Recently I’ve seen not so smart optimizer behavior: one query took long time to parse, and ended with an error hitting PGA_AGGREGATE_LIMIT in few minutes; another query was just parsed for ages while using reasonable (under 2G :)) amount of PGA and still could hit PGA_AGGREGATE_LIMIT but after way more time – up to an hour.
Both cases were similar and involved queries which were accessing views; and those views’ code is generated by an application using lots of IN LISTs and other OR conditions. They both are really ugly SQLs with text length ~100K. When Oracle tried to parse them it took a lot of time and parse attempt had either failed with ORA-4036 soon or hanged for a long time and then failed. Strangely incident trace file generated for ORA-4036 doesn’t include PGA heaps breakdown and you have to manually enable PGA heapdump on error to get an idea what is taking up memory. Here’s what I’ve found in there:
I've encountered a bug at several clients that upgraded to Oracle 12c - 126.96.36.199 - that requires the combination of several new adaptive features introduced with Oracle 12c.It needs an execution plan that makes use of adaptive join methods, plus at runtime the activation of the new "statistics feedback for joins" feature that was also introduced with Oracle 12c. Note that in 11.2 there was already the "cardinality feedback" feature that only applies to single table cardinality misestimates, but not to join cardinality misestimates.In case then the join method used at runtime is a Nested Loop join - not necessarily the join method preferred initially, so a runtime switch from Hash to Nested Loop join also reproduces the problem - the "statistics feedback for joins" feature generates a bad OPT_ESTIMATE hint for the join cardinality that always seems to be one, like the following: OPT_ESTIMATE(...
I finally managed to install and test patch 22652097 (PROVIDE SEPARATE CONTROLS FOR ADAPTIVE PLANS AND ADAPTIVE STATISTICS FEATURES). Actually, I installed and tested two patches. The first was installed on top of “Oracle Database Patch 23054246 – Database Patch Set Update 188.8.131.52.160719” (from now on, 184.108.40.206.160719 PSU). The second was installed on top of “Patch 24448103 – Database Proactive Patch 220.127.116.11.161018” (18.104.22.168.161018 PBP).
From a functional point of view, both patches provides the 12.2 configuration in 12.1 as I describe it in this post. In fact, the behavior of setting the (un)documented initialization parameters as well as all the default values are the same. I was able to find only two differences between 12.2 and a patched 12.1.
In the past I gave a number of 1-day seminars about the new performance features available in Oracle Database 12c Release 1. On the 22nd of February, for the first time, I’ll give an updated version of that seminar with content about both Release 1 and Release 2. Note that because there is more content, I extended it from one day to two days.
Here is another example (besides the fact that Adaptive Cursor Sharing only gets evaluated during a PARSE call (still valid in 12c) and supports a maximum of 14 bind variables) I've recently come across at a client site where the default implementation of Adaptive Cursor Sharing fails to create a more suitable execution plan for different bind variable values.Broken down to a bare minimum the query was sometimes executed using non-existing values for a particular bind variable, but other times these values were existing and very popular. There were two suitable candidate indexes and one of them appeared to the optimizer more attractive in case of the "non-existing" value case.
Having done my mini-series on Nested Loop join logical I/O optimizations a while ago I unfortunately never managed to publish anything regarding the Nested Loop join physical I/O optimizations, which are certainly much more relevant to real-life performance.Therefore the main purpose of this blog post is to point you to Nikolay Savvinov's (whose blog I can recommend in general) great mini-series covering various aspects of these optimizations:Part 1Part 2Part 3SummaryOne point that - at least to me - isn't entirely clear when reading Nikolay's series is which specific plan shape he refers to, in particul
This is just a short 12c update on my post about gathering statistics on tables with many columns from some time ago.I'm currently investigating the "Incremental Statistics" feature in 22.214.171.124 for a client, which probably will be worth one or more other posts, but since we're still in the process of evaluating and installing various patches it's too early to write about that.As part of the investigation I've noticed a significant change in behaviour in 126.96.36.199 compared to previous versions when it comes to gathering statistics on tables with many columns, hence this post here.The key message of the original post was that DBMS_STATS needs potentially several passes when gathering statistics on tables with many columns, which meant a significant increase in overall work and resource consumption, exaggerated by the fact that tables with that many colu
Quite often you can get into trouble with Oracle when you start combining different features.In this case of one my clients it is the combination of user-defined PL/SQL functions that can raise exceptions (think of currency conversion and a non-existent currency code gets passed into the function), DML error logging and attempting to improve performance by wrapping the PL/SQL function call into a scalar subquery to benefit from the built-in scalar subquery caching feature of the SQL runtime engine.As long as the scalar subquery didn't get used everything worked as expected, but after adding the scalar subquery after some while it became obvious that wrong results occurred - in that particular case here it meant rows that should have been rejected and written to the error logging table due to the exception raised in the user-defined PL/SQL function suddenly showed up in the target table, and what was even more worrying - they included a co
One of my customers that recently upgraded to 12c hit a bug (22913528) that I think is good to be aware of. Note that as the title of this post states, the problem only occur in 188.8.131.52. At least, I wasn’t able to reproduce it in any other version.
To reproduce it you simply need a composite partitioned table with a non-partitioned or global-partitioned index. In other words, if all your indexes are local, you shouldn’t be impacted by the bug.
The SQL statements I use to prepare the schema to reproduce it are the following:
As mentioned in the first and second part of this instalment the different available distribution methods of the new parallel FILTER are selected automatically by the optimizer - in this last post of this series I want to focus on that optimizer behaviour.It looks like there are two new optimizer related parameters that control the behaviour of the new feature: "_px_filter_parallelized" is the overall switch to enable/disable the new parallel filter capability - and defaults to "true" in 12c, and "_px_filter_skew_handling" influences how the optimizer determines the distribution methods - the parameter naming suggests that it somehow has to do with some kind of "skew" - note that the internal parameter that handles the new