Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Justin Derrick

Pages: [1] 2 3 4 5 6 ... 119
1
MP Server / Re: Tuning - Migration to Oracle
« on: January 08, 2021, 11:21:48 AM »
Presumably you made the move to Oracle because you have better internal support for it...  So...  the ultimate authority are the local DBAs.  :)

Otherwise, I don't think there's much that needs tuning outside making sure your database indexes are good.

-JD.

2
MP Server / Re: Maually read from Cache Storage possible
« on: January 04, 2021, 10:55:08 AM »
If you didn't migrate the data, then you have to keep the old server intact if you want access to the old data.  Trying to read the objects directly from the cache/TSM isn't supported.

The owner of the data will need to decide if it costs more to keep the old server, or just migrate the rest of the data.

-JD.

3
MP Server / Re: arsdb -rtv fails when creating db (10.5 + oracle)
« on: December 31, 2020, 01:54:37 PM »
Yeah, I wasn't sure if you had substituted these values for anonymized versions, or if you'd overlooked configuring them.

Code: [Select]
HOST=server.hostname
SRVR_INSTANCE=database

-JD.

4
z/OS Server / Re: CMOD Retention based on GDG like versions
« on: December 31, 2020, 01:51:57 PM »
It should be possible to manage this with CMOD's Enhanced Retention Management and some custom code...  Set a blanket hold on all documents, then write a script to search the database and remove holds that don't match the criteria.

I'm not a z/OS guy, so I don't think I can help more than that.

-JD.

5
z/OS Server / Re: Enhanced Retention Management and expiration
« on: December 16, 2020, 04:42:59 PM »
The thing to know about ERM is that it allows you to select documents to be put on hold (say, a regulatory issue, or lawsuit, or audit) or 'implied hold', which means 'hold everything I load until I say it can be deleted'.

To answer your first question...  In the first situation, yes.  Documents without 'holds' applied to them will expire as per your standard schedule.  Documents *with* holds will be held.  In the second situation (implied hold), you will have to remove the implied hold from documents in order to get them expired.

You're going to want to work with a specialist in CMOD on Z, because the implementation details are tricky, and going back to apply ERM on old documents will take some work - either low-level changes to database tables, or export and reload of data.




7
Hi MD.

There's nothing in CMOD's code that will help you validate the contents of a PDF *before* loading -- you'll have to do that yourself with some validation tool, and there are lots of open-source PDF libraries that will do a basic check.

You may want to switch your process to use the PDF Indexer, so that CMOD will try to open the PDF files and read them in order to identify and separate the resources.

-JD.

8
It's the only Windows graphical interface for administration.  There is the 'arsxml' utility, but that's more for batch administration, or moving configurations from system to system.

-JD.

9
z/OS Server / Re: Job to delete out files in the /tmp directory
« on: December 06, 2020, 02:27:24 PM »
If you have unix-style commands available to you, and the cron scheduling tool, you can configure cron to run this command daily, which deletes files that were created x days ago:

find /tmp/ -name FILESPEC -ctime +x rm -f {} \;

PLEASE BE EXTRAORDINARILY CAUTIOUS WITH THIS COMMAND, AS IT FORCES FILES TO BE DELETED. 

Set FILESPEC to the file specification you want to use ("*.AOD") and 'x' to the number of days you want files to exist.

-JD.

10
I'll risk incrimination here...  :)

You can use "SSH tunnels" to create a portal from the jumphost to the CMOD server.  Then you configure your windows client to connect to the jump host, and the jump host forwards the connection through the encrypted tunnel to the other side.

This may or may not violate your local security policy -- even though it obeys the spirit of the policy (all connections into production must go through the jump host), the people in charge of enforcing the policy may not see it that way.

The alternative that most organizations provide is a citrix server that has one foot inside the internal network, and the other foot inside the production network.

-JD.

11
MP Server / Re: Long Running ARSLOAD - How to see progress and kill
« on: December 03, 2020, 11:25:06 AM »
Hey TJ...

For long-running loads, you just have to be patient.  However, I've seen your other posts, and multi-hour loads are unheard of.  Even on-premise to on-cloud loads are measured in minutes.

-JD.

12
I order to avoid incriminating myself, I'll only say:  "SSH Tunnels".  :)

-JD.

13
MP Server / Re: Loading error and runstat error
« on: December 02, 2020, 08:04:21 AM »
You'll want to have your DB2 database administrator investigate the cause to make sure it isn't something serious.  In the past, I've solved this problem by taking CMOD & DB2 offline, running a DB2 full offline backup, then running a re-org on all of the ARS* tables and whichever Application Group tables that were causing problems.

Good luck, and please reply back with the solution when you find it!

-JD.

14
MP Server / Re: CMOD 9.5 MP Load Fails - No System Log Error Message?
« on: December 01, 2020, 04:39:21 PM »
Yup, that should work.  Try it out and report back.  :)

-JD.

15
OD/WEK & JAVA API / Re: Automatically unload when error occurs
« on: December 01, 2020, 04:35:07 PM »
So, just submit the LoadID to arsadmin unload?

-JD.

Pages: [1] 2 3 4 5 6 ... 119