DB2 11 for z/OS - Question about cataloging archive log datasets

Javier Estrada Benavides

DB2 11 for z/OS - Question about cataloging archive log datasets

Hello everyone:

  I have a question about using the parm CATALOG which commands whether or not to catalog archive log datasets...

We're allocating them on VTS, and these days our z/OS sysprog and racf friends are asking us to avoid cataloging these datasets because of the number of entries in the ICF catalogs and the RACF calls to STGADMIN.IGG.DEFINE.RECAT on class FACILITY. I had not actually noticed that the manual is a bit misleading on this, so I'm here with a question:

 The manual says the recommendation for the CATALOG parameter when you allocate archive logs to tape is YES, however, right below that recommendation it states that doing so prevents the DB2 from reading directly from the BSDS and getting the specific volumen, thus provoking the DB2 to read the whole archive log data set. So which value was better for accessing the archive logs? Reading the whole archive log locating it from the ICF catalog or going directly to a specific volume by reading from the BSDS?. What value on this parameter do you guys use?

 

Thank you

 

Javier Estrada

DB2 for z/OS Certified System Administrator. Mexico. (BTW... I've been here for a while but I have not found any other Mexicans in here, am I really the only one??)

Robert Plata

DB2 11 for z/OS - Question about cataloging archive log datasets
(in response to Javier Estrada Benavides)
Orale! No, you’re not the only one….

We have CATALOG set to YES as well. I have been searching for the recommendation from IBM. Other, than saying ‘we just took the default’, I’m at a loss for a suggestion.

Bob Plata

From: Javier Estrada [mailto:[login to unmask email]
Sent: Tuesday, March 07, 2017 10:42 AM
To: [login to unmask email]
Subject: [DB2-L] - DB2 11 for z/OS - Question about cataloging archive log datasets


Hello everyone:

I have a question about using the parm CATALOG which commands whether or not to catalog archive log datasets...

We're allocating them on VTS, and these days our z/OS sysprog and racf friends are asking us to avoid cataloging these datasets because of the number of entries in the ICF catalogs and the RACF calls to STGADMIN.IGG.DEFINE.RECAT on class FACILITY. I had not actually noticed that the manual is a bit misleading on this, so I'm here with a question:

The manual says the recommendation for the CATALOG parameter when you allocate archive logs to tape is YES, however, right below that recommendation it states that doing so prevents the DB2 from reading directly from the BSDS and getting the specific volumen, thus provoking the DB2 to read the whole archive log data set. So which value was better for accessing the archive logs? Reading the whole archive log locating it from the ICF catalog or going directly to a specific volume by reading from the BSDS?. What value on this parameter do you guys use?



Thank you



Javier Estrada

DB2 for z/OS Certified System Administrator. Mexico. (BTW... I've been here for a while but I have not found any other Mexicans in here, am I really the only one??)

-----End Original Message-----

Phil Grainger

DB2 11 for z/OS - Question about cataloging archive log datasets
(in response to Robert Plata)
What are your disaster recovery plans?

I ask, because if you catalog archives and copies, then you likely can't even start DB2 recovery until the ICF catalogs are restored. And if these are not recovered successfully, you have problems

If you DON'T catalog them, then DB2 knows where the datasets should be without looking them up in ICF

Phil Grainger
BMC Software

Sent from my iPhone

On 7 Mar 2017, at 19:21, PLATA Robert <[login to unmask email]<mailto:[login to unmask email]>> wrote:

Orale! No, you're not the only one....

We have CATALOG set to YES as well. I have been searching for the recommendation from IBM. Other, than saying 'we just took the default', I'm at a loss for a suggestion.

Bob Plata

From: Javier Estrada [mailto:[login to unmask email]
Sent: Tuesday, March 07, 2017 10:42 AM
To: [login to unmask email]<mailto:[login to unmask email]>
Subject: [DB2-L] - DB2 11 for z/OS - Question about cataloging archive log datasets


Hello everyone:

I have a question about using the parm CATALOG which commands whether or not to catalog archive log datasets...

We're allocating them on VTS, and these days our z/OS sysprog and racf friends are asking us to avoid cataloging these datasets because of the number of entries in the ICF catalogs and the RACF calls to STGADMIN.IGG.DEFINE.RECAT on class FACILITY. I had not actually noticed that the manual is a bit misleading on this, so I'm here with a question:

The manual says the recommendation for the CATALOG parameter when you allocate archive logs to tape is YES, however, right below that recommendation it states that doing so prevents the DB2 from reading directly from the BSDS and getting the specific volumen, thus provoking the DB2 to read the whole archive log data set. So which value was better for accessing the archive logs? Reading the whole archive log locating it from the ICF catalog or going directly to a specific volume by reading from the BSDS?. What value on this parameter do you guys use?



Thank you



Javier Estrada

DB2 for z/OS Certified System Administrator. Mexico. (BTW... I've been here for a while but I have not found any other Mexicans in here, am I really the only one??)

-----End Original Message-----

-----End Original Message-----
BMC Software Limited Registered Office: Building E2, Eskdale Road, Winnersh, Wokingham, Berkshire, United Kingdom, RG41 5TS Registered in England No. 1927903 The content of this email is confidential. If you are not the addressee, you may not distribute, copy or disclose any part of it. If you receive this message in error, please delete this from your system and notify the sender immediately.

Venkat Srinivasan

RE: DB2 11 for z/OS - Question about cataloging archive log datasets
(in response to Javier Estrada Benavides)

Reading the whole log vs reading directly applies to the special case when the datasets is cut in multiple TAPE volumes. In this case reading it from catalog is likely a good choice against unexpected log errors by reading from BSDS inventory. If you look at the V11 manual that is what it is saying. I dont like to see multiple TAPE vols because it reduces the archive inventory.

I would use catalog yes due to the convenience. Once cataloged dataset can be relocated anywhere. Catalog does use cache and so does a RLISTED protection. The argument that there is a overhead in catalog access / RACF check is not maintainable considering the fact the extra cycles may be minuscule compared to the overhead from workload.

You may also get into unforeseen situations with uncataloged TAPE datasets as they can be mistakenly scratched. To enhance availability, you can even catalog each copies in their own icf catalogs. 

Venkat


In Reply to Javier Estrada:

Hello everyone:

  I have a question about using the parm CATALOG which commands whether or not to catalog archive log datasets...

We're allocating them on VTS, and these days our z/OS sysprog and racf friends are asking us to avoid cataloging these datasets because of the number of entries in the ICF catalogs and the RACF calls to STGADMIN.IGG.DEFINE.RECAT on class FACILITY. I had not actually noticed that the manual is a bit misleading on this, so I'm here with a question:

 The manual says the recommendation for the CATALOG parameter when you allocate archive logs to tape is YES, however, right below that recommendation it states that doing so prevents the DB2 from reading directly from the BSDS and getting the specific volumen, thus provoking the DB2 to read the whole archive log data set. So which value was better for accessing the archive logs? Reading the whole archive log locating it from the ICF catalog or going directly to a specific volume by reading from the BSDS?. What value on this parameter do you guys use?

 

Thank you

 

Javier Estrada

DB2 for z/OS Certified System Administrator. Mexico. (BTW... I've been here for a while but I have not found any other Mexicans in here, am I really the only one??)

Jim Tonchick

DB2 11 for z/OS - Question about cataloging archive log datasets
(in response to Javier Estrada Benavides)
IMHO it's always better to catalog archive logs than let them go uncataloged. Your tape management system can then easily uncataloged them when they expire. It also allows easy investigation by dataset name from ISPF 3.4 especially if you include the date and time in the name.
<span style="font-size: 16px;">
</span>
<span style="font-size: 16px;">Having the logs cataloged has never caused a performance problem with either the ICF USER cat or RACF in our shop.</span>
<span style="font-size: 16px;">
</span>-----Original Message-----
From: Javier Estrada <[login to unmask email]>
To: DB2-L <[login to unmask email]>
Sent: Tue, Mar 7, 2017 12:41 PM
Subject: [DB2-L] - DB2 11 for z/OS - Question about cataloging archive log datasets



<div id="AOLMsgPart_2_132460fb-4ce2-40f8-be24-834f1a18fdce">
<div class="aolReplacedBody"><p>Hello everyone:</p>
<p> I have a question about using the parm CATALOG which commands whether or not to catalog archive log datasets...</p>
<p>We're allocating them on VTS, and these days our z/OS sysprog and racf friends are asking us to avoid cataloging these datasets because of the number of entries in the ICF catalogs and the RACF calls to STGADMIN.IGG.DEFINE.RECAT on class FACILITY. I had not actually noticed that the manual is a bit misleading on this, so I'm here with a question:</p>
<p> The manual says the recommendation for the CATALOG parameter when you allocate archive logs to tape is YES, however, right below that recommendation it states that doing so prevents the DB2 from reading directly from the BSDS and getting the specific volumen, thus provoking the DB2 to read the whole archive log data set. So which value was better for accessing the archive logs? Reading the whole archive log locating it from the ICF catalog or going directly to a specific volume by reading from the BSDS?. What value on this parameter do you guys use?</p>
<p> </p>
<p>Thank you</p>
<p> </p>
<p>Javier Estrada</p>
<p>DB2 for z/OS Certified System Administrator. Mexico. (BTW... I've been here for a while but I have not found any other Mexicans in here, am I really the only one??)</p>
<hr size="1" style="color:#ccc"><div id="aolmail_socfooter" style="font-size:80%"><span style="font-weight:bold">Site Links: </span>
<a target="_blank" rel="noopener noreferrer" href="http://www.idug.org/p/fo/st/?post=180385&anc=p180385#p180385">View post online</a>
<a target="_blank" rel="noopener noreferrer" href="http://www.idug.org/p/fo/si/?topic=19">View mailing list online</a>
<a target="_blank" rel="noopener noreferrer" href="mailto:[login to unmask email]">Start new thread via email</a>
<a target="_blank" rel="noopener noreferrer" href="mailto:[login to unmask email]?Subject=Unsubscribe">Unsubscribe from this mailing list</a>
<a target="_blank" rel="noopener noreferrer" href="http://www.idug.org/p/us/to/">Manage your subscription</a>


This email has been sent to: <a href="mailto:[login to unmask email]">[login to unmask email]</a>
<p>
** ** ** Attend the 2017 IDUG Tech Conference North America ** ** **

---> Anaheim, California, April 30 - May 04, 2017 <---

<a target="_blank" rel="noopener noreferrer" href="http://www.idug.org/na">http://www.idug.org/na </a>

</p>
<p>
Use of this email content is governed by the terms of service at:
<a target="_blank" rel="noopener noreferrer" href="http://www.idug.org/p/cm/ld/fid=2">http://www.idug.org/p/cm/ld/fid=2</a></p>

<hr size="1" style="color:#ccc"></div>
</div>
</div>

Javier Estrada Benavides

RE: DB2 11 for z/OS - Question about cataloging archive log datasets
(in response to Jim Tonchick)

Hey

Thank you very much for your replies, I was able to have a better talk with the sysprog and racf friends in here (well.... in this site I also change hats to be the one of the racf guys myself).

 

Javier Estrada

DB2 11 for z/OS Certified System Administrator. Mexico