BLOG UPDATE: 2014.05.15: The following link supersedes all other references to SLOB kit and patches. This will always be the up-to-date locale: https://kevinclosson.wordpress.com/slob/
BLOG UPDATE 2013.12.26: Quick link to download the kit
BLOG UPDATE 2012.05.05: Updated the tar archive distribution file with some bug fixes. Simply preserve your slob.conf file and extract this tar archive over your prior SLOB install directory.
BLOG UPDATE 2012.05.04: The PDF README will no longer be bundled in with the tar archive. The README can be found here: SLOB2 README.
BLOG UPDATE 2012.05.03: First time visitors should see the introductory page for SLOB.
About SLOB 2
I’ve already socialized the SLOB 2 update via twitter and a lot of friends have had early access to the kit. So, this is just a very brief blog entry to point to SLOB 2.
I’ve written a form of a release note that will be sufficient for current SLOB users to move forward rapidly with new SLOB 2 features. The note can be found here: SLOB 2 README or here.
Download The SLOB2 Kit
To download the software you can access the tar archive on EMC Syncplicity. Click SLOB 2 Tar Archive.
After downloading you should verify the md5sum:
$ md5sum 2013.05.05.slob2.tar e1e67a68bf253a02532ebd556a2ea782 2013.05.05.slob2.tar $
Kevin Closson's Blog: Platforms, Databases and Storage 
hmmm….. shared file not found in that link?
Not SLOB12c? Dissapointed…
Cheers
Tim… 🙂
Not sure if I’m liking the popularity of SLOB….Popularity + Oracle normally equals acquisition….I can see the future now:
C:\Users\connor> sqlplus abc/xyz
SQL*Plus: Release 13.1.0.5.0 Production on Sat May 15 14:57:27 2013
Copyright (c) 1982, 2015, Oracle. All rights reserved.
Connected to:
Oracle13s Enterprise Edition Release 13.1.0.5.0 – 64bit Production
With the Partitioning and SLOB options
and probably $50k per core as well !!
🙂
Hey Kevin,
The syncplicity links for both the TAR archive and the README are both broken. Can you point me to some valid links?
Sorry about that, Mark. The links appear to expire after a fixed number of downloads. I’ve addressed the problem. Please try again now and ping me if you have problems.
Looks good now, thanks!
Hi Kevin,
How do I run SLOB2 for a specified amount of time (say a few hours) for a specified update percentage (say 20%)? I have tried setting 3600 for “RUN_TIME” and 30 for UPDATE_PCT. The run completes in about 1800 seconds instead of specified 3600 seconds. If I kick off a subsequent run keeping RUN_TIME and UPDATE_PCT setting the same in slob.conf file, the test runs randomly end within a few seconds (sometimes 5 seconds, sometimes 8 etc.).
Thank you.
– Saad
Yes long duration testing has a bug. The next version of SLOB2 does not have the bug. If you will ping me in email I can send you a patch (early access for the up-coming updated kit).
Thanks Kevin.
Sent you an email with the request for early access to up-coming updated kit.
– Saad
@Saad : Neto tells me he took care of you.
@Kevin, yes got the patch through Neto. Thank you. If you happen to make any further updates to the kit, please do send me a copy.
Hi Kevin,
Is there a maximum limit to the RUN_TIME parameter in the slob.conf file?
Even though I specified a runtime of 24 hours, it seemed to stop generating any load after 5-6 hours.
I am using the latest version of SLOB2 (2014.12.29.slob_2.2.1.3.tar.gz)
Thanks.
— Prashanth
Hello Prashanth,
Oh no… people do runs as long as many days. You might care to check the alert log if you haven’t already. Another ting to do is to export SLOB_DEBUG=TRUE and re-run ..you’ll get a log file from SLOB called slob_debug.out in your SLOB directory.
Hi Kevin,
Thanks for a very useful IO testing tool.
I tested SLOB2 and somehow the values of “bytes per second” were not extracted correctly by the script awr_info.sh – i think the problem is with the chomp command.
In my server (el6uek.x86_64) the command ‘chomp’ does not seem to work, so ‘bytes per second’ values are zero.
Example:
function get_read_mbs() {
local f=$1
grep ‘^physical read total bytes’ $f | head -1 | chomp | awk ‘{
printf(“%6.0lf\n”, ( $2 / 2 ^ 20) )
}’
}
I changed it to:
grep ‘^physical write total bytes’ $f | head -1 | awk ‘{print $6}’ | awk ‘{ printf(“%6.0lf\n”,$1) }’
May be you can provide a fix for that in the next release.
Br,
Pascal
Hello,
just for your info: there’s a syntax error in misc/awr_info.sh – missing closing double quote at line 235.
Hi Dimitre,
I fixed this. Please go to kevinclosson.net/slob to get the latest.
Thank you!
Thank you for taking your time to point it out. I appreciate the help.
well, Dimitre, I fixed it but embarrassingly didn’t test it. So the fix yesterday was also buggy. I apologize. Please go to kevinclosson.net/slob and download the latest-latest as it were …
Hi Kevin,
I’d like to know how to configure & use SLOB2 to exercise the RAC interconnect [in an AIX environment].
I know this information might be available in the LinkedIn Group “SLOB Oracle Platform Testing” (Yury Velikanov).
I asked for permission to be able to join this group last week, however, have not heard back.
Might you (or anyone) be able to help?
Thanks,
Rich
I think I’m getting somewhere with this – slob against one instance with UPDATE_PCT=0 and another instance of slob against another instance with UPDATE_PCT=100 seems to get a lot of ‘gc cr block busy’ waits from the select heavy instance; what I’d like to do is try to get ‘gc cr grant 2-way’ waits…
Let me give this some thought. But first, I need to know if you are using a round-robin connection method (single SQL*Net server) or are you using a named service per instance?
Hi Kevin,
Sorry for the late reply – holidays.
At first, I tried with SQLNET_SERVICE_MAX=2 (for round-robin) and UPDATE_PCT=50, however, that only got me a relatively small ‘gc current grant busy’ spike which resolves quickly on a first run (makes sense).
So, I went to using 2 separate instances of SLOB using different SQL*Net service names – one with UPDATE_PCT=50 to the first instance and the other with UPDATE_PCT=100 to the second instance. This gets me a bigger ‘gc current block 2-way’ spike, however, it resolves quite quickly as well.
I’ve not tried named service per instance…
I do see a lot of ‘enq: US – contention’ in both cases – would that cause an issue?
yes. I recommend you google for “SLOB recipe”