I recently had to troubleshoot an ORA-4031 issue at a client site. The issue reappeared 3 times within 2 months and only after escalating the SR to SEV1 and being quite persistent for an explanation, the second engineer attempting to solve the issue finally got it right.  Being curious, I digged into the trace files again to confirm and understand the issue here. This blog post describes the method to analyze the issue and how to troubleshoot.

• First, when the issue first appears, it creates a 4031 trace file, which does not really give enough information to solve the problem. So I asked Oracle Support for specific trace events to properly diagnose it. He correctly mentioned event 10235 level 65536 and heapdump level 536870914  but required an instance restart. After explaining that this is not really possible and researching myself, I found these two events, which do not really need an instance restart. Event 10235 was not really necessary in this case, because the description of the allocations did not matter here.

ALTER system SET events '4031 trace name HEAPDUMP level 536870914, lifetime 1; name errorstack level 3, lifetime 1';

• Now, after the issue appeared again, we finally had the relevant diagnostic information.

Allocation Request Summary Informaton
=====================================
Allocation request for: kglsim object batch
...
Heap: 0x6007b278, size: 3896
******************************************************
HEAP DUMP heap name="sga heap(3,0)"  desc=0x6007b278
extent sz=0xfe0 alt=248 het=32767 rec=9 flg=-126 opc=2
...
durations enabled for this heap
reserved granules for root 0 (granule size 134217728)
Total heap size    =2684352800. (2559 MB)
Total free space   =    38360
Total reserved free space   =120837856 (115MB) 
Permanent space    =2563474760
 
==============================================
TOP 10 MEMORY USES FOR SGA HEAP SUB POOL 1
----------------------------------------------
"free memory               "       6597 MB 71%
"gcs resources             "       1080 MB 12%
"gcs shadows               "        748 MB  8%
"db_block_hash_buckets     "        200 MB  2%
"kglsim object batch       "        189 MB  2%
"kglsim heap               "        112 MB  1%
"gcs res hash bucket       "         64 MB  1%
"ges big msg buffers       "         54 MB  1%
"dbktb: trace buffer       "         51 MB  1%
"KGLH0                     "         48 MB  1%
-----------------------------------------
free memory                        6597 MB
memory alloc.                      2747 MB
Sub total                          9344 MB
==============================================
TOP 10 MEMORY USES FOR SGA HEAP SUB POOL 2
----------------------------------------------
"free memory               "       6641 MB 70%
"gcs resources             "       1080 MB 11%
"gcs shadows               "        747 MB  8%
"db_block_hash_buckets     "        512 MB  5%
"gcs res hash bucket       "         64 MB  1%
"kglsim object batch       "         59 MB  1%
"dbktb: trace buffer       "         51 MB  1%
"KGLH0                     "         44 MB  0%
"kglsim heap               "         35 MB  0%
"ges enqueues              "         22 MB  0%
-----------------------------------------
free memory                        6641 MB
memory alloc.                      2831 MB
Sub total                          9472 MB
==============================================
TOP 10 MEMORY USES FOR SGA HEAP SUB POOL 3
----------------------------------------------
"free memory               "       6488 MB 71%
"gcs resources             "       1080 MB 12%
"gcs shadows               "        747 MB  8%
"db_block_hash_buckets     "        200 MB  2%
"kglsim object batch       "        105 MB  1%
"gc name table             "         72 MB  1%
"gcs res hash bucket       "         64 MB  1%
"kglsim heap               "         63 MB  1%
"dbktb: trace buffer       "         51 MB  1%
"KGLH0                     "         46 MB  1%
-----------------------------------------
free memory                        6488 MB
memory alloc.                      2600 MB
Sub total                          9088 MB
==============================================
TOP 10 MEMORY USES FOR SGA HEAP SUB POOL 4
----------------------------------------------
"free memory               "       6663 MB 70%
"gcs resources             "       1080 MB 11%
"gcs shadows               "        749 MB  8%
"db_block_hash_buckets     "        512 MB  5%
"kglsim object batch       "         59 MB  1%
"dbktb: trace buffer       "         51 MB  1%
"KGLH0                     "         47 MB  0%
"kglsim heap               "         35 MB  0%
"FileOpenBlock             "         35 MB  0%
"ges enqueues              "         22 MB  0%
-----------------------------------------
free memory                        6663 MB
memory alloc.                      2809 MB
Sub total                          9472 MB
TOTALS ---------------------------------------
Total free memory                    26 GB
Total memory alloc.                  11 GB
Grand total                          37 GB

• This shows that 4031 was triggered for memory allocation request in sga heap (3,0) with size of 3896. This is smaller than 4400 bytes (_shared_pool_reserved_min_alloc), so the request can not be served the the shared pool reserved area.  Next, we can see that we have 4 subpools (kghdsidx_count) and each of them has around 6 GB of free memory. So, in total, we have 37 GB shared pool with 26 GB of it free, but one mini-subpool (duration 0) of subpool 3 still has run out of free contigous memory bigger than 3792 bytes.
• Running Tanel´s heapdump_analyzer, I could verify that the largest chunk for sga heap(3,0) was indeed 3792 bytes and that all bigger chunks were allocated by “PERM” allocations.

 ./heapdump_analyzer  ./PROD1_ora_27396.trc > heap.txt
 
grep -e "---" -e "Total_size" -e "sga heap(3,0)"  heap.txt
 
-- Heapdump Analyzer v1.00 by Tanel Poder ( http://www.tanelpoder.com )
 
Total_size #Chunks  Chunk_size,        From_heap,       Chunk_type,  Alloc_reason
---------- ------- ------------ ----------------- ----------------- -----------------
 
114123720      17    6713160 ,    sga heap(3,0),           R-free,
28442152       1   28442152 ,    sga heap(3,0),             perm,  perm
...
7485000       1    7485000 ,    sga heap(3,0),             perm,  perm
6713080       1    6713080 ,    sga heap(3,0),           R-free,
6712104       1    6712104 ,    sga heap(3,0),           R-perm,  perm
...
3028568       1    3028568 ,    sga heap(3,0),             perm,  perm
...
 
8928       4       2232 ,    sga heap(3,0),             perm,  perm
7864       1       7864 ,    sga heap(3,0),             perm,  perm
5488       2       2744 ,    sga heap(3,0),             perm,  perm
3792       1       3792 ,    sga heap(3,0),             free,  <<<<<<<<<<<<<<
...

This shows that the first chunk of chunk_type “free” is of 3792 bytes, but request was for 3896 bytes. The
subpool (3,0) only has R-free (for shared_pool_reserved area) and PERM allocations of bigger size. The subpool (3,0) consists of 20 Extents á (128M Granule size). Each extent has

Researching the issue with “PERM” allocations brought me to:

ORA-4031: unable to allocate 4160 bytes of shared memory (“shared pool”,”unknown object”,”sga heap(4,0)”,”modification “) (Doc ID 1675470.1)

This note also confirms that the classification in 4 durations leads to a weak spot of duration 0 in any subpool for memory allocations due to high PERM utilization. The note mentions patch 8857940, which was also finally recommended by the support engineer.

HEAP DUMP heap name="sga heap(1,0)"  desc=0x600680e8
Total heap size    =2818570440
Total free space   =    28008 <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< 
Total reserved free space   =127549960
HEAP DUMP heap name="sga heap(1,1)"  desc=0x60069940
Total heap size    =805305840
Total free space   =632590232
Total reserved free space   = 40243520
HEAP DUMP heap name="sga heap(1,2)"  desc=0x6006b198
Total heap size    =1879046960
Total free space   =1734275840
Total reserved free space   = 93975928
HEAP DUMP heap name="sga heap(1,3)"  desc=0x6006c9f0
Total heap size    =4294964480
Total free space   =4072007248
Total reserved free space   =214821040
HEAP DUMP heap name="sga heap(2,0)"  desc=0x600719b0
Total heap size    =2952788080
Total free space   =    19144 <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Total reserved free space   =147689440
HEAP DUMP heap name="sga heap(2,1)"  desc=0x60073208
Total heap size    =671088200
Total free space   =524561448
Total reserved free space   = 33565720
HEAP DUMP heap name="sga heap(2,2)"  desc=0x60074a60
Total heap size    =1879046960
Total free space   =1738701544
Total reserved free space   = 93984160
HEAP DUMP heap name="sga heap(2,3)"  desc=0x600762b8
Total heap size    =4429182120
Total free space   =4202264536
Total reserved free space   =221534200
HEAP DUMP heap name="sga heap(3,0)"  desc=0x6007b278
Total heap size    =2684352800
Total free space   =    38360 <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Total reserved free space   =120837856
HEAP DUMP heap name="sga heap(3,1)"  desc=0x6007cad0
Total heap size    =671088200
Total free space   =526361840
Total reserved free space   = 33565720
HEAP DUMP heap name="sga heap(3,2)"  desc=0x6007e328
Total heap size    =1879046960
Total free space   =1736230384
Total reserved free space   = 93977784
HEAP DUMP heap name="sga heap(3,3)"  desc=0x6007fb80
Total heap size    =4294964480
Total free space   =4076323080
Total reserved free space   =214821040
HEAP DUMP heap name="sga heap(4,0)"  desc=0x60084b40
Total heap size    =2952788080
Total free space   =     8536 <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Total reserved free space   =140977632
HEAP DUMP heap name="sga heap(4,2)"  desc=0x60087bf0
Total heap size    =1879046960
Total free space   =1735347144
Total reserved free space   = 93984160
HEAP DUMP heap name="sga heap(4,3)"  desc=0x60089448
Total heap size    =4429182120
Total free space   =4198496144
Total reserved free space   =221534200

For reference, these are the relevant parameters. Please note that all of these are set to default, except _kghdsidx_count which was reduced from 4 earlier and _ksmg_granule_size which was reduced to 128M.

NAME                                     VALUE
---------------------------------------- --------------------
__shared_pool_size                       38923141120
_dm_max_shared_pool_pct                  1
_enable_shared_pool_durations            TRUE
_io_shared_pool_size                     4194304
_kghdsidx_count                          4
_ksmg_granule_size                       134217728
_memory_imm_mode_without_autosga         TRUE
_shared_pool_max_size                    0
_shared_pool_minsize_on                  FALSE
_shared_pool_reserved_min_alloc          4400
_shared_pool_reserved_pct                5
shared_pool_reserved_size                1946157056
shared_pool_size                         38654705664

Thanks to Tanel, Riyaj and Hatem Mahmoud

I was troubleshooting an issue where the Oracle Database Result Cache did not get used when the RESULT_CACHE hint was specified inside a subquery. As the main query has a sysdate inside, the RESULT_CACHE Hint can not be specified for the main query, but only for the subquery.

1
2

SELECT SQ.*, sysdate FROM
(SELECT * FROM tableX, ....) SQ;

When using “/*+ RESULT_CACHE */ inside of Subquery, result cache was not enabled.

1
2
3

 
SELECT SQ.*, sysdate FROM
(SELECT /*+ RESULT_CACHE */ * FROM tableX, ....) SQ;

I found out that there is a trace event which can be set for diagnostics.

1

ALTER SESSION SET EVENTS '43905 trace name context forever, level 1';

This generated a trace file containing a clue to the reason:

1

QKSRC: ViewText[ecode=942] = SELECT  /*+ RESULT_CACHE */ ......

So, it seemed that there was an internal ORA-942 somewhere that prevented result caching from being used. The solution was to qualify the objects inside the subquery. Now the result cache was used correctly:

1
2

SELECT SQ.*, sysdate FROM
(SELECT /*+ RESULT_CACHE */ * FROM ownerY.tableX, ....)

During the Upgrade of Oracle Linux 7.2 to 7.3 with “yum update”, when updating the microcode_ctl package, the system crashed and firmware reported an uncorrectable CPU error after reboot.

It turned out that the update of the package in combination with specific Intel CPUs causes the issue. In the meantime, there is a red hat bug and solution available.

References:

• https://community.oracle.com/thread/3997160
• https://bugzilla.redhat.com/show_bug.cgi?id=1398698
• MOS: The system could not start since suddenly rebooted during upgrade to microcode_ctl package (Doc ID 2217876.1)

I recently had an issue where a database with 240GB SGA (1 huge shared memory segment) configured with hugepages on a system with 512G RAM was suddenly becoming instable and new logons were denied with these error message:


ORA-01034: ORACLE not available
ORA-27123: unable to attach to shared memory segment
Linux-x86_64 Error: 22: Invalid argument
Additional information: 2667
Additional information: 1736718
Additional information: 215016800256


This was strange because ipcs -a showed all shared memory segments and all processes existed. It turned out that there is a known issue with the Linux feature “prelink” and very big SGAs (>200g). This MOS Note gives some details:

Connected to an idle instance, while database instance is running (Process J000 died, see its trace file kkjcre1p: unable to spawn jobq slave process ) (Doc ID 1578491.1)

Obviously, exadata customers also suffered from this issue and therefore prelink is disabled beginning with exadata version 11.2.3.3. Redhat is reporting the same issue in combination with Informix databases in Knowledgebase article: https://access.redhat.com/solutions/1186953

In previous versions, it was “best practice” to always get the most current opatch (patch 6880880) from MOS. Unfortunately, with Enterprise Manager Cloud Control 13c, this is problematic at the moment. The reason is that OMS 13.1 is shipped with OPatch 13.6:

[oracle@em13c ~]$ opatch version
OPatch Version: 13.6.0.0.0
 
OPatch succeeded.

Currently, OPatch 13.6 is not available in MOS. Only OUI Nextgen 13.2 is available, but not opatch 13.6. Do not overwrite OPatch 13.6 with OUI Nextgen 13.2 because it will break it.

When trying to patch OMS 12.1.0.5 in a VirtualBox environment with latest OMS PSU, I came across a strange problem which took quite a while to solve. The opatchauto apply and also -analyze commands failed every time with this error after several minutes of hanging:

opatchauto failed with error code 231

Manual connect to WLS console with relevant protocol/host/port/username/password was working fine. Then I realized that there was an issue with the entropy pool on the VirtualBox VM.

I followed this Note to resolve the isse:

E1: OS: Linux Servers Hang or Have Delays on Any JAVA Process Affecting Performance (Doc ID 1525645.1)

After implementing rngd, the patching worked successfully without any hangs.

During implementation of AWR Warehouse, I discovered that AWR warehouse is using temporary staging schemas in the AWR warehouse repository database. These schemas life approximately for the duration of a datapump import job and are then dropped again. Due to the fact that the used password is not compliant with customers password verification function, the jobs failed.

v_sql := ‘ CREATE USER ‘ || STAGING_SCHEMA || ‘ IDENTIFIED BY SYS_GUID ‘ ||
‘ DEFAULT TABLESPACE ‘ || tbsname;

The staging schemas are created with the password “SYS_GUID” in capital lettters. This looks to me like the developer tried to generate a random string as password but instead overlooked that the password is set to fixed string “SYS_GUID” instead.

Oracle support has noticed this issue and filed an enhancement request. It is a pity, that this is not filed as a bug, but an enhancement.

Well, I hope this improves in a future version together with dynamic retention and purging options as well as customizable staging directories.

Happy AWR´ing.

For one of my clients, I experienced several issues with applying PSU 12.1.0.2.5 with opatch 12.1.0.1.9.

There were some code changes in opatchauto, which are not yet production-ready. I am looking forward to seeing a new opatch release (maybe 12.1.1.10 or 11) which has these issues fixed.

• Bug 22091017 : OPATCHAUTO -ANALYZE COMMAND SHUTS DOWN THE RDBMS HOME IN 12C
  “opatchauto apply -analyze” is normally used before patching to verify if there are any conflicts. In our case, the “opatchauto apply -analyze” did a shutdown of the database instances. Not good.
• BUG 22202019 – OPATCHAUTO FAILS WITH “TOO MANY ARGUMENTS” FROM CHECKSPFFILE SCRIPT
  When applying GI PSU patch to RDBMS Home and database instances are already stopped, patch apply fails. The reason is that opatchauto is trying to query database instances for information before shutting down instances. If instances are already stopped, these queries fail and opatchauto aborts.
• OPatchauto is shutting down database instance with “srvctl stop home -stopoption TRANSACTIONAL”. I was of the opinion that this is doing a “shutdown transactional”, which in turn will wait until the last session is performing a commit. This can take a long time. Support Engineer explained that the wait will be limited to 10min, after which the system is begin stopped with “shutdown abort”. This is not documented in “http://docs.oracle.com/database/121/RACAD/srvctladmin.htm#RACAD5040” and in my opinion a bug. I expect the shutdown to be “immediate” whithout delaying the patching process.

Customers have been waiting for a Standard Edition release for version 12.1.0.2. Then, in July, a support engineer leaked that there will be a whole new type of “Edition”, called “Standard Edition 2” to replace both “Standard Edition” and “Standard Edition One”. The respective MOS Note disappeared shortly after the accidental publication. Now today the release finally became official and can be downloaded via OTN. The “SE2” is priced exactly as the previous “Standard Edition”. SE2 is limited to a single server with 2 CPU sockets, or when used with RAC to two servers with one socket each. In addition, there seems to be a limit of 16 CPU threads (e.g. 8 cores with hyperthreading) when using one host and 8 CPU threads each for both nodes when using RAC.

So beginning with 12.1.0.2, there will be no more “Standard Edition (4 CPU socket limit)” or “Stanard Edition One (2 socket limit)” releases anymore. There is no way to avoid going to “Standard Edition 2 (2 socket limit)” in the future.

This might mean for SE customers, which were using 4 CPU sockets (occupied) on a single host or a 2-node RAC Cluster with 2 CPU sockets (occupied) per host, that they have to remove half the (occupied) sockets from their installation. I am not sure what happens with already bought SE licences and how they can be “migrated” to SE2 licenses. I hope that this will not start a new wave of license-driven Oracle->PostgreSQL/MySQL/MS SQL Migrations.

Now would probably be the right time for 11gR2 Standard Edition users to start upgrade tests for 12c to gather experience. On the other hand, waiting until end of year and bugs found by “early adaptors” might safe you some troubles.

There is one very interesting feature of Microsoft Hyper-V called Replica, which allows replication of Virtual Machines without shared storage. With Hyper-V Replica, one can set up Oracle Database VMs to replicate between independent nodes. This document (in german) describes the setup in more detail:

Hyper-V Replica für Oracle Database