can to rich the tests.
> > 
> > 2. Stress tests
> >     We perform tests with a extremely large number of dirents, a maxium
> > size of dirent record length and also a enormous directory depth and
> > numbers in filesystem.
> > 
> > 3. Boundary tests
> >    1.) Test to right extend a inlined directory into indexed one.
> > 
> >    2) Test to right extend a inlined index record block to extent.
> > 
> > 4. Performance comparison tests.
> >    Collect the timing measurement data for all kinds of directory
> > manipulations(Unlink,readdir,insert/creation,rename,find,ls,mkdir,rmdir,etc...)with different workloads, then generate a comparison report between unindexed directory and indexed directory to see if substantial performance improvementes gained.
> > 
> > =========Testing scenario for ocfs2 indexed directory ends===========
> > 
> > It will be greatly appreciated if you take your time to review the above
> > testplan, or let me know any other corner testcases and testing
> > limitations missed to make the tests more thorough and cover more
> > corresponding codes as much as possible(to create more white-box
> > testcases).
> > 
> > I'm sure there must be some important testcases missing in boundary
> > tests due to my poor understanding to your indexed directory framework
> > right now. 
> > 
> > I'm going to create a wiki to have these documented after you've
> > commented on this, which also helps to keep the track of testing status
> > and result in following days.
> > 
> > 
> > Regards,
> > Tristan
> > 
> > 
> > On Fri, 2009-01-30 at 13:42 -0800, Mark Fasheh wrote:
> > > The following patches implement indexed directory support in Ocfs2, mostly
> > > according to the design doc I wrote up a while ago:
> > > 
> > > http://oss.oracle.com/osswiki/OCFS2/DesignDocs/IndexedDirectories
> > > 
> > > The patches have been rebased on top of 2.6.29-rc2. It should be trivial to
> > > put them into merge_window. Things are what I'd call complete now. I'd like
> > > to get these into the merge_window branch sooner rather than later, so any
> > > reviews are appreciated :)
> > > 
> > > Changes from last time:
> > > 
> > > - Added inline indexing support. This turned out to be a good idea - many
> > >   directories just keep their index inlined.
> > > 
> > > - Rebased so it fully complies with metaecc and quota requirements.
> > > 
> > > - Lots of bugs fixed.
> > > 
> > > 
> > > There's a 'dx_dirs' branch in ocfs2.git:
> > > 
> > > git pull git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2.git dx_dirs
> > > 
> > > 
> > > I've tested this batch of patches with kernel builds, and various bash loops
> > > to create large numbers of files. Last time around, it survived a two day
> > > exorcist run. I'll schedule another one for the weekend.
> > > 
> > > 
> > > Tools support for this is available in ocfs2-tools.git:
> > > 
> > > git pull git://oss.oracle.com/git/ocfs2-tools.git dx_dirs
> > > 
> > > I won't post the exact tools patches just yet. There's enough support in
> > > there that we can mkfs, tunefs (only turning the feature on) and debugfs. I
> > > need to add better write support and of course things like fsck, etc. The
> > > good news is tha the debugfs.ocfs2 support is what I'd call complete - it
> > > should allow us to fully debug and visualize the on-disk structures in
> > > question.
> > > 	--Mark
> > > 
> > > --
> > > Mark Fasheh
> > > 
> > > _______________________________________________
> > > Ocfs2-devel mailing list
> > > Ocfs2-devel at oss.oracle.com
> > > http://oss.oracle.com/mailman/listinfo/ocfs2-devel
> 
> 
> _______________________________________________
> Ocfs2-test-devel mailing list
> Ocfs2-test-devel at oss.oracle.com
> http://oss.oracle.com/mailman/listinfo/ocfs2-test-devel