Recent content by Rob Fitzpatrick

I once had an experience of trying to rebuild indexes of a large table on a resource-constrained machine. I didn't have enough RAM to prevent disk I/O when building all eight indexes at the same time. In the end it was faster to run idxbuild multiple times, with different sets of indexes, to...

It was also back-ported to late 10.2B (SP06 - 08?). I have confirmed that it is in SP08 HF 48.

Tuning explanation and advice for idxbuild: https://pugchallenge.org/downloads/2013/261_Index_Rebuild_Perf.pptx (audio recording also available) -r and -i will give the same performance benefit. -r allows crash recovery in some scenarios where -i does not. Re: threading, you can also look...

In my experience, binary load followed by idxbuild is faster and easier to optimize than using "build indexes" on the load. Be sure to back up the database between the load and the build, and then again after the build. -SG has a maximum value of 64. -r gets you the same boost as -i with less...

Conveying signals via colour alone is a potential pitfall if you need to worry about accessibility, i.e. you need to provide colour-blindness support.

No, but I think this thread belongs in the Development forum, rather than General Announcements | Chit Chat. You may get more engagement there.

Yes it does. Each index on a table contains as many index entries as there are records. Each index entry consists of the key value (the field value or set of fields values, for each of the fields in the index definition) and the rowid. Dumping records from a source database and loading...

Here are a related Community discussion and KB article: https://community.progress.com/s/question/0D7Pb00000ROx93KAD/detail https://community.progress.com/s/article/Fill-in-borders-are-displayed-in-white-color-on-Windows-11

When you restore from a backup, you are creating a new database. That database will have the same data storage areas, and assignment of objects to those areas, as the source database from which the backup was taken. As to where the extents of those areas are stored, you have two options: You...

Delete the target database and start over. You can't shrink extents. Regarding data recovery, it is good that you are thinking about that! I will give you my preferred approach. The database schema is tightly coupled with the application code. A particular version of your application has a...

Note that adding "-csoutput -verbose" provides the same output to stdout as without it. But it also creates a file called dbname.viewB2.txt in the current directory. This makes the data much easier to parse programmatically, or import into a spreadsheet or table.

I assume you mean a dump of data, sequence values, and schema. It is the schema that contains the mapping of schema objects (tables, indexes, and LOBs) into storage areas. If you loaded the same schema into your target database that you dumped from your source, then the area assignments...

Generally speaking, enabling AI should be something you do once, not on a regular basis. Stop the database, set the AI block size, enable AI, enable the AI archiver, set the appropriate parameters for the AIMD (-aiarcinterval and -aiarcdir), and then you're done. If there is power loss, etc...

I know that in the past there have been comments from Progress devs on the internal differences between _progres.exe and prowin.exe/prowin32.exe. The gist was that for batch mode, a character client is more efficient as it doesn't deal with window messages in its event loop. I am having...

It looks like prosoapview was discontinued after 11.7. https://community.progress.com/s/article/P120871 The article suggests to try Fiddler.