SAN i/o in RAC [message #573987] |
Fri, 04 January 2013 13:42 |
|
Hi,
We have 11.2.0.3 RAC on Oracle Virtual Box,OEL 5.8. Storage is ASM on EMC SAN.
Even when the database is not connected to application and thus open but idle and none is working on it, and all other virtual machines are powered off, we are still getting a continuous i/o in SAN.Since I am new to RAC setup I need to know the few things...
1.Is it normal behaviour?
2.If yes, what is the reason of this i/o? Is it due to continuous ping on voting disk?..write in OCR? automatic backup of OCR?
Please find the attached screenshot.If anyone kindly explain the reason, I will be grateful.
Regards,
Saikat
|
|
|
|
Re: SAN i/o in RAC [message #573990 is a reply to message #573987] |
Fri, 04 January 2013 14:34 |
John Watson
Messages: 8963 Registered: January 2010 Location: Global Village
|
Senior Member |
|
|
Can you confirm your environment? Do you mean that your VirtualBox VMs are running from VDI files on file systems mounted from the EMC SAN?
|
|
|
Re: SAN i/o in RAC [message #573994 is a reply to message #573990] |
Fri, 04 January 2013 21:09 |
|
Quote: Can you confirm your environment? Do you mean that your VirtualBox VMs are running from VDI files on file systems mounted from the EMC SAN?
Yes.exactly.
|
|
|
Re: SAN i/o in RAC [message #573995 is a reply to message #573989] |
Fri, 04 January 2013 21:18 |
|
Quote:Why are you focused on I/O activity; as opposed to CPU or Network activity?
Why does system continuously burn CPU cycles? What is causing all this activity?
yes..I am also unware why ntw and cpu too showing this behavior. If anyone from their experience/test setup kindly let me know their output and explain the reason I will be grateful.
Regards,
Saikat
|
|
|
Re: SAN i/o in RAC [message #574009 is a reply to message #573994] |
Sat, 05 January 2013 01:32 |
John Watson
Messages: 8963 Registered: January 2010 Location: Global Village
|
Senior Member |
|
|
I use VirtualBox all the time, I'm running a RAC right now on my PC: two VMs for the RAC nodes, and third VM as the DHCP/DNS/storage server. There is always continuous i/o and CPU usage. But I would never use VB for anything other than test/demo/development environments, certainly not for any application that required EMC storage. So sorry, I can't advise further.
|
|
|
Re: SAN i/o in RAC [message #574046 is a reply to message #574009] |
Sun, 06 January 2013 02:27 |
|
Hi John,
Thanks a lot for your reply.
Actually this is not a production server. This is only for testing, learning , practice and building competency on oracle products. We work on various products apart from oracle.That's why we used virtual box.
Since our budget could afford EMC storage we used it to simulate near real-time environment.
Regards,
Saikat
|
|
|
Re: SAN i/o in RAC [message #579662 is a reply to message #574046] |
Wed, 13 March 2013 18:08 |
|
alan.kendall@nfl.com
Messages: 163 Registered: June 2012 Location: Culver City, California
|
Senior Member |
|
|
You can look at the historic physical I/O saved within the hourly snapshots (if statistics_level=typical) with the following sql:
ENDOCP1P > @physical
DATE PHYSICAL_READS_LAST_HOUR OBJECT_NAME
-------------------- ------------------------ --------------------
2013-11-03 Monday 6054 D_1F000D5D80000910
2013-11-03 Monday 26179 D_1F000D5D80000005
2013-11-03 Monday 40332 D_1F000D5D80000901
2013-11-03 Monday 46808 DM_SYSOBJECT_S_COMP2
2013-11-03 Monday 64930 DM_RELATION_S
2013-11-03 Monday 78228 NFL_CONTENT_R_COMP1
2013-11-03 Monday 79647 DM_SYSOBJECT_S_COMP1
2013-11-03 Monday 135421 DMR_CONTENT_S
2013-11-03 Monday 165087 NFL_CONTENT_R
2013-11-03 Monday 243480 DM_SYSOBJECT_R_COMP1
2013-11-03 Monday 548052 D_1F000D5D8000010A
2013-11-03 Monday 1475076 DM_SYSOBJECT_R_COMP2
ENDOCP1P > list
1 select to_char(c.BEGIN_INTERVAL_TIME,'YYYY-MM-DD Day') "DATE",
2 sum(b.PHYSICAL_READS_DELTA) PHYSICAL_READS_LAST_HOUR,
3 a.object_name
4 from dba_objects a,dba_hist_seg_stat b,sys.wRM$_SNAPSHOT c
5 where b.SNAP_ID =(select max(SNAP_ID) from sys.wRM$_SNAPSHOT)
6 and a.object_id=b.OBJ#
7 and b.PHYSICAL_READS_DELTA>0
8 and c.instance_number=(select instance_number from v$instance)
9 and c.snap_id=b.snap_id
10 group by to_char(c.BEGIN_INTERVAL_TIME,'YYYY-MM-DD Day'),a.object_name
11* order by 2
If you are not saving hourly shapshots, you can see what is being written by taking a backup of gv$segment_statistics and comparing it with gv$segment_statistics.
ECSCDAP1P > @io605
INSTANCE IO_PER_MINUTE STATISTIC_NAME OBJECT_NAME
--------- ------------- ----------------------- -------------------------------------
CSCDAP1 12 physical reads SYS.IDL_UB2$.
CSCDAP1 12 physical reads SYS.IDL_CHAR$.
CSCDAP1 12 physical read requests SYS.IDL_UB2$.
CSCDAP1 12 physical read requests SYS.IDL_CHAR$.
CSCDAP1 84 physical reads SYS.I_OBJAUTH2.
CSCDAP1 84 physical read requests SYS.I_OBJAUTH2.
CSCDAP3 240 physical read requests CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP3 252 physical reads direct CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP3 252 physical reads CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP3 276 physical reads CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP3 276 physical read requests CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP3 276 physical reads direct CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP4 288 physical read requests CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP1 288 physical reads direct CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP1 288 physical reads CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP1 288 physical read requests CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP1 300 physical reads direct CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP1 300 physical reads CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP1 300 physical read requests CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP4 312 physical reads direct CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP4 312 physical reads CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP4 360 physical reads direct CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP4 360 physical reads CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP4 360 physical read requests CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP2 384 physical read requests CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP2 396 physical reads CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP2 396 physical reads direct CDA_SHARED.SYS_LOB0000077712C00008$$.
CSCDAP2 480 physical read requests CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP2 492 physical reads CDA_SHARED.SYS_LOB0000077712C00004$$.
CSCDAP2 492 physical reads direct CDA_SHARED.SYS_LOB0000077712C00004$$.
-------------
sum 8184
io605.sql follows:
set termout off
set lines 160
break on report
col instance for a9
compute sum of io_per_minute break on report
drop table gv$segment_statistics1;
drop table gv$segment_statistics2;
column statistic_name format a23
create table gv$segment_statistics1 as select * from gv$segment_statistics
where upper(statistic_name) like '%PHYSICAL%' and owner not like 'E%';
execute dbms_lock.sleep(5);
create table gv$segment_statistics2 as select * from gv$segment_statistics
where upper(statistic_name) like '%PHYSICAL%' and owner not like 'E%';
set wrap off
set termout on
select i.instance_name instance,(b.value-a.value)*12 IO_PER_MINUTE,
a.statistic_name,
a.owner||'.'||a.object_name||'.'||a.subobject_name object_name
from gv$segment_statistics2 b,gv$segment_statistics1 a,gv$instance i
where a.statistic_name=b.statistic_name
and a.inst_id=i.inst_id and b.inst_id=i.inst_id
and a.inst_id=b.inst_id and a.object_type=b.object_type
and a.owner||'.'||a.object_name||'.'||a.subobject_name=
b.owner||'.'||b.object_name||'.'||b.subobject_name
and upper(a.statistic_name) like '%PHYSICAL%'
and a.object_name not like 'GV$SEGMENT_STATISTICS%'
and b.value-a.value > 0
order by b.value-a.value;
set termout off
drop table gv$segment_statistics1 purge;
drop table gv$segment_statistics2 purge;
set termout on
|
|
|