Copy of Logical logs for DR
Posted in 2008
Vikas wanted logical logs copied to a remote server for DR on IDS 11.10 (Solaris 10), with HDR/ER ruled out by cost and network-mounted mirroring unsupported. Suggestions: Continuous Log Restore, virtual-tape/backup products (EMC CDL + Networker), but the accepted cheap approach was ontape -a backup to a directory driven by ALARMPROGRAM on event class 23 (logical log complete), then scp/rcp the file (Art Kagel's eventalarm.c from utils3_ak was mentioned). His script always lagged one log behind; the fix was to move the LOGFILE=`ls -ltr ... | tail -1` line inside the IF, just before the scp. Follow-up CLR questions were moved to a new thread.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Topics: High Availability & Replication, Storage & Space Management, Logging & Checkpoints, Platform-Specific Issues
Hello all, IDS 11.10.FC2W4 on Solaris 10 Not so good question but i would like to know any ways to replicate/copy my logical log files on to a remote server HDR is the best way to do it but NOT an option for me :( Mirroring the logdbs dbspace is another option but "Mirroring is not supported on disks that are managed over a network" so if i have a backup server over a network where i wish to mirror my logical logs then,will the NFS mount work for me?? what am i left with, manually scp the logical log files to the other location,after regular intervals - shell script running through cron :( always with a risk of loosing few logs. kindly suggest me if there is a better way of doing this (other than HDR/ER). I reffered one similar thread. http://www.iiug.org/forums/ids/index.cgi/noframes/read/2876 Regards vikas
Why HDR/RSS can not suite your needs?. Anyway, you can consider Continuous Log Restore (CLR). But in order to help you, we need to understand what are your requirements... Regards On Thu, Sep 11, 2008 at 11:34 AM, VIKAS HIVARKAR <vikas.hivarkar@gmail.com>wrote: > Hello all, > > IDS 11.10.FC2W4 on Solaris 10 > > Not so good question but i would like to know any ways to replicate/copy my > logical log files on to a remote server > HDR is the best way to do it but NOT an option for me :( > > Mirroring the logdbs dbspace is another option but > "Mirroring is not supported on disks that are managed over a network" > so if i have a backup server over a network where i wish to mirror my > logical > logs then,will the NFS mount work for me?? > > what am i left with, manually scp the logical log files to the other > location,after regular intervals - shell script running through cron :( > always with a risk of loosing few logs. > > kindly suggest me if there is a better way of doing this (other than > HDR/ER). > > I reffered one similar thread. > http://www.iiug.org/forums/ids/index.cgi/noframes/read/2876 > > Regards > vikas > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > -- -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently...
On Thu, Sep 11, 2008 at 7:40 AM, Fernando Nunes <domusonline@gmail.com> wrote: > Why HDR/RSS can not suite your needs?. > Anyway, you can consider Continuous Log Restore (CLR). > But in order to help you, we need to understand what are your > requirements... > > Regards > > On Thu, Sep 11, 2008 at 11:34 AM, VIKAS HIVARKAR > <vikas.hivarkar@gmail.com>wrote: > >> Hello all, >> >> IDS 11.10.FC2W4 on Solaris 10 >> >> Not so good question but i would like to know any ways to replicate/copy my >> logical log files on to a remote server >> HDR is the best way to do it but NOT an option for me :( >> >> Mirroring the logdbs dbspace is another option but >> "Mirroring is not supported on disks that are managed over a network" >> so if i have a backup server over a network where i wish to mirror my >> logical >> logs then,will the NFS mount work for me?? >> >> what am i left with, manually scp the logical log files to the other >> location,after regular intervals - shell script running through cron :( >> always with a risk of loosing few logs. >> >> kindly suggest me if there is a better way of doing this (other than >> HDR/ER). >> >> I reffered one similar thread. >> http://www.iiug.org/forums/ids/index.cgi/noframes/read/2876 >> >> Regards >> vikas >> >> >> >> > ******************************************************************************* >> Forum Note: Use "Reply" to post a response in the discussion forum. >> >> > > -- > -- > Fernando Nunes > Portugal > > http://informix-technology.blogspot.com > My email works... but I don't check it frequently... > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > Some people like to do this as an extra form of insurance on top of HDR/SDS/etc. We use the alarm script to key off event class 23 (LOGICAL LOG FILE COMPLETED) and kick off a script to do just that. Jim
Have you checked use EMC's Clariion Disk Library (CDL) a virtual tape product that allows Networker (which it is used to back up the logical logs, and everthing else BTW) to control writes to tapes. There must be other products too. I think that you are looking for something that is out of the Database Management scope and more on the backup infrastructure side. Not a cheap solution,but what the heck DR is never cheap. Walter Milan DBA -----Original Message----- From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Fernando Nunes Sent: Thursday, September 11, 2008 6:41 AM To: ids@iiug.org Subject: Re: Copy of Logical logs for DR [13338] Why HDR/RSS can not suite your needs?. Anyway, you can consider Continuous Log Restore (CLR). But in order to help you, we need to understand what are your requirements... Regards On Thu, Sep 11, 2008 at 11:34 AM, VIKAS HIVARKAR <vikas.hivarkar@gmail.com>wrote: > Hello all, > > IDS 11.10.FC2W4 on Solaris 10 > > Not so good question but i would like to know any ways to replicate/copy my > logical log files on to a remote server > HDR is the best way to do it but NOT an option for me :( > > Mirroring the logdbs dbspace is another option but > "Mirroring is not supported on disks that are managed over a network" > so if i have a backup server over a network where i wish to mirror my > logical > logs then,will the NFS mount work for me?? > > what am i left with, manually scp the logical log files to the other > location,after regular intervals - shell script running through cron :( > always with a risk of loosing few logs. > > kindly suggest me if there is a better way of doing this (other than > HDR/ER). > > I reffered one similar thread. > http://www.iiug.org/forums/ids/index.cgi/noframes/read/2876 > > Regards > vikas > > > > ************************************************************************ ******* > Forum Note: Use "Reply" to post a response in the discussion forum. > > -- -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... ************************************************************************ ******* Forum Note: Use "Reply" to post a response in the discussion forum.
@Fernando >Why HDR/RSS can not suite your needs?. >Anyway, you can consider Continuous Log Restore (CLR). >But in order to help you, we need to understand what are your >requirements... HDR suits my needs 100% but not to the business needs, i tried to convience for having HDR but .. Continues log restore is absouletly a perfect substitue for HDR but then it would require another instance ie exact replica of the primary instance,which calls for another server with alteast 500 GB spcace and you can understand if no HDR then would i get it. My requirement is very much as explained by Jim. ahh i have to thank you for your wonderfull article on AUDITING.. @Jim >Some people like to do this as an extra form of insurance on top of >HDR/SDS/etc. We use the alarm script to key off event class 23 >(LOGICAL LOG FILE COMPLETED) and kick off a script to do just that. well your way of making the copy of logical logs suits my requirement and would appreciate if you could share more of it. @Walter >Have you checked use EMC's Clariion Disk Library (CDL) a virtual tape >product that allows Networker (which it is used to back up the logical >logs, and everthing else BTW) to control writes to tapes. There must be >other products too. I think that you are looking for something that is >out of the Database Management scope and more on the backup >infrastructure side. >Not a cheap solution,but what the heck DR is never cheap. I wish i could go for these High end solutions :( Thanks Fenando,Jim and Walter I really appreciate your replies.. Regards vikas
On Thu, Sep 11, 2008 at 3:30 PM, VIKAS HIVARKAR
<vikas.hivarkar@gmail.com>wrote:
> @Fernando
> >Why HDR/RSS can not suite your needs?.
> >Anyway, you can consider Continuous Log Restore (CLR).
> >But in order to help you, we need to understand what are your
> >requirements...
>
> HDR suits my needs 100% but not to the business needs, i tried to convience
> for having HDR but ..
>
> Continues log restore is absouletly a perfect substitue for HDR but then it
> would require another instance ie exact replica of the primary
> instance,which
> calls for another server with alteast 500 GB spcace and you can understand
> if
> no HDR then would i get it.
I see... hardware costs money...
>
>
> My requirement is very much as explained by Jim.
> ahh i have to thank you for your wonderfull article on AUDITING..
>
Appreciation always feel good. Thanks.
>
> @Jim
> >Some people like to do this as an extra form of insurance on top of
> >HDR/SDS/etc. We use the alarm script to key off event class 23
> >(LOGICAL LOG FILE COMPLETED) and kick off a script to do just that.
>
> well your way of making the copy of logical logs suits my requirement and
> would appreciate if you could share more of it.
>
> @Walter
> >Have you checked use EMC's Clariion Disk Library (CDL) a virtual tape
> >product that allows Networker (which it is used to back up the logical
> >logs, and everthing else BTW) to control writes to tapes. There must be
> >other products too. I think that you are looking for something that is
> >out of the Database Management scope and more on the backup
> >infrastructure side.
>
> >Not a cheap solution,but what the heck DR is never cheap.
>
> I wish i could go for these High end solutions :(
>
> Thanks Fenando,Jim and Walter
> I really appreciate your replies..
>
Well... On IDS 11 you can backup to directories... After backing up you can
copy the files to another server... Can't get cheaper than this...
Every time a logical log fills up you get an ALARM event. You can setup your
ALARMPROGRAM to do something... like ontape... followed by copy to another
server...
If you just want to put the logical logs in another server it's pretty
flexible.... But this will work for logs already filled... Not for the
current log...
Regards,
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
I do not know how you are archiving your logical logs currently, but if you
use the ALARMPROGRAM to do that, the program or script you register as the
ALARMPROGRAM can capture the event (23 IB) and run an ontape -a or onbar
then once the archive is completed rename and compress the archive file (the
new name should indicate the logical log(s) that it contains to simplify
restores), rcp or scp the file to a remote repository and then recreate the
empty TAPEDEV file (if you are using ontape).
We did this at Bloomberg for many years using my ALARMPROGRAM eventalarm.c
which you can download. It is in the package utils3_ak downloadable from
the Oninit web site (www.oninit.com/utils) or from the IIUG Software
Repository (www.iiug.org/software).
Art
On Thu, Sep 11, 2008 at 6:34 AM, VIKAS HIVARKAR
<vikas.hivarkar@gmail.com>wrote:
> Hello all,
>
> IDS 11.10.FC2W4 on Solaris 10
>
> Not so good question but i would like to know any ways to replicate/copy my
> logical log files on to a remote server
> HDR is the best way to do it but NOT an option for me :(
>
> Mirroring the logdbs dbspace is another option but
> "Mirroring is not supported on disks that are managed over a network"
> so if i have a backup server over a network where i wish to mirror my
> logical
> logs then,will the NFS mount work for me??
>
> what am i left with, manually scp the logical log files to the other
> location,after regular intervals - shell script running through cron :(
> always with a risk of loosing few logs.
>
> kindly suggest me if there is a better way of doing this (other than
> HDR/ER).
>
> I reffered one similar thread.
> http://www.iiug.org/forums/ids/index.cgi/noframes/read/2876
>
> Regards
> vikas
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Art S. Kagel
Oninit (www.oninit.com)
IIUG Board of Directors (art@iiug.org)
Disclaimer: Please keep in mind that my own opinions are my own opinions and
do not reflect on my employer, Oninit, the IIUG, nor any other organization
with which I am associated either explicitly or implicitly. Neither do those
opinions reflect those of other individuals affiliated with any entity with
which I am affiliated nor those of the entities themselves.
I'm using "ontape -a -d" for backing up my logical logs with directory feature.
I tried to following additions to the alarmprogram.sh script
RHOST=testdb05
LOGPATH=/usr/informix/logicallog # Dir for logical log files
LOGFILE=`ls -ltr $LOGPATH |tail -1 | awk -F" " '{print $9}'`
if ( `test x${BACKUPLOGS} = xY` ) then
$BACKUP_CMD 2>&1 >> /dev/null
EXIT_STATUS=$?
scp -q $LOGPATH/$LOGFILE $RHOST:$LOGPATH &
else
echo "Script will not backup the logical logs."
fi
I also tried with
if ( `test x${BACKUPLOGS} = xY` ) then
$BACKUP_CMD 2>&1 >> /dev/null
EXIT_STATUS=$?
if [ $EXIT_STATUS -eq 0 ]; then
scp -q $LOGPATH/$LOGFILE $RHOST:$LOGPATH &
fi
else
echo "Script will not backup the logical logs."
fi
But it copies the 2nd last file in $LOGFILE and the last backed up file is not
copied to remote location untill there is a new file in the $LOGFILE
so at any time i have 1 log file less copied at the $RHOST location.
surely I'm missing some thing very basic..
>I do not know how you are archiving your logical logs currently, but if you
>use the ALARMPROGRAM to do that, the program or script you register as the
>ALARMPROGRAM can capture the event (23 IB) and run an ontape -a or onbar
>then once the archive is completed rename and compress the archive file (the
>new name should indicate the logical log(s) that it contains to simplify
>restores), rcp or scp the file to a remote repository and then recreate the
>empty TAPEDEV file (if you are using ontape).
>We did this at Bloomberg for many years using my ALARMPROGRAM eventalarm.c
>which you can download. It is in the package utils3_ak downloadable from
>the Oninit web site (www.oninit.com/utils) or from the IIUG Software
>Repository (www.iiug.org/software).
>Art
>On Thu, Sep 11, 2008 at 6:34 AM, VIKAS HIVARKAR
<vikas.hivarkar@gmail.com>wrote:
>> Hello all,
>>
>> IDS 11.10.FC2W4 on Solaris 10
>>
>> Not so good question but i would like to know any ways to replicate/copy my
>> logical log files on to a remote server
>> HDR is the best way to do it but NOT an option for me :(
>>
>> Mirroring the logdbs dbspace is another option but
>> "Mirroring is not supported on disks that are managed over a network"
>> so if i have a backup server over a network where i wish to mirror my
>> logical
>> logs then,will the NFS mount work for me??
>>
>> what am i left with, manually scp the logical log files to the other
>> location,after regular intervals - shell script running through cron :(
>> always with a risk of loosing few logs.
>>
>> kindly suggest me if there is a better way of doing this (other than
>> HDR/ER).
>>
>> I reffered one similar thread.
>> http://www.iiug.org/forums/ids/index.cgi/noframes/read/2876
>>
>>Regards
>> vikas
>>
>>
>>
oh, i have also tried delaying the scp by adding sleep 10. my logical log file are of 20M and does not take much time for a file to be created on the system..
My senior just gave me the solution it was to add the line in the IF clause just before the scp command. LOGFILE=`ls -ltr $LOGPATH |tail -1 | awk -F" " '{print $9}'` Thanks...
Hello
IDS 11.10.FC2W4
I want to try out the continous log restore using ontape.
The version of Dynamic Server is identical on both the primary and secondary
systems.
From the manual i read the following steps:
1.On the primary system, perform a level-0 archive with the following command:
ontape -s -L 0
2.On the secondary system, copy the files or mount the tape (as assigned by
LTAPEDEV) and perform a physical restore with the following command:
ontape -p
Respond to the following prompts:
Continue restore? Y
Do you want to back up the logs? N
Restore a level 1 archive? N
After the physical restore completes, the database instance waits in fast
recovery mode to restore logical logs.
3.On the primary system, back up logical logs with the following command:
ontape -a
-- I'm using the directory feature to backup the logical logs and
alarmprogram.sh to triger the logical log backup on event 23 so i have a
directory where all the logs from the last level 0 are available
4.On the secondary system, copy the files or mount the tape that contains the
backed up logical logs from the primary system. Perform a logical log restore
with the following command:
ontape -l -C
-- After the step 2 the db is in fast recovery mode,now do i have to run
ontape -l -C ( considering that I'm copying every logical log generated on theprimray to the secondry in directory to which the LTAPEDEV points )in this
mode?
5.Repeat steps 3 and 4 for all logical logs that are available to back up and
restore.
--supposing that step 4 holds good and i finish restoring all the logical logs
from primary on to this secondry then do i have to follow the step 6 as it is?
6.If you are doing continuous log restore on a secondary system as an
emergency standby, run the following commands to complete restoring logical
logs and quiesce the server.
If logical logs are available to restore:
ontape -l
After all available logical logs are restored:
ontape -l -X
-- whats does the above 2 command accomplish?
If a new log file is generated on primary and is copied on to secondry then
will that log be auto restored on the secondry ?
What would be the state of secondry as long as continous log restoration is
enabled? would it be read only as HDR or always in quescient mode ?
Let me know if i have not been able to clearly mention my doubts and Pls
simplify the working of continous logical log restoration...
Thanking you
Regards
Vikas.
---------------------------------------------------------------------
Why HDR/RSS can not suite your needs?.
Anyway, you can consider Continuous Log Restore (CLR).
But in order to help you, we need to understand what are your
requirements...
Regards
pls ignore the last request as i have started a new thread. thanks