Interesting change in ALTER EXTRACT command
Today, while helping a customer, we had to rebuild an extract. The integrated extract that we rebuilt was stuck in a loop and displaying it was an initial load extract from the HTML5 page (AdminService). The adminclient (cmd line), said it was working fine. After the customer rebooted the their GoldenGate Service (GGS) environment, the integrated extract was still having problems.
At this point, we executed an INFO EXTRACT <extract>, DETAIL and retrieved the sequence number (EXTSEQ) and the relative byte address (EXTRBA). With this information in hand, we removed the extract and the associated parameter file. After all, we were going to rebuilt the extract.
The customer the rebuilt the extract using Microsoft VSCode with the RESTful plug-in (makes it really easy and scriptable). With the integrated extract rebuilt, we attempted to ALTER EXTRACT from the admin service, but there is no options (image 1).
Image 1: alter extract no seqno/rba option
This lead us to look at altering the extract from AdminClient within GGS. The alter extract command that we use was:
adminclient> alter extract <extract>, extseq <num>, extrba <num>
This command caused a syntax error. For anyone and myself, doing GoldenGate for better part of 15 plus years, this was odd. So, we went and looked up the documentation on ALTER EXTRACT in 19c (here). The command syntax for ALTER EXTRACT for Admin Client is clearly the same as any veteran to Oracle GoldenGate would remember.
Admin Client Syntax (19c):
ALTER EXTRACT group-name
[, BEGIN (NOW | yyyy-mm-dd[ hh:mi:[ss[.cccccc]]]} |
EXTSEQNO sequence-number
[, EXTRBA archive-offset-number] [, ADD_EXTRACT_attribute] |
SCN value]
[, DESC
[, UPGRADE INTEGRATED TRANLOG]
[, DOWNGRADE INTEGRATED TRANLOG [THREADS number]]
[, THREAD number]
[, ETROLLOVER]
[, ENCRYPTIONPROFILE encryption-profile-name ]
[CRITICAL [ YES | NO ]
[PROFILE profile-name
| [AUTOSTART [ YES | NO ]
[DELAY delay-number]
[AUTORESTART [ YES | NO ]|
[RETRIES retries-number ]|
[WAITSECONDS wait-number ]|
[RESETSECONDS reset-number ]|
[DISABLEONFAILURE [ YES | NO ] ] ]
]
Then we realized or remembered that we were using GoldenGate Service (GGS) and there might have been a few things different from the 19c release to the 21c release. After all, Oracle GoldenGate Service (GGS) is running on 21c. This prompted me to look at the 21c docs and I was sure it didn’t change; users needed a way to position an extract after rebuilding. The documentation for 21c (here) provide what the syntax is for ALTER EXTRACT in 21c – there are minor differences:
Admin Client Syntax (21c):
ALTER EXTRACT group-name
[, BEGIN {NOW | yyyy-mm-dd[ hh:mi:[ss[.cccccc]]]} |
[, EXTRBA archive-offset-number] [, ADD_EXTRACT_attribute] |
SCN value]
[, DESC
[, THREAD number]
[, ETROLLOVER]
[, ENCRYPTIONPROFILE encryption-profile-name ]
[CRITICAL [ YES | NO ]
[PROFILE profile-name
| [AUTOSTART [ YES | NO ]
[DELAY delay-number]
[AUTORESTART [ YES | NO ]|
[RETRIES retries-number ]|
[WAITSECONDS wait-number ]|
[RESETSECONDS reset-number ]|
[DISABLEONFAILURE [ YES | NO ] ] ]
]
[, LOGNUM lognum]
[, LOGPOS logpos]
What this meant for the customer, is that the extract could be rebuilt; however, to find the correct position to start the extract from we needed to know how to get the correct System Change Number (SCN) or the correct Relative Byte Address (RBA).
In discussions with the customer, we decided that it was best to use a known System Change Number (SCN). With the information we had, the customer knew we could go back an hour. From here, we used an old post which I wrote in 2014 on how to convert a timestamp to SCN (here). After the retrieving the SCN, we start the rebuilt extract as follows:
adminclient> alter extract <extract>, scn <scn>
The extract started successfully, remained on the correct trail file (we were over 660ish files) and captured data as expected.
Lesson learned here was, between versions Oracle likes to change things and we need to keep up. At the same time, we wish subtle differences like this do not get over looked in the release notes.
Current Oracle Certs
Bobby Curtis
I’m Bobby Curtis and I’m just your normal average guy who has been working in the technology field for awhile (started when I was 18 with the US Army). The goal of this blog has changed a bit over the years. Initially, it was a general blog where I wrote thoughts down. Then it changed to focus on the Oracle Database, Oracle Enterprise Manager, and eventually Oracle GoldenGate.
If you want to follow me on a more timely manner, I can be followed on twitter at @dbasolved or on LinkedIn under “Bobby Curtis MBA”.
order priligy online Annals of Oncology, 27 suppl_6
buprenorphine and thioridazine both increase sedation medicamento priligy estudios clinicos 25 mg kg 1 for 30 days, while compounds 119 and 120 achieved the same effect at 10 mg kg 1 PO once a day
Compound I 100 was prepared from compound 82 donde comprar priligy mexico
It is a truth universally acknowledged, that a doctor in possession of a cellulitis patient, must be in want of an antibiotic recommendation from pharmacy how to buy priligy in usa reviews
pastillas priligy en mexico 2001, 113 772 778
2002, 162 2113 2123 priligy dapoxetina 30mg nos eua Lysates were analyzed by immunoblotting with anti Cre and anti ОІ actin antibodies and gels were run under the same experimental conditions
I think you used wrong syntax, the alter should work. You need to use EXTSEQNO instead of EXTSEQ
This web site is really a walk-through for all of the info you wanted about this and didn’t know who to ask. Glimpse here, and you’ll definitely discover it.
19 from Berlin four years ago, won the 100 meters last weekend having completed the sprint double twice at the Olympics and also in the 2009 world championships buy priligy tablets EMBASE was searched 1980 week 33, 2009, using the search terms tamoxifen and warfarin, for articles limited to use in humans and published in English, then further narrowed to drug interactions