Re: January RU Combo Patch

Yes, except because you don't typically have a database in your template home you won't hit the ORA-00600 then. You'd need to test actual upgrades of course.

On Mon, Feb 24, 2020, 08:12 Ls Cheng <exriscer_at_gmail.com> wrote:

> Hi
>
> Ok, so I guess this is a sort of "you fix one software copy (the gold
> image) and deploy it to other places/servers/clusters". Because the ORA-600
> would very probably happen when patching to the Gold Image.
>
> I have previously upgraded three 18.5 clusters to 19.6 without any issues.
> I hit the ORA-600 when I patched two 19.4 clusters (both got ORA-600). I
> guess is a 19c bug....
>
> Thanks
>
>
>
>
>
> On Mon, Feb 24, 2020 at 8:59 AM <niall.litchfield_at_gmail.com> wrote:
>
>> Absolutely. Makes rollback simple as well :) I appreciate it's not simple
>> to adjust processes in the first place.
>>
>> On Mon, Feb 24, 2020 at 7:48 AM Ls Cheng <exriscer_at_gmail.com> wrote:
>>
>>> Hi Niall
>>>
>>> With Gold Image I guess everything (patchset, RU, RUR, one-off) is
>>> out-of-place upgrade?
>>>
>>> Thank you
>>>
>>>
>>> On Mon, Feb 24, 2020 at 8:27 AM <niall.litchfield_at_gmail.com> wrote:
>>>
>>>> HI LS
>>>>
>>>> We've been using Gold Images (tarballs of a patched Oracle Home + home
>>>> cloning) to rollout known sets of patches like this for some time. If you
>>>> have the appropriate license this functionality is also baked into both OEM
>>>> and Fleet Patching and Maintenance. That's a longer-term answer. The short
>>>> answer to your question is that whilst GI setup allows the application of
>>>> one-offs the database installer doesn't yet offer this functionality.
>>>>
>>>> On Sat, Feb 22, 2020 at 9:01 PM Ls Cheng <exriscer_at_gmail.com> wrote:
>>>>
>>>>> Hi
>>>>>
>>>>> I found yesterday a patch which is related to the ORA-600 error:
>>>>>
>>>>> *Patch 30663646* STRESS:FA:PRJ-COMBO:ORA 600 [PFRI.C#103: BAD LU
>>>>> TYPE] PFRI.C PFRINS ( SYN TRANSLATION ) - ASM_FEATURE_PUB.IS_FEATURE_ENABLED
>>>>> Linux x86-64 for Oracle Database 19.6.0.0.0 DBRU
>>>>>
>>>>> And just 5 minutes ago the SR I opened yesterday suggests to apply
>>>>> this patch. However does anyone know how to apply RU and a one-off patch at
>>>>> the same time so I can ensure the software binaries are patched with Patch
>>>>> 30663646 therefore ORA-600 is avoided?
>>>>>
>>>>> Similar to gridSetup.sh -applyRUR
>>>>>
>>>>> Thanks
>>>>>
>>>>>
>>>>>
>>>>> On Fri, Feb 21, 2020 at 4:47 PM Ls Cheng <exriscer_at_gmail.com> wrote:
>>>>>
>>>>>> I compared 30501910 which comes with
>>>>>>
>>>>>> p30463609_190000_Linux-x86-64.zip
>>>>>> p30501910_190000_Linux-x86-64.zip
>>>>>>
>>>>>> and they have same size
>>>>>>
>>>>>>
>>>>>> du -sk 30463609/30501910
>>>>>> 3745320 30463609/30501910
>>>>>>
>>>>>>
>>>>>> du -sk 30501910
>>>>>> 3745320 30501910
>>>>>>
>>>>>>
>>>>>> I have a SR opened as well
>>>>>>
>>>>>> datapatch was ran against MGMTDB and this is the error
>>>>>>
>>>>>>
>>>>>> Stopping dequeue for queue AQ_SRVNTFN_TABLE_Q_1
>>>>>> begin
>>>>>> *
>>>>>> ERROR at line 1:
>>>>>> ORA-00600: internal error code, arguments: [pfri.c#103: bad lu type],
>>>>>> [22], [],
>>>>>> [], [], [], [], [], [], [], [], []
>>>>>> ORA-06512: at "SYS.DBMS_AQADM_SYS", line 15446
>>>>>> ORA-06512: at "SYS.DBMS_AQADM_SYSCALLS", line 778
>>>>>> ORA-06512: at "SYS.DBMS_AQADM", line 769
>>>>>> ORA-06512: at line 5
>>>>>> ORA-06512: at line 5
>>>>>>
>>>>>>
>>>>>>
>>>>>> This is the code
>>>>>>
>>>>>> begin
>>>>>> for queue in (select name from dba_queues where owner = 'SYS' AND
>>>>>> name like 'AQ_SRVNTFN_TABLE_Q%') loop
>>>>>> begin
>>>>>> dbms_output.put_line('Stopping dequeue for queue ' || queue.name);
>>>>>> DBMS_AQADM.STOP_QUEUE (
>>>>>> queue_name => queue.name,
>>>>>> enqueue => FALSE,
>>>>>> dequeue => TRUE,
>>>>>> wait =>TRUE);
>>>>>> dbms_output.put_line('Dequeue stopped for queue ' || queue.name);
>>>>>> end;
>>>>>> end loop;
>>>>>> exception when others then
>>>>>> dbms_output.put_line('Exception while stopping queues ' || SQLCODE ||
>>>>>> ' ' || SQLERRM);
>>>>>> end;
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Fri, Feb 21, 2020 at 1:33 PM Jay Hostetter <
>>>>>> hostetter.jay_at_gmail.com> wrote:
>>>>>>
>>>>>>> This was the exact problem that I ran into when running datapatch.
>>>>>>> The workaround for me was to NOT use the combo patch. I downloaded the
>>>>>>> Grid RU and OJVM RU patches separately:
>>>>>>>
>>>>>>> p30501910_190000_Linux-x86-64.zip
>>>>>>> p30484981_190000_Linux-x86-64.zip
>>>>>>>
>>>>>>> Jay
>>>>>>>
>>>>>>> On Fri, Feb 21, 2020 at 6:36 AM Ls Cheng <exriscer_at_gmail.com> wrote:
>>>>>>>
>>>>>>>> Hi
>>>>>>>>
>>>>>>>> Did you get this error when patching the management database? I got
>>>>>>>> error for two clusters and for other 2 cluster the patch went through
>>>>>>>>
>>>>>>>>
>>>>>>>> Installing patches...
>>>>>>>> Patch installation complete. Total patches installed: 3
>>>>>>>>
>>>>>>>> Validating logfiles...done
>>>>>>>> Patch 30557433 apply (pdb CDB$ROOT): SUCCESS
>>>>>>>> logfile:
>>>>>>>> /u01/app/grid/cfgtoollogs/sqlpatch/30557433/23305305/30557433_apply__MGMTDB_CDBROOT_2020Feb21_10_18_25.log
>>>>>>>> (no errors)
>>>>>>>> Patch 30557433 apply (pdb PDB$SEED): SUCCESS
>>>>>>>> logfile:
>>>>>>>> /u01/app/grid/cfgtoollogs/sqlpatch/30557433/23305305/30557433_apply__MGMTDB_PDBSEED_2020Feb21_10_19_56.log
>>>>>>>> (no errors)
>>>>>>>> Patch 30557433 apply (pdb GIMR_DSCREP_10): WITH ERRORS
>>>>>>>> logfile:
>>>>>>>> /u01/app/grid/cfgtoollogs/sqlpatch/30557433/23305305/30557433_apply__MGMTDB_GIMR_DSCREP_10_2020Feb21_10_19_56.log
>>>>>>>> (errors)
>>>>>>>> -> Error at line 42: script rdbms/admin/bug_29784106_prepatch.sql
>>>>>>>> - ORA-00600: internal error code, arguments: [pfri.c#103: bad
>>>>>>>> lu type], [22], [],
>>>>>>>> - [], [], [], [], [], [], [], [], []
>>>>>>>> - ORA-06512: at "SYS.DBMS_AQADM_SYS", line 15446
>>>>>>>> - ORA-06512: at "SYS.DBMS_AQADM_SYSCALLS", line 778
>>>>>>>> - ORA-06512: at "SYS.DBMS_AQADM", line 769
>>>>>>>> - ORA-06512: at line 5
>>>>>>>> - ORA-06512: at line 5
>>>>>>>>
>>>>>>>> Thanks
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> On Fri, Feb 7, 2020 at 9:37 PM Jay Hostetter <
>>>>>>>> hostetter.jay_at_gmail.com> wrote:
>>>>>>>>
>>>>>>>>> My support tech didn't say if it will be republished. Since I
>>>>>>>>> have a workaround, the SR is now closed.
>>>>>>>>>
>>>>>>>>> Jay
>>>>>>>>>
>>>>>>>>> On Fri, Feb 7, 2020 at 2:09 AM Ls Cheng <exriscer_at_gmail.com>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>>> Good to know, are they going to re-publish the patch?
>>>>>>>>>>
>>>>>>>>>> TIA
>>>>>>>>>>
>>>>>>>>>> On Thu, Feb 6, 2020 at 8:13 PM Jay Hostetter <
>>>>>>>>>> hostetter.jay_at_gmail.com> wrote:
>>>>>>>>>>
>>>>>>>>>>> I ran into an ORA-0600 when running datapatch for the January
>>>>>>>>>>> 2020 RU for 19c. After going back and forth with Support and doing some
>>>>>>>>>>> testing, it seems as though the combo patch for Linux x86-64 (30463609)
>>>>>>>>>>> may have an issue. My specific error was related to Streams.
>>>>>>>>>>>
>>>>>>>>>>> I downloaded the GI and OJVM patches separately and was able to
>>>>>>>>>>> successfully apply the the RU.
>>>>>>>>>>>
>>>>>>>>>>> Hopefully this info may save some aggravation if you are
>>>>>>>>>>> preparing to install this Release Update.
>>>>>>>>>>>
>>>>>>>>>>> Jay
>>>>>>>>>>>
>>>>>>>>>>
>>>>
>>>> --
>>>> Niall Litchfield
>>>> Oracle DBA
>>>> http://www.orawin.info
>>>>
>>>
>>
>> --
>> Niall Litchfield
>> Oracle DBA
>> http://www.orawin.info
>>
>

--
http://www.freelists.org/webpage/oracle-l