The Demon-Haunted World, Carl Sagan, 1996
21.07.2025 09:20 β π 6 π 0 π¬ 0 π 0@chris.antognini.ch
Striving for Optimal Performance https://antognini.ch
The Demon-Haunted World, Carl Sagan, 1996
21.07.2025 09:20 β π 6 π 0 π¬ 0 π 0For those who can't past the paywall, you can also find the list of books here:
fiendatcourt.com/six-books-ab...
The AI age looks likely to belong not to enablers of technology like Accenture but to its originators
28.06.2025 15:32 β π 8 π 3 π¬ 1 π 0Today is a special day for me: it is the last working day of 2025. I will do my best to spend the rest of the year doing whatever makes me and my dearest ones happyπ
26.06.2025 06:46 β π 19 π 0 π¬ 4 π 0The issue was due to the COLUMN TRANSFORM clause. According to my tests:
- COLUMN TRANSFORM is needed when NO offset/length is specified.
- COLUMN TRANSFORM is NOT needed when offset/length is specified.
Even though I don't understand why such an inconsistency exists, now I know how to implement it.
The non-working example provided in the documentation can be found in "Oracle Database Utilities Guide - 18.7.2.2 Loading LOBs from LOBFILE Files".
docs.oracle.com/en/database/...
Excerpt of "Oracle Database Utilities Guide" describing how data is processed when the LLS clause is used.
I'm looking for a working example of the LLS Clause in #OracleDatabase, but I'm unable to find it. Also the one provided in the doc is wrong (doesn't work).
When no offset/length is specified, it works as expected.
With offset/length, I always get a KUP-04001 error referencing an invalid file name.
In other #databs news, Salesforce is buying Informatica. I guess Salesforce is where aging data tools go to die.
27.05.2025 13:52 β π 11 π 1 π¬ 1 π 1They did it just right, and it deserved to be done just right.
My latest from #RolandGarros for #Tennis & Beyond
#Nadal #clay #farewell
open.substack.com/pub/christop...
To mark publication of my new book THE WARRIOR. I took a stroll with @benrothenberg.bsky.social. We visited landmarks I focus on in the book, which goes deep on #Nadal's career & #RolandGarros.
This unique post is a guided tour & audio tour. Thanks Ben!
christopherclarey.substack.com/cp/164293213
I also don't use mine that often. But, I'm not ready to get rid of it. Let's see how long it will take...
21.05.2025 09:55 β π 0 π 0 π¬ 0 π 0Now that the conference is over, let's spend two days at the Kennedy Space Center!
16.05.2025 11:02 β π 1 π 0 π¬ 0 π 0#QlikConnect appreciation event took place in Disney's Animal Kingdom. I'm not a big fan of such parks, but it was fun. Avatar Flight is really impressive! Thanks to #Qlik and diamond sponsors #AWS and #Accenture.
16.05.2025 10:57 β π 0 π 0 π¬ 0 π 0First time I attended a conference using the silent-disco approach. E.g. on the photo, there are three "rooms" with parallel sessions. I must say it works really well.
#QlikConnect
Crypto once dreamed of being above politics. But now the industry has developed a grubby relationship with the American governmentβand become synonymous with self-dealing econ.st/43iWAOC
15.05.2025 13:57 β π 215 π 64 π¬ 12 π 7#QlikConnect day 2 - general session with guest speaker Katie Ledecky
15.05.2025 14:02 β π 2 π 0 π¬ 0 π 0#QlikConnect day 1 - general session
14.05.2025 12:35 β π 2 π 0 π¬ 0 π 0On my way to #QlikConnect25 in #Orlando. I can't believe the last time I took a plane for a conference was in December 2019 for #UKOUG in Brighton.
12.05.2025 10:33 β π 5 π 0 π¬ 2 π 0Hey #Oracle, can you please remove from the 23 documentation references to Java classes (e.g. oracle.sql.STRUCT/OPAQUE/ARRAY/REF) that were deprecated more than one decade ago?
docs.oracle.com/en/database/...
If it's related to wind, I guess that there are still plenty of electrical overhead lines. In CH, except for trains and HT, they are getting very rare...
23.04.2025 18:17 β π 1 π 0 π¬ 1 π 0Out of curiosity, what's the frequency of power cuts?
23.04.2025 17:42 β π 0 π 0 π¬ 1 π 0According to Oracle Support it's bug 36628068 - HIGH CONTENTION ON SHARED POOL LATCH WHEN ALLOCATING FSO SCN FIELDS. Waiting on a backport to verify how good the patch works.
22.04.2025 09:53 β π 1 π 0 π¬ 0 π 0Qlik Connect 2025 - Highly automated and customizable data migration to Qlik Talend
Somehow, I find it strange to see my name on a conference agenda... I can't even remember the last time it happened (except for #tvdpdays #rip). Anyway, I'm happy to talk about what we developed in the last 5 years.
#ecaas #qlik #qlikconnect25
SR created.
02.04.2025 09:06 β π 1 π 0 π¬ 0 π 0Today I reproduce the issue with the latest RUs: 19.26 and 21.17.
I'll discuss about the SR with my customer.
The fix for bug 35999726 (available in 19.23) doesn't help.
27.03.2025 13:44 β π 2 π 0 π¬ 0 π 0On the customer's database, I have limited access. Hence, I only took a systemstate (level 266).
On my notebook, I used perf to confirm that what I see seems equivalent to the customer's database.
It seems that the issue is caused by kdlwiOpen1/kdlwiClose1.
Next, I will try the fix for bug 35999726.
The customer is using 19.18. On my notebook I have 19.19.
27.03.2025 08:08 β π 2 π 0 π¬ 1 π 0To be clear, the contention is due to the 8 parallel processes themselves; not to other concurrent operations. The following images illustrate what I observe on my notebook when moving a small 1-million-row table...
26.03.2025 15:58 β π 0 π 0 π¬ 2 π 0I would love to know why in #OracleDatabase only statement #2 is subject to shared-pool latch contention. If you have a clue, thank you for sharing.
#1: ALTER TABLE t MOVE LOB (l) STORE AS SECUREFILE (NOCOMPRESS) PARALLEL 8
#2: ALTER TABLE t MOVE LOB (l) STORE AS SECUREFILE (COMPRESS) PARALLEL 8