https://media.notthebee.com/articles/657b2e2993ba1657b2e2993ba2.jpg

I love it.

Not the Bee

https://cdn.arstechnica.net/wp-content/uploads/2023/12/enemycloset-760×380.jpg

While Doom can sometimes feel like an overnight smash success, the seminal first-person shooter was far from the first game created by id co-founders John Carmack and John Romero. Now, in a rare joint interview that was livestreamed during last weekend’s 30th-anniversary celebration, the pair waxed philosophical about how Doom struck a perfect balance between technology and simplicity that they hadn’t been able to capture previously and have struggled to recapture since.

Carmack said that Doom-precursor Wolfenstein 3D, for instance, "was done under these extreme, extraordinary design constraints" because of the technology available at the time. "There just wasn’t that much we could do."

One of the biggest constraints in Wolfenstein 3D was a grid-based mapping system that forced walls to be at 90-degree angles, leading to a lot of large, rectangular rooms connected by long corridors. "Making the levels for the original Wolfenstein had to be the most boring level design job ever because it was so simple," Romero said. "Even [2D platformer Commander Keen] was more rewarding to make levels for."

By the time work started on Doom, Carmack said it was obvious that "the next step in graphics was going to be to get away from block levels." The simple addition of angled walls let Doom hit "this really sweet spot," Carmack said, allowing designers to "create an unlimited number of things" while still making sure that "everybody could draw in this 2D view… Lots of people could make levels in that."

Romero expanded on the idea, saying that working on top of the Doom engine was, at the time, "the easiest way to make something that looks great. If you want to get anything that looks better than this, you’re talking 10 times the work."

Was Quake too complex?

Then came Quake, with a full 3D design that Carmack admitted was "more ambitious" and "did not reach all of its goals." When it came to modding and designing new levels, Carmack lamented how, with Quake, "a lot of potentially great game designers just hit their limit as far as compositional aesthetic in terms of what something is going to look like."

"When you have the ability to do a full six degree-of-freedom modeling, you not only have to be a game designer, you have to be an architect, a modeler working through your composition," Carmack continued. "[Doom] helped you along by keeping you from wasting time doing some crazy things that you would have had to be a master of a different craft to pull off… Going to full 3D made this something that not everybody does on a lark, but something you set out time, almost set a career arc to make mods for newer games."

Romero reminisced during the chat about "the Everest of Quake" and "the insane amount of technology we took on" during its development. "Even adding QuakeC on top of [a] client/server [architecture] on top of full 3D, it was so much tech. It was a whole new engine, it wasn’t Doom at all. It was all brand new."

Looking back, Carmack allowed that "there’s a couple different steps we could have taken [with Quake], and we probably did not pick the optimal direction, but we kept wanting to make this, just throw everything at it and say, ‘If you can think of something that’s going to be better, we should just strive our hardest to do that.’ When Doom came together, it was just a perfect storm of ‘everything went right.’.. [it was] as close to a perfect game as anything we made."

Ars Technica – All content

https://opengraph.githubassets.com/1f9367f276bb74707b68c5b387ff964b9e84a2b12ad55594b2a533de7f130699/maantje/pulse-database

composer require maantje/pulse-database

return [
    // ...
    
    'recorders' => [
        \Maantje\Pulse\Database\Recorders\DatabaseRecorder::class => [
            'connections' => [
                'mysql_another' => [
                    'values' => [
                        'Connections',
                        'Threads_connected',
                        'Threads_running',
                        'Innodb_buffer_pool_reads',
                        'Innodb_buffer_pool_read_requests',
                        'Innodb_buffer_pool_pages_total',
                        'Max_used_connections'
                    ],
                    'aggregates' => [
                        'avg' => [
                            'Threads_connected',
                            'Threads_running',
                            'Innodb_buffer_pool_reads',
                            'Innodb_buffer_pool_read_requests',
                            'Innodb_buffer_pool_pages_total',
                        ],
                        'max' => [
                            //
                        ],
                        'count' => [
                            //
                        ],
                    ],
                ],
                'mysql' => [
                    'values' => [
                        'Connections',
                        'Threads_connected',
                        'Threads_running',
                        'Innodb_buffer_pool_reads',
                        'Innodb_buffer_pool_read_requests',
                        'Innodb_buffer_pool_pages_total',
                        'Max_used_connections'
                    ],
                    'aggregates' => [
                        'avg' => [
                            'Threads_connected',
                            'Threads_running',
                            'Innodb_buffer_pool_reads',
                            'Innodb_buffer_pool_read_requests',
                            'Innodb_buffer_pool_pages_total',
                        ],
                        'max' => [
                            //
                        ],
                        'count' => [
                            //
                        ],
                    ],
                ]
            ]
        ],
    ]
]

<x-pulse>
    <livewire:pulse.servers cols="full" />
    
+ <livewire:database cols='6' title="Active threads" :values="['Threads_connected', 'Threads_running']" :graphs="[
+ 'avg' => ['Threads_connected' => '#ffffff', 'Threads_running' => '#3c5dff'],
+ ]" />

+ <livewire:database cols='6' title="Connections" :values="['Connections', 'Max_used_connections']" />

+ <livewire:database cols='full' title="Innodb" :values="['Innodb_buffer_pool_reads', 'Innodb_buffer_pool_read_requests', 'Innodb_buffer_pool_pages_total']" :graphs="[
+ 'avg' => ['Innodb_buffer_pool_reads' => '#ffffff', 'Innodb_buffer_pool_read_requests' => '#3c5dff'],
+ ]" />

    <livewire:pulse.usage cols="4" rows="2" />

    <livewire:pulse.queues cols="4" />

    <livewire:pulse.cache cols="4" />

    <livewire:pulse.slow-queries cols="8" />

    <livewire:pulse.exceptions cols="6" />

    <livewire:pulse.slow-requests cols="6" />

    <livewire:pulse.slow-jobs cols="6" />

    <livewire:pulse.slow-outgoing-requests cols="6" />

</x-pulse>

Laravel News Links

https://media.notthebee.com/articles/6579bf8b4fa716579bf8b4fa72.jpg

Do you need further proof of man’s total depravity and his need for a Savior?

Not the Bee

https://media.notthebee.com/articles/657883b4a4a3b657883b4a4a3c.jpg

Christmas carols have officially peaked, you guys. Watch this video and tell me otherwise.

Not the Bee

https://s.yimg.com/os/creatr-uploaded-images/2023-12/7fc6a590-9873-11ee-bb6b-c5c58a96588d

Apple’s AI-powered Journal app is finally here. The new diary entry writing tool was first teased for iOS 17 back in June, but it only became available on Monday with the new iPhone update — nearly three months after iOS 17 itself came out. After Apple released iOS 17.2, iPhone users can now access to the Journal app, which allows users to jot down their thoughts in a digital diary. Journaling is a practice that can improve mental wellbeing and it can also be used to fuel creative projects.

You can create traditional text entries, add voice recordings to your notes, or include recent videos or pictures. If you need inspiration, AI-derived text prompts can offer suggestions for what to write or create an entry for next. The app also predicts and proposes times for you to create a new entry based on your recent iPhone activity, which can include newer photos and videos, location history, recently listened-to playlists, and workout habits. This guide will walk you through how to get started with the Journal app and personalize your experience.

How to create a new entry in the Journal app on iPhone

Malak Saleh

When you open the Journal app, tap the + button at the bottom of the page to create a new entry. If you want to start with a blank slate, when you tap ‘New Entry’ an empty page will appear and from there you can start typing text. You can add in recent photos from your library when you tap the photos icon below the text space, take a photo in the moment and add it to your entry or include a recorded voice memo when you tap the voice icon. You can also add locations to your entry when you tap the arrow icon at the bottom right of an entry page. This feature might be helpful for travel bloggers looking back at their trips abroad. You can edit the date of an entry at the top of the page.

Alternatively, you can create a post based on recent or recommended activities that your phone compiled — say, pictures, locations from events you attended, or contacts you recently interacted with. The recent tab will show you, in chronological order, people, photos and addresses that can inspire entries based on recent activities. The recommended tab pulls from highlighted images automatically selected from your photo memories. For example, a selection of portraits from 2022 can appear as a recommendation to inspire your next written entry. Some suggestions underneath the recommendation tab may appear within the app with ‘Writing prompts.’ For example, a block of text may appear with a question like, “What was the highlight of your trip?”

Malak Saleh

Scheduling, bookmarking and filtering

If you’re not free to write when a suggestion is made, you can also save specific moments you want to journal about and write at a later time. Using the journaling schedule feature, you can set a specific time to be notified to create an entry, which will help a user make journaling a consistent practice. Go to the Settings app on your iPhone and search for the Journal app. Turn on the ‘Journaling schedule’ feature and personalize the days and times you would like to be reminded to write entries. As a side note, in Settings, you can also opt to lock your journal using your device passcode or Face ID.

Malak Saleh

You can also organize your entries within the app using the bookmarking feature, so you can filter and find them at your own convenience. After creating an entry, tap the three dots at the bottom of your page and scroll down to tap the bookmark tab. This is the same place where you can delete or edit a journal entry.

Later on, if you want to revisit a bookmarked entry, tap the three-line icon at the corner of the main journal page to select the filter you would like applied to your entries. You can select to only view bookmarked entries, entries with photos, entries with recorded audio and see entries with places or locations. This might be helpful when your journal starts to fill up with recordings.

Adding music, workouts and other off-platform entries into your journal app

Using your streaming app of choice, (Apple Music, Spotify or Amazon Music), you can integrate specific tracks or podcast episodes into your entries by tapping three buttons at the bottom of your screen that opens up the option to ‘share your music.’ The option to share a track to the Journal app should appear and it will sit at the top of a blank entry when you open the app.

You can use the same method with other applications, like Apple’s Fitness app. You can share and export a logged workout into your journal and start writing about that experience.

Malak Saleh

This article originally appeared on Engadget at https://www.engadget.com/how-to-use-apples-new-journal-app-with-the-ios-172-update-164518403.html?src=rssEngadget

https://photos5.appleinsider.com/gallery/57678-117473-qnapnasbook-xl.jpg

Article Hero Image

QNAP has introduced its Thunderbolt 4 All-Flash NASbook, a high-density storage device aimed at video production that can provide high-speed transfers to connected Macs.

Firmly aimed at streamlining video production in pre-production and post-production, the QNAP Thunderbolt 4 All-Flash NASbook is an M.2 SSD device meant to hold fast storage drives, as well as delivering the data to users as quickly as possible.

Measuring 8.46 inches by 7.83 inches by 2.36 inches, the NASbook is a very small storage appliance. At 4.94 pounds, it’s also reasonably portable for a dedicated storage device, meaning it could be brought to filming locations instead of being kept in an office.

Capable of holding five E1S/M.2 PCIe NVMe SSDs, the unit offers automatic RAID disk replacement, so SSDs can be replaced without any system downtime. It also uses 13th-gen Intel Core hybrid architecture 8-core, 12-thread or 12-core, 16-thread processors for dealing with video production tasks, while the built-in GPU can help accelerate video transcoding.

On the connectivity side, it has two Thunderbolt 4 ports to connect to Mac and PC workstations, while 2.5GbE and 10GbE ports can be used for more conventional network-based NAS usage. A pair of USB 3.2 Gen 2 10Gbps ports can deal with ingesting video from other storage, or for connecting a DAS or JBOD for archiving projects.

The included HDMI port allows teams to display raw footage stored on the NASbook on an external display at up to 4K resolution.

The entire unit runs on the ZFS-based QuTS hero operating system, which has self-healing properties to detect and repair corrupted data in RAW images.

The QNAP Thunderbol 4 All-Flash NASbook starts from $1,199 for the 8-core version, $1,449 for the 12-core model, and are available now.

As part of a launch offer, a free one-year 1TB allocation of myQNAPcloud cloud storage is being provided to early buyers from QNAP’s website.

AppleInsider News

https://i.kinja-img.com/image/upload/c_fill,h_675,pg_1,q_80,w_1200/6a12586b70dc3414af93d6de525cd139.png

Gizmodo

https://www.percona.com/blog/wp-content/uploads/2023/12/pt-table-sync-for-Replica-Tables-With-Triggers-200×119.jpg

In Percona Managed Services, we manage Percona for MySQL, Community MySQL, and MariaDB. Sometimes, the replica server might have replication errors, and the replica might be out of sync with the primary. In this case, we can use Percona Toolkit’s pt-table-checksum and pt-table-sync to check the data drift between primary and replica servers and make the replica in sync with the primary. This blog gives you some ideas on using pt-table-sync for replica tables with triggers.

In my lab, we have two test nodes with replication setup, and both servers will have Debian 11 and Percona Server for MySQL 8.0.33 (with Percona Toolkit) installed.

The PRIMARY server is deb11m8 (IP: 192.168.56.188 ), and the REPLICA server name is deb11m8s (IP: 192.168.56.189).

1. Creating the test tables and the AFTER INSERT trigger

Create the below table and trigger on PRIMARY, and it will replicate down to REPLICA. We have two tables: test_tab and test_tab_log. When a new row is inserted into test_tab, the trigger will fire and put the data and the user who did the insert into the test_tab_log table.

Create database testdb;
Use testdb; 
Create table test_tab (id bigint NOT NULL , test_data varchar(50)  NOT NULL ,op_time TIMESTAMP  NOT NULL , PRIMARY KEY (id,op_time));              
Create table test_tab_log (id bigint  NOT NULL , test_data varchar(50)  NOT NULL ,op_user varchar(60)  NOT NULL  ,op_time TIMESTAMP  NOT NULL , PRIMARY KEY (id,op_time)); 
DELIMITER $$
CREATE DEFINER=`larry`@`%` TRIGGER after_test_tab_insert  AFTER INSERT
ON test_tab FOR EACH ROW
BEGIN
   INSERT INTO test_tab_log(id,test_data,op_user,op_time) VALUES(new.id, NEW.test_data, USER(),NOW());
END$$
DELIMITER ;

2. Let’s fill in some test data

We do an insert as a root user. You can see that after data is inserted, the trigger fires as expected.

mysql> insert into test_tab (id,test_data,op_time) values(1,'lt1',now());
Query OK, 1 row affected (0.01 sec)
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
+----+-----------+---------------------+
1 row in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
+----+-----------+----------------+---------------------+
1 row in set (0.00 sec)
We Insert another row insert into test_tab (id,test_data,op_time) values(2,'lt2',now());
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
2 rows in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+
2 rows in set (0.00 sec)

3. Let’s get percona.dsns ready for pt-table-checksum and pt-table-sync

CREATE TABLE percona.dsns (`id` int(11) NOT NULL AUTO_INCREMENT,`parent_id` int(11) DEFAULT NULL, `dsn` varchar(255) NOT NULL, PRIMARY KEY (`id`)) ENGINE=InnoDB DEFAULT CHARSET=utf8;
INSERT INTO percona.dsns (dsn) VALUES ('h=192.168.56.190');

4. Simulate the out of sync on 192.168.56.190 by removing one row (id=1) in test_tab

mysql> use testdb;
Database changed
mysql> set sql_log_bin=0;
Query OK, 0 rows affected (0.00 sec)
mysql> delete from test_tab where id=1;
Query OK, 1 row affected (0.00 sec)
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
1 row in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+
2 rows in set (0.00 sec)

5. Run pt-table-checksum to report the difference

root@deb11m8:~/test_pt_trigger# pt-table-checksum h=192.168.56.189 --port=3306 --no-check-binlog-format 
--no-check-replication-filters --replicate percona.checksums_test_tab  
--recursion-method=dsn=D=percona,t=dsns 
--tables testdb.test_tab 
--max-load Threads_running=50 
--max-lag=10 --pause-file /tmp/checksums_test_tab
Checking if all tables can be checksummed ...
Starting checksum ...
            TS ERRORS  DIFFS     ROWS  DIFF_ROWS  CHUNKS SKIPPED    TIME TABLE
11-26T10:02:58      0      1        2          1       1       0   4.148 testdb.test_tab
root@deb11m8:~/test_pt_trigger#

on REPLICA, deb11m8s, we can see the checksum reports the difference.

mysql> SELECT db, tbl, SUM(this_cnt) AS total_rows, COUNT(*) AS chunks
    -> FROM percona.checksums_test_tab
    -> WHERE (
    -> master_cnt <> this_cnt
    -> OR master_crc <> this_crc
    -> OR ISNULL(master_crc) <> ISNULL(this_crc))
    -> GROUP BY db, tbl;
+--------+----------+------------+--------+
| db     | tbl      | total_rows | chunks |
+--------+----------+------------+--------+
| testdb | test_tab |          1 |      1 |
+--------+----------+------------+--------+
1 row in set (0.00 sec)

6. Let’s try pt-table-sync to fix it; we will run pt-table-sync under user ‘larry’@’%’

Pt-table-sync says Triggers are defined on the table and will not continue to fix it.

root@deb11m8:~/test_pt_trigger# pt-table-sync h=192.168.56.190,P=3306 --sync-to-master 
--replicate percona.checksums_test_tab 
--tables=testdb.test_tab 
--verbose --print
# Syncing via replication P=3306,h=192.168.56.190
# DELETE REPLACE INSERT UPDATE ALGORITHM START    END      EXIT DATABASE.TABLE
Triggers are defined on the table at /usr/bin/pt-table-sync line 11306. while doing testdb.test_tab on 192.168.56.190
#      0       0      0      0 0         10:03:31 10:03:31 1    testdb.test_tab

Pt-table-sync has an option –[no]check-triggers- to that will skip trigger checking. The print result is good.

root@deb11m8:~/test_pt_trigger# pt-table-sync --user=larry --ask-pass  h=192.168.56.190,P=3306 --sync-to-master --nocheck-triggers 
--replicate percona.checksums_test_tab 
--tables=testdb.test_tab 
--verbose --print
# Syncing via replication P=3306,h=192.168.56.190
# DELETE REPLACE INSERT UPDATE ALGORITHM START    END      EXIT DATABASE.TABLE
REPLACE INTO `testdb`.`test_tab`(`id`, `test_data`, `op_time`) VALUES ('1', 'lt1', '2023-11-26 09:59:19') /*percona-toolkit src_db:testdb src_tbl:test_tab src_dsn:P=3306,h=192.168.56.189 dst_db:testdb dst_tbl:test_tab dst_dsn:P=3306,h=192.168.56.190 lock:1 transaction:1 changing_src:percona.checksums_test_tab replicate:percona.checksums_test_tab bidirectional:0 pid:4169 user:root host:deb11m8*/;
#      0       1      0      0 Nibble    10:03:54 10:03:55 2    testdb.test_tab

When we run pt-table-sync with –execute under user ‘larry’@’%’:

root@deb11m8:~/test_pt_trigger# pt-table-sync --user=larry --ask-pass  h=192.168.56.190,P=3306 --sync-to-master --nocheck-triggers 
--replicate percona.checksums_test_tab 
--tables=testdb.test_tab 
--verbose --execute
# Syncing via replication P=3306,h=192.168.56.190
# DELETE REPLACE INSERT UPDATE ALGORITHM START    END      EXIT DATABASE.TABLE
#      0       1      0      0 Nibble    10:05:26 10:05:26 2    testdb.test_tab
-------PRIMARY -------
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
2 rows in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  1 | lt1       | larry@deb11m8  | 2023-11-26 10:05:26 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+
3 rows in set (0.00 sec)
-----REPLICA
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
2 rows in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  1 | lt1       |                | 2023-11-26 10:05:26 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+
3 rows in set (0.00 sec)

We can see a new row inserted into the test_tab_log table. The reason is that the trigger fired on the primary and replicated to the REPLICA when we ran pt-table-sync.

7. If we do not want that to happen (new row inserted  in test_tab_log table)

Option 1: Do the pt-table-checksum/pt-table-sync for the test_tab_log  table again. This might fix the issue.

Option 2: We might need to do some work on the trigger like below (or there might be another better way).

Let‘s recreate the trigger as below; the trigger will check if it’s run by ‘larry’.

Drop trigger  after_test_tab_insert;
DELIMITER $$
CREATE  DEFINER=`larry`@`%`  TRIGGER after_test_tab_insert
AFTER INSERT
ON test_tab FOR EACH ROW
BEGIN
   IF left(USER(),5) <> 'larry' and trim(left(USER(),5)) <>'' THEN
     INSERT INTO test_tab_log(id,test_data, op_user,op_time)
     VALUES(new.id, NEW.test_data, USER(),NOW());
   END IF;
END$$
DELIMITER ;

And restore the data to its original out-of-sync state.

The PRIMARY

mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
2 rows in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+

The REPLICA

mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
1 row in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+
2 rows in set (0.00 sec)

Run pt-table-sync under user ‘larry’@’%’.  

root@deb11m8s:~# pt-table-sync --user=larry --ask-pass h=192.168.56.190,P=3306 --sync-to-master --nocheck-triggers --replicate percona.checksums_test_tab --tables=testdb.test_tab --verbose --execute
Enter password for 192.168.56.190: 
# Syncing via replication P=3306,h=192.168.56.190,p=...,u=larry
# DELETE REPLACE INSERT UPDATE ALGORITHM START    END      EXIT DATABASE.TABLE
#      0       1      0      0 Nibble    21:02:26 21:02:27 2    testdb.test_tab

We can use pt-table-sync, which will fix the data drift for us, and the trigger will not fire when pt-table-sync is run under user larry.

root@deb11m8s:~# pt-table-sync --user=larry --ask-pass h=192.168.56.190,P=3306 --sync-to-master --nocheck-triggers --replicate percona.checksums_test_tab --tables=testdb.test_tab --verbose --execute
Enter password for 192.168.56.190: 
# Syncing via replication P=3306,h=192.168.56.190,p=...,u=larry
# DELETE REPLACE INSERT UPDATE ALGORITHM START    END      EXIT DATABASE.TABLE
#      0       1      0      0 Nibble    21:02:26 21:02:27 2    testdb.test_tab

—The PRIMARY
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
2 rows in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+
2 rows in set (0.00 sec)

—The REPLICA
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
+----+-----------+---------------------+
1 row in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
+----+-----------+----------------+---------------------+
2 rows in set (0.00 sec)

8. If we still insert other data into the table test_tab under another user (e.g. root@localhost), the trigger will still fire

mysql> select user();
+----------------+
| user()         |
+----------------+
| root@localhost |
+----------------+
1 row in set (0.00 sec)
mysql> insert into test_tab (id,test_data,op_time) values(3,'lt3',now());
Query OK, 1 row affected (0.01 sec)
-— The PRIMARY 
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
|  3 | lt3       | 2023-11-26 21:04:26 |
+----+-----------+---------------------+
3 rows in set (0.00 sec)
+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
|  3 | lt3       | root@localhost | 2023-11-26 21:04:26 |
+----+-----------+----------------+---------------------+
3 rows in set (0.00 sec)

— The REPLICA 
mysql> select * from test_tab; select * from test_tab_log;
+----+-----------+---------------------+
| id | test_data | op_time             |
+----+-----------+---------------------+
|  1 | lt1       | 2023-11-26 09:59:19 |
|  2 | lt2       | 2023-11-26 10:01:30 |
|  3 | lt3       | 2023-11-26 21:04:26 |
+----+-----------+---------------------+
3 rows in set (0.00 sec)

+----+-----------+----------------+---------------------+
| id | test_data | op_user        | op_time             |
+----+-----------+----------------+---------------------+
|  1 | lt1       | root@localhost | 2023-11-26 09:59:19 |
|  2 | lt2       | root@localhost | 2023-11-26 10:01:30 |
|  3 | lt3       | root@localhost | 2023-11-26 21:04:26 |
+----+-----------+----------------+---------------------+
3 rows in set (0.00 sec)

In our test case, we just cover one AFTER INSERT trigger. In a live production system, there might be more complex scenarios (e.g. a lot of different types of triggers defined on the table you are going to do pt-table-sync, auto-increment value, the table has foreign key constraints, etc.). It would be better to test on a test environment before you go to production and make sure you have a valid backup before making a system change.

I hope this will give you some ideas on pt-table-sync on a table with triggers.

Percona Distribution for MySQL is the most complete, stable, scalable, and secure open source MySQL solution available, delivering enterprise-grade database environments for your most critical business applications… and it’s free to use!

Try Percona Distribution for MySQL today!

Percona Database Performance Blog

https://laracoding.com/wp-content/uploads/2023/12/laravel-eloquent-encryption-cast-type.png

Using the Eloquent ‘encrypted’ cast type, you can instruct Laravel to encrypt specific attributes before storing them in the database. Later, when accessed through Eloquent, the data is automatically decrypted for your application to use.

Encrypting fields in a database enhances security by scrambling sensitive data. This measure shields information like emails, addresses, and phone numbers, preventing unauthorized access and maintaining confidentiality even if data is exposed.

In this guide you’ll learn to use Eloquent’s built-in ‘encrypted’ cast to encrypt sensitive data within an ‘Employee’ model to ensure personal data is stored securely.

Important note: Encryption and decryption in Laravel are tied to the APP_KEY found in the .env file. This key is generated during the installation and should remain unchanged. Avoid running ‘php artisan key:generate‘ on your production server. Generating a new APP_KEY will render any encrypted data irretrievable.

While keeping that in mind. Let’s get started and apply encryption!

Step 1: Create a Laravel Project

Begin by creating a new Laravel project if you haven’t done so already. Open your terminal and run:

composer create-project laravel/laravel model-encrypt
cd model-encrypt

Step 2: Add Database Credentials to the .env file

Open the .env file in your project and add the database credentials you wish to use:

.env

DB_CONNECTION=mysql
DB_HOST=127.0.0.1
DB_PORT=3306
DB_DATABASE=your-db
DB_USERNAME=your-db-user
DB_PASSWORD=your-db-password

Step 3: Create a Model and Migration

Begin by generating an Employee model and its corresponding migration using Artisan commands:

php artisan make:model Employee -m

Step 4: Add Migration Code

Open the generated migration file and add the code below to define the table and its columns, including those that we will apply encryption to later on.

As stated in the documentation on Eloquent encrypted casting all columns that will be encrypted need to be of type ‘text’ or larger, so make sure you use the correct type in your migration!

database/migrations/2023_12_10_172859_create_employees_table.php

<?php

use Illuminate\Database\Migrations\Migration;
use Illuminate\Database\Schema\Blueprint;
use Illuminate\Support\Facades\Schema;

return new class extends Migration
{
 public function up(): void
 {
 Schema::create('employees', function (Blueprint $table) {
 $table->id();
 $table->string('name'); // The 'name' column which we won't encrypt
 $table->text('email'); // The 'email' column, which we will encrypt
 $table->text('phone'); // The 'phone' column, which we will encrypt
 $table->text('address'); // The 'address' column, which we will encrypt
 // Other columns...
 $table->timestamps();
 });
 }

 public function down(): void
 {
 Schema::dropIfExists('employees');
 }
};

Step 5: Run the Migration

Run the migration to create the ‘employees’ table:

Step 6: Add encrypted casts to Model

Open the Employee model and add the code below to specify the attributes to be encrypted using the $casts property. We’ll also define a fillable array to make sure fields will support mass assignment to ease creation of models with data:

app/Models/Employee.php

<?php

namespace App\Models;

use Illuminate\Database\Eloquent\Model;

class Employee extends Model
{
 protected $casts = [
 'email' => 'encrypted',
 'phone' => 'encrypted',
 'address' => 'encrypted',
 // Other sensitive attributes...
 ];

 protected $fillable = [
 'name',
 'email',
 'phone',
 'address',
 ];

 // Other model configurations...
}

Step 7: Saving and Retrieving Encrypted Data

Once configured, saving and retrieving data from the encrypted attributes remains unchanged. Eloquent will automatically handle the encryption and decryption processes.

To test this out I like to use Laravel tinker. To follow along open Tinker by running:

Then paste the following code:

use App\Models\Employee;

$employee = Employee::create([
 'name' => 'Paul Atreides',
 'email' => 'paul@arrakis.com', // Data is encrypted before storing
 'phone' => '123-456-7890', // Encrypted before storing
 'address' => 'The Keep 12', // Encrypted before storing
]);

echo $employee->email; // Automatically decrypted
echo $employee->phone; // Automatically decrypted
echo $employee->address; // Automatically decrypted

The output shows the Laravel Eloquent was able to decrypt the contents properly:

> echo $employee->email;
paul@arrakis.com⏎
> echo $employee->phone;
123-456-7890⏎
> echo $employee->address;
The Keep 12⏎

If we view the contents in our database we can verify that the sensitive data in email, phone and address is in fact encrypted:

Conclusion

By using Laravel Eloquent’s built-in cast “encrypted” we can easily add a layer of security that applies encryption to sensitive data.

In our example we learned how to encrypt sensitive data of employee’s like email, address and phone and demonstrated how the application can still use them.

Now you can apply this technique to your own applications and ensure the privacy of your users is up to todays standards. Happy coding!

References

Laravel News Links