[Info-Ingres] Ingres and Dell-Equalogic SAN

[Database Administrator]

Hi Karl et al,

Well I'd have to say that no I havent tested the WDTD recently...I
suppose now is going to be a good opportunity!

The Dell guys were a little surprised at the entire SSH key exchange
thing. But have provided a perl module which talks to the SAN using
telnet and something vaguely akin to 'expect' scripts.
I have to hard code the password of the SAN manager into my scripts but
other than that it seems to work.

Backup snapshots are easy, but the recovery of the snapshot just stalled
in an uninteruptable disk IO state.

Hmm, back to the drawing board...


Marty



-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl
& Betty Schendel
Sent: 06 November 2008 12:36
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:

> Hi Karl et al,
>
> The deletion phase of the recovery can be controlled by the WDTD flag.
> So rather than do a delete of the data directory, I was simply
> planning
> to do:
> WDTD: /bin/echo At WDTD: Do nothing

That's useful. I assume you've tested this... :-)
Once upon a time, the WDTD only affected added locations for a database,
it would still unconditionally delete the files from the root location!
Which was pretty stupid, of course. I guess that got fixed. That last
time I played with SAN checkpointing in anger was probably Ingres 2.5-
ish.

> At the moment the problem is getting the SSH key exchange to work. It
> just aint. Thus all my SSH commands fail...which is a fairly limiting
> feature!

IMHO, the whole SSH thing is the modern day version of the modem.
It's an invention of the devil, meant to produce frustration... :-)

Karl

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

[Database Administrator]

I haven't done a rollforward in a long time (knock on wood) but as far as I know, the
root data location does get wiped out with 2.6.

Let me know if that has changed in 2006/r3.

Andre

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Martin Bowes
Sent: November 6, 2008 10:34 AM
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN

Hi Karl et al,

Well I'd have to say that no I havent tested the WDTD recently...I suppose now is going to be a good opportunity!

The Dell guys were a little surprised at the entire SSH key exchange thing. But have provided a perl module which talks to the SAN using telnet and something vaguely akin to 'expect' scripts.
I have to hard code the password of the SAN manager into my scripts but other than that it seems to work.

Backup snapshots are easy, but the recovery of the snapshot just stalled in an uninteruptable disk IO state.

Hmm, back to the drawing board...


Marty



-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl & Betty Schendel
Sent: 06 November 2008 12:36
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:

> Hi Karl et al,
>
> The deletion phase of the recovery can be controlled by the WDTD flag.
> So rather than do a delete of the data directory, I was simply
> planning to do:
> WDTD: /bin/echo At WDTD: Do nothing

That's useful. I assume you've tested this... :-) Once upon a time, the WDTD only affected added locations for a database, it would still unconditionally delete the files from the root location!
Which was pretty stupid, of course. I guess that got fixed. That last time I played with SAN checkpointing in anger was probably Ingres 2.5- ish.

> At the moment the problem is getting the SSH key exchange to work. It
> just aint. Thus all my SSH commands fail...which is a fairly limiting
> feature!

IMHO, the whole SSH thing is the modern day version of the modem.
It's an invention of the devil, meant to produce frustration... :-)

Karl

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

[Database Administrator]

Why are you testing recovery. It is so much more fun
to do that when you have 5 other emergencies going on
at the same time and a flooded computer room.

;-)

Martin Bowes wrote:
> Hi Karl et al,
>
> Well I'd have to say that no I havent tested the WDTD recently...I
> suppose now is going to be a good opportunity!
>
> The Dell guys were a little surprised at the entire SSH key exchange
> thing. But have provided a perl module which talks to the SAN using
> telnet and something vaguely akin to 'expect' scripts.
> I have to hard code the password of the SAN manager into my scripts but
> other than that it seems to work.
>
> Backup snapshots are easy, but the recovery of the snapshot just stalled
> in an uninteruptable disk IO state.
>
> Hmm, back to the drawing board...
>
>
> Marty
>
>
>
> -----Original Message-----
> From: info-ingres-bounces@kettleriverconsulting.com
> [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl
> & Betty Schendel
> Sent: 06 November 2008 12:36
> To: Ingres and related product discussion forum
> Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN
>
>
> On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:
>
>> Hi Karl et al,
>>
>> The deletion phase of the recovery can be controlled by the WDTD flag.
>> So rather than do a delete of the data directory, I was simply
>> planning
>> to do:
>> WDTD: /bin/echo At WDTD: Do nothing
>
> That's useful. I assume you've tested this... :-)
> Once upon a time, the WDTD only affected added locations for a database,
> it would still unconditionally delete the files from the root location!
> Which was pretty stupid, of course. I guess that got fixed. That last
> time I played with SAN checkpointing in anger was probably Ingres 2.5-
> ish.
>
>> At the moment the problem is getting the SSH key exchange to work. It
>> just aint. Thus all my SSH commands fail...which is a fairly limiting
>> feature!
>
> IMHO, the whole SSH thing is the modern day version of the modem.
> It's an invention of the devil, meant to produce frustration... :-)
>
> Karl
>
> _______________________________________________
> Info-Ingres mailing list
> Info-Ingres@kettleriverconsulting.com
> http://www.kettleriverconsulting.com...fo/info-ingres
>
> _______________________________________________
> Info-Ingres mailing list
> Info-Ingres@kettleriverconsulting.com
> http://www.kettleriverconsulting.com...fo/info-ingres
>

[Database Administrator]

Why are you testing recovery. It is so much more fun
to do that when you have 5 other emergencies going on
at the same time and a flooded computer room.

;-)

Martin Bowes wrote:
> Hi Karl et al,
>
> Well I'd have to say that no I havent tested the WDTD recently...I
> suppose now is going to be a good opportunity!
>
> The Dell guys were a little surprised at the entire SSH key exchange
> thing. But have provided a perl module which talks to the SAN using
> telnet and something vaguely akin to 'expect' scripts.
> I have to hard code the password of the SAN manager into my scripts but
> other than that it seems to work.
>
> Backup snapshots are easy, but the recovery of the snapshot just stalled
> in an uninteruptable disk IO state.
>
> Hmm, back to the drawing board...
>
>
> Marty
>
>
>
> -----Original Message-----
> From: info-ingres-bounces@kettleriverconsulting.com
> [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl
> & Betty Schendel
> Sent: 06 November 2008 12:36
> To: Ingres and related product discussion forum
> Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN
>
>
> On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:
>
>> Hi Karl et al,
>>
>> The deletion phase of the recovery can be controlled by the WDTD flag.
>> So rather than do a delete of the data directory, I was simply
>> planning
>> to do:
>> WDTD: /bin/echo At WDTD: Do nothing
>
> That's useful. I assume you've tested this... :-)
> Once upon a time, the WDTD only affected added locations for a database,
> it would still unconditionally delete the files from the root location!
> Which was pretty stupid, of course. I guess that got fixed. That last
> time I played with SAN checkpointing in anger was probably Ingres 2.5-
> ish.
>
>> At the moment the problem is getting the SSH key exchange to work. It
>> just aint. Thus all my SSH commands fail...which is a fairly limiting
>> feature!
>
> IMHO, the whole SSH thing is the modern day version of the modem.
> It's an invention of the devil, meant to produce frustration... :-)
>
> Karl
>
> _______________________________________________
> Info-Ingres mailing list
> Info-Ingres@kettleriverconsulting.com
> http://www.kettleriverconsulting.com...fo/info-ingres
>
> _______________________________________________
> Info-Ingres mailing list
> Info-Ingres@kettleriverconsulting.com
> http://www.kettleriverconsulting.com...fo/info-ingres
>

[Database Administrator]

I forgot to mention we copy the snapshots to a separate set of physical disks
before they are backed up to tape.
A restore would involve straight data tranfers between real disks.

Is this how your set up works ? Or is your restore some type of rebuild based directly
on the snapshot ?

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Martin Bowes
Sent: November 6, 2008 10:34 AM
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN

Hi Karl et al,

Well I'd have to say that no I havent tested the WDTD recently...I suppose now is going to be a good opportunity!

The Dell guys were a little surprised at the entire SSH key exchange thing. But have provided a perl module which talks to the SAN using telnet and something vaguely akin to 'expect' scripts.
I have to hard code the password of the SAN manager into my scripts but other than that it seems to work.

Backup snapshots are easy, but the recovery of the snapshot just stalled in an uninteruptable disk IO state.

Hmm, back to the drawing board...


Marty



-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl & Betty Schendel
Sent: 06 November 2008 12:36
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:

> Hi Karl et al,
>
> The deletion phase of the recovery can be controlled by the WDTD flag.
> So rather than do a delete of the data directory, I was simply
> planning to do:
> WDTD: /bin/echo At WDTD: Do nothing

That's useful. I assume you've tested this... :-) Once upon a time, the WDTD only affected added locations for a database, it would still unconditionally delete the files from the root location!
Which was pretty stupid, of course. I guess that got fixed. That last time I played with SAN checkpointing in anger was probably Ingres 2.5- ish.

> At the moment the problem is getting the SSH key exchange to work. It
> just aint. Thus all my SSH commands fail...which is a fairly limiting
> feature!

IMHO, the whole SSH thing is the modern day version of the modem.
It's an invention of the devil, meant to produce frustration... :-)

Karl

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

[Database Administrator]

Hi Mike,

Yes I'm a big wussy, consumed by due care and diligence...that and the
lights on the SAN box look so cool.

Marty

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of
Michael Leo
Sent: 06 November 2008 15:46
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN

Why are you testing recovery. It is so much more fun
to do that when you have 5 other emergencies going on
at the same time and a flooded computer room.

;-)

Martin Bowes wrote:
> Hi Karl et al,
>
> Well I'd have to say that no I havent tested the WDTD recently...I
> suppose now is going to be a good opportunity!
>
> The Dell guys were a little surprised at the entire SSH key exchange
> thing. But have provided a perl module which talks to the SAN using
> telnet and something vaguely akin to 'expect' scripts.
> I have to hard code the password of the SAN manager into my scripts
but
> other than that it seems to work.
>
> Backup snapshots are easy, but the recovery of the snapshot just
stalled
> in an uninteruptable disk IO state.
>
> Hmm, back to the drawing board...
>
>
> Marty
>
>
>
> -----Original Message-----
> From: info-ingres-bounces@kettleriverconsulting.com
> [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of
Karl
> & Betty Schendel
> Sent: 06 November 2008 12:36
> To: Ingres and related product discussion forum
> Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN
>
>
> On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:
>
>> Hi Karl et al,
>>
>> The deletion phase of the recovery can be controlled by the WDTD
flag.
>> So rather than do a delete of the data directory, I was simply
>> planning
>> to do:
>> WDTD: /bin/echo At WDTD: Do nothing
>
> That's useful. I assume you've tested this... :-)
> Once upon a time, the WDTD only affected added locations for a
database,
> it would still unconditionally delete the files from the root
location!
> Which was pretty stupid, of course. I guess that got fixed. That
last
> time I played with SAN checkpointing in anger was probably Ingres 2.5-

> ish.
>
>> At the moment the problem is getting the SSH key exchange to work. It
>> just aint. Thus all my SSH commands fail...which is a fairly limiting
>> feature!
>
> IMHO, the whole SSH thing is the modern day version of the modem.
> It's an invention of the devil, meant to produce frustration... :-)
>
> Karl
>
> _______________________________________________
> Info-Ingres mailing list
> Info-Ingres@kettleriverconsulting.com
> http://www.kettleriverconsulting.com...fo/info-ingres
>
> _______________________________________________
> Info-Ingres mailing list
> Info-Ingres@kettleriverconsulting.com
> http://www.kettleriverconsulting.com...fo/info-ingres
>

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

[Database Administrator]

Hi Andre,

Yep, the root location gets wiped out in 2006R2 regardless of the settings of WDxD. It seems that option only applies to secondary data locations.

Which sucks. It must have something to do with managing the configuration file...

Marty

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Laframboise, André
Sent: 06 November 2008 15:46
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


I haven't done a rollforward in a long time (knock on wood) but as far as I know, the
root data location does get wiped out with 2.6.

Let me know if that has changed in 2006/r3.

Andre

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Martin Bowes
Sent: November 6, 2008 10:34 AM
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN

Hi Karl et al,

Well I'd have to say that no I havent tested the WDTD recently...I suppose now is going to be a good opportunity!

The Dell guys were a little surprised at the entire SSH key exchange thing. But have provided a perl module which talks to the SAN using telnet and something vaguely akin to 'expect' scripts.
I have to hard code the password of the SAN manager into my scripts but other than that it seems to work.

Backup snapshots are easy, but the recovery of the snapshot just stalled in an uninteruptable disk IO state.

Hmm, back to the drawing board...


Marty



-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl & Betty Schendel
Sent: 06 November 2008 12:36
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:

> Hi Karl et al,
>
> The deletion phase of the recovery can be controlled by the WDTD flag.
> So rather than do a delete of the data directory, I was simply
> planning to do:
> WDTD: /bin/echo At WDTD: Do nothing

That's useful. I assume you've tested this... :-) Once upon a time, the WDTD only affected added locations for a database, it would still unconditionally delete the files from the root location!
Which was pretty stupid, of course. I guess that got fixed. That last time I played with SAN checkpointing in anger was probably Ingres 2.5- ish.

> At the moment the problem is getting the SSH key exchange to work. It
> just aint. Thus all my SSH commands fail...which is a fairly limiting
> feature!

IMHO, the whole SSH thing is the modern day version of the modem.
It's an invention of the devil, meant to produce frustration... :-)

Karl

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

[Database Administrator]

Hi Andre,

This restore attempt is based on a rebuild using the snapshot. Keep it simple at this point and see if it all works.

I was also going to play with the thought of cloning the snapshot to a separate volume and then backing up that volume to tape. Which seems the best way of still keeping a tape backup of the database and minimising the impact on the real data. My thinking is that this will be OK as long as we can backup the database in a reasonable period. When writing to tape I get better than 1G/minute but that still means that 1Tb takes 16hours. I'm hoping that the tape device we need for this volume will be faster as well, but its still looking like a considerable portion of a day.

The final option is to look at SAN based replication to a remote host. I'd probably start considering that as the database backup and recovery just takes too long too achieve.

Which means more playing...sure beats working for a living.

Marty

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Laframboise, André
Sent: 06 November 2008 15:53
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


I forgot to mention we copy the snapshots to a separate set of physical disks
before they are backed up to tape.
A restore would involve straight data tranfers between real disks.

Is this how your set up works ? Or is your restore some type of rebuild based directly
on the snapshot ?

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Martin Bowes
Sent: November 6, 2008 10:34 AM
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN

Hi Karl et al,

Well I'd have to say that no I havent tested the WDTD recently...I suppose now is going to be a good opportunity!

The Dell guys were a little surprised at the entire SSH key exchange thing. But have provided a perl module which talks to the SAN using telnet and something vaguely akin to 'expect' scripts.
I have to hard code the password of the SAN manager into my scripts but other than that it seems to work.

Backup snapshots are easy, but the recovery of the snapshot just stalled in an uninteruptable disk IO state.

Hmm, back to the drawing board...


Marty



-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl & Betty Schendel
Sent: 06 November 2008 12:36
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


On Nov 6, 2008, at 5:37 AM, Martin Bowes wrote:

> Hi Karl et al,
>
> The deletion phase of the recovery can be controlled by the WDTD flag.
> So rather than do a delete of the data directory, I was simply
> planning to do:
> WDTD: /bin/echo At WDTD: Do nothing

That's useful. I assume you've tested this... :-) Once upon a time, the WDTD only affected added locations for a database, it would still unconditionally delete the files from the root location!
Which was pretty stupid, of course. I guess that got fixed. That last time I played with SAN checkpointing in anger was probably Ingres 2.5- ish.

> At the moment the problem is getting the SSH key exchange to work. It
> just aint. Thus all my SSH commands fail...which is a fairly limiting
> feature!

IMHO, the whole SSH thing is the modern day version of the modem.
It's an invention of the devil, meant to produce frustration... :-)

Karl

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres

[Database Administrator]

On Nov 7, 2008, at 5:27 AM, Martin Bowes wrote:

> Hi Andre,
>
> Yep, the root location gets wiped out in 2006R2 regardless of the
> settings of WDxD. It seems that option only applies to secondary
> data locations.
>
> Which sucks. It must have something to do with managing the
> configuration file...
>
>

The trick I recall using was to have a wrapper script move the root
location into a holding directory; let rollforwarddb wipe it out;
and then move the contents back as part of the restore action.

I really don't think it should be deleting the root location
unconditionally, though. Something to put on the ever growing
todo list, I guess.

Karl

[Database Administrator]

Hi Karl,

I've just made an extension to my recovery script. It now saves aside
the aaaaaaaa.rfc from the primary data location, lets the system wipe
out the data location and then after the restore completes at the EETE I
get it to restore the aaaaaaaa.rfc.

I'm hoping this will work, can't test it on the SAN as Dell just came
and took it away.

Marty

-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Karl
& Betty Schendel
Sent: 07 November 2008 12:35
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] Ingres and Dell-Equalogic SAN


On Nov 7, 2008, at 5:27 AM, Martin Bowes wrote:

> Hi Andre,
>
> Yep, the root location gets wiped out in 2006R2 regardless of the
> settings of WDxD. It seems that option only applies to secondary
> data locations.
>
> Which sucks. It must have something to do with managing the
> configuration file...
>
>

The trick I recall using was to have a wrapper script move the root
location into a holding directory; let rollforwarddb wipe it out;
and then move the contents back as part of the restore action.

I really don't think it should be deleting the root location
unconditionally, though. Something to put on the ever growing
todo list, I guess.

Karl

_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com...fo/info-ingres