Changing the Timezone for a particular Database *only* in a RAC

April 16th, 2019 No comments

Sometimes there are requests that multiple databases running on the same RAC return different time values — so they must run in different time zones.

This goal cannot achieved by changing the server time zone because that would affect ALL databases.

The solution is fairly simple:

bash-4.2$ srvctl setenv database -db ORCL -env "TZ=CET-1"
bash-4.2$ srvctl stop database -d ORCL
bash-4.2$ srvctl start  database -d ORCL

This effectively changes the environment variables in the clusterware for this particular database:

 

bash-4.2$ sqlplus c##ronny/ronny@ORCL

SQL*Plus: Release 18.0.0.0.0 - Production on Wed Feb 6 09:29:03 2019

Version 18.3.0.0.0

Copyright (c) 1982, 2018, Oracle.  All rights reserved.

Connected to:

Oracle Database 18c Enterprise Edition Release 18.0.0.0.0 - Production

Version 18.3.0.0.0

SQL> select to_char(sysdate, 'HH24:MI') from dual;

TO_CH
-----
08:29

SQL> !date

Wed Feb  6 09:29:16 CET 2019

Note that this only works if you connect via SQLNet. Local connetions are unaffected!

Categories: Oracle in general Tags:

ORA-01422: exact fetch returns more than requested number of rows ORA-06512: at “SYS.CDBVIEW_INTERN during upgrade

March 1st, 2019 No comments

During a database upgrade vom 11.2.0.4 to 18.4.1 i encountered the following problem:

REASON:
      ERRORS FOUND: During Upgrade
         FILENAME: /data/oracle/product/cfgtoollogs/dbua/upgrade2019-02-22_10-18-20AM/DB/catupgrd1R0.log AT LINE NUMBER: 666432
------------------------------------------------------
Identifier CATALOG 19-02-22 11:00:37
SCRIPT    = [/data/oracle/product/18.4/rdbms/admin/cdcore_cols.sql]
ERROR     = [ORA-01422: exact fetch returns more than requested 
             number of rows ORA-06512: at "SYS.CDBVIEW_INTERNAL", line 199
ORA-06512: at "SYS.CDBVIEW_INTERNAL", line 64
ORA-06512: at "SYS.CDBVIEW", line 6
ORA-06512: at line 1
]
STATEMENT = [BEGIN CDBView.create_cdbview(false,'SYS','DBA_TAB_COLS','CDB_TAB_COLS'); END;
]
------------------------------------------------------

Note that the view name (here: DBA_TAB_COLS) can change!

 

The reason for this to fail is in the file $ORACLE_HOME/rdbms/admin/catcdbviews.sql:

 

  -- Check if oldview already exist, qualify with owner# to avoid ORA-1422
  -- error,if two dictionary view with the same name exists across different
  -- schemas like DBA_DV_STATUS which exists in both SYS and DVSYS schemas.

    begin
      execute immediate 'SELECT 1 FROM OBJ$ WHERE NAME = :1' ||
                       ' AND TYPE# in (2, 4) and owner# = :2'
               into cCheck using oldview, ownerId;

 

The developer already identified one issue with his code, but overlooked another: Whenever a database link gets referenced in a piece of pl/sql code (user code or internal code – does not matter), a object with the same name as referenced by the database link is created in OBJ$.

 

SQL> select obj#, name, remoteowner,linkname from obj$ where TYPE# in (2, 4) and name=’DBA_TAB_COLS’;

 

      OBJ# NAME           REMOTEOWNER   LINKNAME
---------- -------------- ------------- --------------------------
  14000219 DBA_TAB_COLS   SYS           STAGING_SYSTEM
      3334 DBA_TAB_COLS

As you can see we have a second object in OBJ$ named DBA_TAB_COLS for the database link. And this causes the upgrade to fail.

 

The fix is relatively simple: Either get a patch from Oracle or fix the code in catcdbviews.sql yourself:

begin
      execute immediate 'SELECT 1 FROM OBJ$ WHERE NAME = :1' ||
          ' AND TYPE# in (2, 4) and remoteowner is null and owner# = :2'
               into cCheck using oldview, ownerId;

After editing the scirpt, install it by executing catcdbview.sql to apply the changes and restart the upgrade. The upgrade will start from the failed step onwards.

Note that the package you edit here is only used during the database upgrade to transfer comment from non-CDB to CDB views. It is deleted after the upgrade and is not needed or used during normal database operation.

Categories: Oracle in general Tags:

RMAN shows: could not locate the LNM save file ‘fqnacfte_1_1’ on server ‘nsrserverhost’. (2:9:2)

September 22nd, 2018 No comments

During an RMAN restore of a backup made to a networker server the restore fails retrieving a backup piece:

restore datafile 2;

channel ORA_SBT_TAPE_1: starting datafile backup set restore
channel ORA_SBT_TAPE_1: specifying datafile(s) to restore from backup set
channel ORA_SBT_TAPE_1: restoring datafile 00002 to +DATA/dwh/datafile/sysaux.257.764502905
channel ORA_SBT_TAPE_1: reading from backup piece DWH_fqnacfte_1_1
channel ORA_SBT_TAPE_1: ORA-19870: error while restoring backup piece DWH_fqnacfte_1_1
ORA-19507: failed to retrieve sequential file, handle="DWH_fqnacfte_1_1", parms=""
ORA-27029: skgfrtrv: sbtrestore returned error
ORA-19511: Error received from media manager layer, error text:
   Could not locate the LNM save file 'DWH_fqnacfte_1_1' on server 'nsrserverhost'. (2:9:2)

Lets´s check if the backup set exists on the server:

root@nsrserverhost # mminfo -a -v -r name,ssid,cloneid,savetime,client,state,location,ssretent,ssbrowse,ssflags -q client=clientname | grep DWH_fqnacfte_1_1
RMAN:DWH_fqnacfte_1_1     3349640857  1336374016 05/07/12 clientname  E      JUKEBOX         05/14/12 05/14/12 vrEF
RMAN:DWH_fqnacfte_1_1     3349640857  1336374016 05/07/12 clientname         JUKEBOX         05/14/12 05/14/12 vrEF

So we can see the save set actually exists but is marked “Expired” (E). So let´s try to change that back:

nsrmm -S 3349640857 -w 07/28/12 -e 07/28/12
Cannot change browse/retention time of recyclable save set 3349640857

root@nsrserverhost # nsrmm -o notrecyclable -v -S 3349640857
Save set 3349640857 cannot be marked as notrecyclable.
Please specify the ssid/cloneid of the particular clone instance.
root@nsrserverhost # nsrmm -o notrecyclable -v -S 3349640857/1336374016
Mark save set copy 3349640857/1336374016 as not recyclable? y
root@nsrserverhost # nsrmm -S 3349640857 -w 07/28/12 -e 07/28/12
Cannot exceed the browse time of a recoverable save set 3349640857

Does not work as the information is already purged from the client database. In that case we need to restore the client index from a date before the expiry and change the browse and retention:

root@nsrserverhost # nsrck -L7 -t 05/07/12 clientname
nsrck: checking index for 'clientname'
9343:nsrck: The file index for client 'clientname' will be recovered.
Requesting 1 recover session(s) from server.
Recover completion time: Tue May 15 19:22:37 2012
nsrck: /var/nsr/index/clientname contains 69 records occupying 37 KB
nsrck: Completed checking 1 client(s)
root@scrkva16 # mminfo -a -v -r name,ssid,cloneid,savetime,client,state,location,ssretent,ssbrowse,ssflags -q client=scrdwh02 | grep DWHKVASY_fqnacfte_1_1
RMAN:DWH_fqnacfte_1_1     3349640857  1336374016 05/07/12 clientname         JUKEBOX         05/23/12 05/23/12 vF
RMAN:DWH_fqnacfte_1_1     3349640857  1336374016 05/07/12 clientname         JUKEBOX         05/23/12 05/23/12 vF
root@nsrserverhost # nsrmm -S 3349640857 -w 07/28/12 -e 07/28/12

And now the restore works as expected:

restore datafile 2;

channel ORA_SBT_TAPE_1: starting datafile backup set restore
channel ORA_SBT_TAPE_1: specifying datafile(s) to restore from backup set
channel ORA_SBT_TAPE_1: restoring datafile 00002 to +DATA/dwh/datafile/sysaux.257.764502905
channel ORA_SBT_TAPE_1: reading from backup piece DWH_fqnacfte_1_1

Categories: Networker, Oracle in general Tags:

Direct NFS: Failed to set socket buffer size.wtmax=[1048576] rtmax=[1048576], errno=-1 on Solaris

August 17th, 2016 No comments

This error occured on Solaris (SPARC) using Direct NFS (dNFS). The MOS Note is incomplete and only recommends to increase the send and receive buffers, but the solution is below:

 

/usr/sbin/ndd -set /dev/tcp tcp_xmit_hiwat 1056768
/usr/sbin/ndd -set /dev/tcp tcp_recv_hiwat 1056768
/usr/sbin/ndd -set /dev/tcp tcp_max_buf 2097152

To check the values, use:

/usr/sbin/ndd /dev/tcp tcp_max_buf
/usr/sbin/ndd /dev/tcp tcp_xmit_hiwat
/usr/sbin/ndd /dev/tcp tcp_recv_hiwat
Categories: Oracle in general Tags:

Exadata outage due to ASM disks missing

August 1st, 2016 No comments

On some older Exadatas it can be that Grid Disks belonging to a cell are not found anymore. And not even FORCEing the moutn helps:

FAILGROUP HEADER_STATU COUNT(*)
------------------------------ ------------ ----------
EXA01CEL02 MEMBER 34
EXA01CEL03 MEMBER 34
EXA01CEL04 MEMBER 34
EXA01CEL05 MEMBER 34
EXA01CEL06 MEMBER 34
EXA01CEL07 MEMBER 34
EXA01CEL08 UNKNOWN 34
EXA01CEL09 UNKNOWN 34
EXA01CEL10 UNKNOWN 32
EXA01CEL11 UNKNOWN 34
EXA01CEL12 MEMBER 34
EXA01CEL13 MEMBER 34
EXA01CEL14 MEMBER 34



SQL> alter diskgroup DBFS_DG mount force;
alter diskgroup DBFS_DG mount force
*
ERROR at line 1:
ORA-15042: ASM disk "71" is missing from group number "1"
ORA-15042: ASM disk "70" is missing from group number "1"
ORA-15042: ASM disk "69" is missing from group number "1"
ORA-15042: ASM disk "68" is missing from group number "1"
ORA-15042: ASM disk "67" is missing from group number "1"
ORA-15042: ASM disk "66" is missing from group number "1"
ORA-15042: ASM disk "65" is missing from group number "1"
ORA-15042: ASM disk "64" is missing from group number "1"
ORA-15042: ASM disk "63" is missing from group number "1"
ORA-15042: ASM disk "62" is missing from group number "1"
ORA-15042: ASM disk "61" is missing from group number "1"
ORA-15042: ASM disk "60" is missing from group number "1"
ORA-15042: ASM disk "59" is missing from group number "1"
ORA-15042: ASM disk "58" is missing from group number "1"
ORA-15042: ASM disk "57" is missing from group number "1"
ORA-15042: ASM disk "56" is missing from group number "1"
ORA-15042: ASM disk "55" is missing from group number "1"
ORA-15042: ASM disk "54" is missing from group number "1"
ORA-15042: ASM disk "53" is missing from group number "1"
ORA-15042: ASM disk "52" is missing from group number "1"
ORA-15042: ASM disk "51" is missing from group number "1"
ORA-15042: ASM disk "50" is missing from group number "1"
ORA-15042: ASM disk "39" is missing from group number "1"
ORA-15042: ASM disk "38" is missing from group number "1"
ORA-15042: ASM disk "37" is missing from group number "1"
ORA-15042: ASM disk "36" is missing from group number "1"
ORA-15042: ASM disk "35" is missing from group number "1"
ORA-15042: ASM disk "34" is missing from group number "1"
ORA-15042: ASM disk "33" is missing from group number "1"
ORA-15042: ASM disk "32" is missing from group number "1"
ORA-15042: ASM disk "31" is missing from group number "1"
ORA-15042: ASM disk "30" is missing from group number "1"

This is fixed with Bug:

16769943 EXADATA: SYSTEM OUTAGE - FOUR CELL NODES IN UNKNOWN STATUS TO ASM
Categories: Oracle in general Tags:

ORA-29760: instance_number parameter not specified — without INSTANCE_NUMBER set

July 29th, 2016 No comments

Oracle is complaining about:

ORA-29760: instance_number parameter not specified

And you have triple checked that the INSTANCE_NUMBER parameter is specified, but you still get the error.

The reason for that is, that your ORACLE_SID contains a underscore, e.g. ORA_SB. Starting with Oracle 12c the *underscore* parameter is a reserved character to indicate a server pool in a RAC.
Blase do not use the underscore parameter in the SID and the problem is gone.

Categories: Oracle in general Tags:

ORA-00902: invalid datatype on MDSYS.SDO_GEORASTER during CREATE TABLE in 12c

May 19th, 2015 No comments

Since i haven´t found this on the web:

When you try to create a table with MDSYS.SGO_GEORASTER and you´re receiving an ORA-00902: invalid datapye try the following:

SQL> @create_tab.sql
        "GEORASTER" "MDSYS"."SDO_GEORASTER" ,
                            *
ERROR at line 12:
ORA-00902: invalid datatype

Try to grant EXECUTE permissions on the data type to the user creating the table.

Categories: Oracle in general Tags:

RAC refuses to start – or how installing TSM caused a cluster outage

April 20th, 2015 No comments

Some days back i setup a new server for a customer of mine. Installation and patching ran fine, cluster was stable – no issues.

Then suddenly the cluster started to act strange. Commands did not work anymore; after a reboot the closure refused to start. The log files showed:

[root@node1 trace]# cat crsctl_60236.trc
Trace file /u01/app/oracle/diag/crs/node1/crs/trace/crsctl_60236.trc
Oracle Database 12c Clusterware Release 12.1.0.2.0 - Production Copyright 1996, 2014 Oracle. All rights reserved.
2014-11-18 18:55:52.765853 :  OCRMSG:2014938688: prom_waitconnect: CONN NOT ESTABLISHED (0,9,1,2)
2014-11-18 18:55:52.765870 :  OCRMSG:2014938688: GIPC error [9] msg [gipcretPermissions]
   CLWAL:2014938688: clsw_Initialize: Error [32] from procr_init_ext
  CLWAL:2014938688: clsw_Initialize: Error [PROCL-32: Oracle High Availability Services on the local node is not running Messaging error [gipcretPermissions] [9]] from procr_init_ext
2014-11-18 18:55:52.766123 :    GPNP:2014938688: clsgpnpkww_initclswcx: [at clsgpnpkww.c:351] Result: (56) CLSGPNP_OCR_INIT. (:GPNP01201:)Failed to init CLSW-OLR context. CLSW Error (3): CLSW-3: Error in the cluster registry (OCR) layer. [32] [PROCL-32: Oracle High Availability Services on the local node is not running Messaging error [gipcretPermissions] [9]]
2014-11-18 18:55:52.781979 : GIPCNET:2014938688:  gipcmodNetworkProcessConnect: [network]  failed connect attempt endp 0x139a810 [000000000000003e] { gipcEndpoint : localAddr 'clsc://(ADDRESS=(PROTOCOL=ipc)(KEY=)(GIPCID=00000000-00000000-0))', remoteAddr 'clsc://(ADDRESS=(PROTOCOL=ipc)(KEY=OCSSD_LL_node1_)(GIPCID=00000000-00000000-0))', numPend 0, numReady 1, numDone 0, numDead 0, numTransfer 0, objFlags 0x0, pidPeer 0, readyRef (nil), ready 0, wobj 0x136c6e0, sendp 0x136c4a0 status 13flags 0xa02c0712, flags-2 0x1, usrFlags 0x14000 }, req 0x13684c0 [0000000000000048] { gipcConnectRequest : addr 'clsc://(ADDRESS=(PROTOCOL=ipc)(KEY=OCSSD_LL_node1_)(GIPCID=00000000-00000000-0))', parentEndp 0x139a810, ret gipcretPermissions (9), objFlags 0x0, reqFlags 0x2 }
2014-11-18 18:55:52.781998 : GIPCNET:2014938688:  gipcmodNetworkProcessConnect: slos op  :  sgipcnDSConnectHelper
2014-11-18 18:55:52.782001 : GIPCNET:2014938688:  gipcmodNetworkProcessConnect: slos dep :  Permission denied (13)
2014-11-18 18:55:52.782003 : GIPCNET:2014938688:  gipcmodNetworkProcessConnect: slos loc :  connect
2014-11-18 18:55:52.782009 : GIPCNET:2014938688:  gipcmodNetworkProcessConnect: slos info:  failed to /var/tmp/.oracle/sOCSSD_LL_node1_
2014-11-18 18:55:52.782215 : CSSCLNT:2014938688: clsssConnect: gipc request failed with 9 (0x31)
2014-11-18 18:55:52.782274 : CSSCLNT:2014938688: clsssInitNative: connect to (ADDRESS=(PROTOCOL=ipc)(KEY=OCSSD_LL_node1_)) failed, rc 9
2014-11-18 18:55:52.782442 :  CRSCTL:2014938688: GetParser::getClustermode: crsctl_cssctx failed with 6

The interesting bit is the “permission denied” thing. It most cases it points to a file system permission issue. As you can see from the code the affected directory is “/var/tmp/.oracle” which (at least on an Exacta) is a link to “/tmp”.

After looking carefully i saw that the directory permission for “/tmp” got changed:

drwxr-xr--   12 root root      4096 Oct  8 23:56 tmp

Which is not right! Luckily the fix is easy:

chmod 1777 /tmp

But what caused the issue? It turned out one of the things i´ve done was to install TSM. I´ve extracted the TAR archive in /tmp and installed it from there. I tried to reproduce the issue and succeeded indeed. By extracting the TAR file the directory permissions of the parent directory was changed:

BEFORE:

drwxrwxrwt   12 root root      4096 Nov 19 12:24 tmp

AFTER:

drwxr-xr--   12 root root      4096 Oct  8 23:56 tmp

Great work, IBM!

Categories: Oracle in general Tags:

Parity-based redundancy (RAID5/6/triple parity and beyond) on BTRFS and MDADM (Dec 2014)

December 10th, 2014 4 comments

While reading the following posts (second post) from Adam Leventhal

I have wondered what is exactly the status of parity-based redundancy in general and triple-parity in particular on Linux in MDADM and BTRFS.

TL;DR: There are patches to extend the linux kernel to support up to 6 parity disks but BTRFS does not want them because it does not fit their “business case” and MDADM would want them but somebody needs to develop patches for the MDADM component. The kernel raid implementation is ready and usable. If someone volunteers to do this kind of work i would support with equipment and myself as a test ressource.

BTRFS has some preliminary support for RAID5/6 but it is completely unusable in it´s current state. Some highlights:

  • As of kernel version 3.14, it works if everything goes right (that is: no errors like failing disks or corruptions), but the error handling is still lacking.
  • scrub cannot fix issues with raid5/6 yet. This means that if you have any checksum problem, your filesystem will be in a bad state.
  • btrfs does not yet seem to recognise that if you remove a drive from an array and then later plug it back in the drive is then out of date.
  • btrfs does not handle very well a drive that is present but not working : for example, an attempt to remove a faulty drive from the array (btrfs device delete) fails because it causes or requires reading from the faulty drive itself and thus btrfs will continue to attempt to access the faulty drive forever.
  • … plus some minor but not to critical things

So much for the parity-based status of BTRFS in general – it is evident it is still in its infancy and needs to mature as in its current state it is unusable. (As of December 2014 i´ve seen some patches appearing on the mailing list intended to fix some of these problems but they are not yet in the code.)

During my research I also found some patches from Andrea Mazzoleni that would extend the kernels ability to have up to six parity disks – per file system. Patches to add this functionality to BTRFS were also supplied (see here for the patches).
For use cases where six parity disks are not needed you can of course also pick three or four parity disks – it´s up to you to decide. This would enable BTRFS to create rather large local file systems with a very good resiliency against data loss.

I have tested the patches and they work just fine. So i wondered why they were not added as they would provide a unique feature to BTRFS no other operating system or file system (not even ZFS) has.

My question on the mailing list and the answer can be see here:

LINK #1, LINK #2 and the answer here. Tests with the patches installed can be found here.

Still not satisfied with the answer i mailed some people offline and got the following responses i´d like to show here::

Our plan is based on the upper distributed fs or storage, which can provide
higher reliability than local filesystems, so we think RAID1/RAID10/RAID5/RAID6
is enough for us.

Your work is very very good, it just doesn’t fit our business case.

And (from a different person):

It would be there some day in the long run I guess. But for now
bringing in the support for enterprise use cases and stability
to the btrfs has been our focus.

If course I do agree with the point that stability matters most given the ongoing problems and bugs with BTRFS that can easily cause a loss of a whole file system (e.g. the most recent problems with snapshots in 3.16 and 3.17 kernels, system lockups, poor performance, space problems and so on) but if a file system also includes the functionality of a volume manager then it needs to provide at least some basic features such as:

  • parity-based redundancy
  • hot-spare disks or n-way-mirroring (see below)
  • mirroring
  • striping
  • n-way-mirroring if there is not support for hot-spare disks (that means more than one copy of the data to protect against a failure of more than one disk in a mirrored configuration)

Currently the *only* working implementation which offers resiliency against disk failures that BTRFS provides is mirroring. Ultimately this means RAID1 for a file system with exactly two disks or RAID1+0 for a file system with more than two disks. A failure of more than one disk is fatal in every case. Hot-spares? Not available and nobody is working on it. N-Way-Mirroring? Not available and nobody claimed it. RAID5/6? You guess it… not available/usable at the moment. No mirroring at all? That´s not what i´d use BTRFS for as the whole idea of a checksum based file system is to detect and repair corruptions which is impossible in a scenario with no redundancy.

The claimed on ongoing feature development for BTRFS can be found here. ##link here to the wiki list## As you see – nobody is working on hot-spare, n-way-mirroring support or RAID5/RAID6.

So returning to the answer I got offline i´d like to ask what the overall goal of BTRFS is and for whom it is developed – the so called “(our) business case” in the cited text. To me it seems the focus is not to be a general purpose linux file system; instead it seems to be more a file system for large enterprises where the data redundancy is already made at lower (= storage) levels. Support for large local file systems like ZFS offers is not available and (according to my impression) not desired.

Now if you read through all of this and you want to comment like

“… then develop and post patches – they´d be highly welcomed”

then I tell you: They are obviously not. At least on BTRFS.

The patches posted were working fine and can easily be adapted for the upstream kernel. Given the premature status of the RAID5/6 support in BTRFS it´s the ideal time to add them – before a lot of work is put into the RAID5/6 code. So why hesitating?

Neil Brown who is supporting the kernel functions for the raid code would send the patches upstream immediately – if someone (that means either MDADM itself or BTRFS) would use them. MDADM patches are not ready (yet?) and BTRFS does´t want them it seems.

So yeah.. so much for BTRFS and parity-based redundancy.

For MDADM it looks a bit better. The effort needed to change the code in MDADM is a non-trivial one and currently Andrea Mazzoleni who developed the patches does not have any time to work on a MDADM patch. I cannot do it myself as I am a pretty bad software developer… If you think you can do it i´d be happy to help you as a tester or with equipment so that at least MDADM get´s a much better parity implementation. Please contact me if you are willing to do this.

So which options are left for large local file systems on Linux?

You could either use ZFS on Linux which support up to three parity disks per VDEV (a number of disks are grouped into a VDEV onto which the chosen parity is applied) and then scale by adding more VDEVs – or you could use several smaller RAID6 (i would not recommend RAID5 due to the high risk of losing the raid if a disk fails and you need to rebuild) and group them together with LVM on top. Leaving the licensing discussion aside ZFS on Linux works quite well but requires a lot of manual work to get it working and running. If it runs it is stable – just make sure you use the latest ZFS code. On top of that ZFS is by far more mature than BTRFS while offering all of it´s features and more.

Categories: UNIX Tags:

Finally: I am now Oracle Certified Master 11g

November 28th, 2014 3 comments

Yes! I made it. I successfully accomplished the exam and i am now a proud member of the OCM family:

OCM 11g Zertifikat Small

Categories: Oracle in general Tags: