Dan Langille @dvl@mastodon.social

Looking for someone to help the Warsaw Expo Refugee Center. They need someone with Django and React experience. They have an existing app which needs modification & data loaded in from a spreadsheet.

If interested, please contact me directly & I will put you in touch. Please boost.

Today I realized I don't need to manually do a pkg upgrade after a upgrade between FreeBSD versions.

https://dan.langille.org/2022/03/27/making-my-pkg-conf-configuration-version-independent/

Yeah, I know I don't HAVE to do such an upgrade, because things will still run, but that's now how things are done around here.

Last weekend, I found rotten wood in the workbench. Today I removed it, found lots more rotten wood, and some dead animals.

https://www.flickr.com/photos/dlangille/albums/72177720296666241

Today: update the Bacula regression testing scripts. Sometime between Bacula 9 and Bacula 11, Linuxisms have crept in. I've been given patches which I'll include in what I have at home.

Fixes will be sent upstream.

email: https://sourceforge.net/p/bacula/mailman/bacula-devel/thread/4941c285-4b46-db1c-bad1-aa8becbb5970%40langille.org/#msg37360380

tests: https://regress.bacula.org/index.php?project=Bacula-11.0

Today: update the Bacula regression testing scripts. Sometime between Bacula 9 and Bacula 11, Linuxisms have crept in. I've been given patches which I'll include in what I have at home.

Fixes will be sent upstream.

email: https://sourceforge.net/p/bacula/mailman/bacula-devel/thread/4941c285-4b46-db1c-bad1-aa8becbb5970%40langille.org/#msg37360380

tests: https://regress.bacula.org/index.php?project=Bacula-11.0

I think this is a duplicate function issue. The same function is declared twice.

One of the functions works, the other does not.

This is not a bug with PostgreSQL.

It's a bug in my file of stored procedures.

Guess what?

Works on PostgreSQL 14.

I'm going to do another test on PostgreSQL 13, using the same approach as for PostgreSQL 13 - I wonder if I've overlooked something in the process.

CONTEXT: SQL statement "INSERT INTO element_pathname (element_id, pathname) VALUES (NEW.id, l_pathname)"
PL/pgSQL function element_pathname_insert() line 7 at SQL statement
SQL statement "insert into element(id, parent_id, directory_file_flag, name, status)
values(element_id, element_parent_id, 'D', element_name, 'A')"
PL/pgSQL function element_add(text,character) line 89 at SQL statement
freshports.org=#

With source db PostgreSQL 12 and all other tools being PostgreSQL 13, same source database:

pgdump (13) pg_restore (13), the PostgreSQL 13 db gives:

freshports.org=# select element_add('/src/things/dan', 'F');
NOTICE: for NEW.id = "1181875", I am inserting "" into element_pathname
ERROR: duplicate key value violates unique constraint "element_pathname_pathname"
DETAIL: Key (pathname)=() already exists.

With all PostgreSQL 12 tools:

pg_dump and pg_restore:

freshports.org=# select element_add('/src/things/dan', 'F');
element_add
-------------
1181966
(1 row)

freshports.org=#

HEADS UP: even using the correct procedure, this does not work on a PG13 database.

Next test: load this db up on PostgreSQL 12 and verify that restore performs as expected.

For my next trick I'll try restoring the PostgreSQL 12 dump again and see if I can replicate the problem a second time. Tune in next time.

The anomaly: a trigger upon insert on table A will insert a row into table B - there is a one-to-one relationship between the two tables. In the newly-restored PostgreSQL 13 database, that insert always resulted in a duplicate key violation in table B.

Dumping with a PostgreSQL 13 client and loading into a fresh PostgreSQL 13 instance does not replicate the original problem.

I encountered a PostgreSQL trigger anomaly which seems to be related to my incorrect dumping of the database.

The issue: I did a pg_dump with a PostgreSQL 12 client and loaded into a PostgreSQL 13 database.

The correct procedure is to dump with a PostgreSQL 13 client. I failed.

The first steps to avoid this warning:

poudriere: Warning: Using ‘-‘ in a SETNAME is not recommended as it causes ambiguities with parsing the build name of 122amd64-default-master-list

https://dan.langille.org/2021/08/28/poudriere-warning-using-in-a-setname-is-not-recommended-as-it-causes-ambiguities-with-parsing-the-build-name-of-122amd64-default-master-list/

#FreeBSD

Today I've been using 'mkjail update -a' to update all my jails with respect to recent published known vulns on #FreeBSD.

Why update the jail and the host? Because I can.

Patch everything when a patch is available. One vuln might have unknown leverage. I can't tell. So I patch.

How I took my name servers offline by adding a new hostname

https://dan.langille.org/2021/08/13/how-i-took-my-name-servers-offline-by-adding-a-new-hostname/

Is this a new device or not?

gpart shows a partition.

smarctl says 4 power on hours, 56 power cycles, 104GB written.

Could that just be initial testing at the factory?

https://dan.langille.org/2021/08/09/i-bought-a-new-nvme-drive-or-did-i/