Autofocus – remote flash unit for focusing in an off-camera flash scenario

So you could press a button to turn on the light, focus, press another button to turn off the light, and then take the picture?

When using a studio flash, you usually have a modeling lamp. You don't have to turn the modeling light off to take pictures. They are bright enough so that you can get an idea of ​​the lighting and the focus. However, the flash will cover the modeling lamp so much that your correct exposure, if the flash does not fire, will not record anything. These lamps are not that bright.

Copy the idea if you're using a flash. Place a dimmable lamp on a floor lamp and give yourself just enough light to focus. Don't worry about turning it off, it will be overwhelmed.

forms – scenario for validation in two transmission times (the second validates and shows a previous hidden field)

I use alter_form to & # 39; user_login_form & # 39; to change. The problem is that I am hiding a field and I want to show it to the user and validate it ONLY AFTER login and passport are validated (The user previously clicked the "Login" button.)

How can I tell the form that I am not done yet and have to fill in a new field that can now be viewed and validated?

I'm not sure which approach is the best? Did someone have the same situation before

Architecture – application design for a given scenario

Given the following scenario, how should you develop this application to take advantage of modern standards / technologies?

You need to create an application where users can log in to see data visualizations for their favorite sports teams. You will find a reliable API to get the required sports data. Now, however, you have to decide how you want to design your application.

Your application essentially consists of three components: a front end, a back end and a cronjob service that writes the required data from the API provider into your database.

I am only familiar with the client / server architecture, but I think the cronjob service should be isolated in its own repo / container, in which you would only share the DB instances / models with the backend component. However, I am interested in hearing from experienced engineers and whether I can use certain cloud technologies to improve performance and maintainability. I am considering using graphql and MongoDB for the backend component if this is relevant.

Any discussion would be welcome. Thanks a lot! Also, links to relevant reading material would be great … if you were trying to learn something.

updates – A stunning scenario with features and areas

Background: Yesterday I updated a plugin called WP Security Audit Log to the latest version 4.0.0 (from

After the upgrade, the website was completely blown up. After doing some research, I found that the biggest damage was done to another plugin called Polylang, which is supposed to translate everything.

The main problem was that it was completely impossible to call Polylang's functions like pll_current_language with the bug

PHP Fatal error:  Uncaught Error: Call to undefined function pll_current_language() 

After long hours, I decided to restore a backup and everything went back to normal.

At this point you might think:

You could have simply deactivated the WP Security Audit Log plugin and everything should
go back to normal …

This is the overwhelming part (for me). Even if Polylang and 100% of the other plugins remained disabled, the Polylang functions like pll_current_language were still not accessible after the WP Security Audit Log was updated

At first, I preferred not to restore the backup because things would be lost, but I couldn't easily understand how this could be done.

How can a plugin update completely break the "callability" of other plugins functions even if this plugin is deactivated ???

There are only two sensible options:

  1. It has introduced some configuration file, .htaccess,
    wp-config.php or wherever there are some ways to avoid this that I don't have
    able to track down.
  2. Some mods have been added to the database, which is unlikely since the database never technically influences the functional behavior.

I am writing this because I lost more than 5 hours yesterday trying to track the problem to no avail. And when I am 100% sincere, I found that the problem with updating the WP Security Audit log was caused by sheer accident of queuing plugins one by one to check their impact on the site.

Any insights into what exactly could have led to this or how I can only track it down for learning purposes?

Security – Is this scenario an exception to the rule that passwords are never saved in plain text?

I am creating a full stack web application for a professor. At his request, the passwords and user names are generated programmatically and cannot be changed or reset by the students. (If you forget your password, ask the professor who can look it up.) Does this tightly controlled system eliminate the need to follow all common best practices for storing passwords in a database?

If relevant, the app contains no associations or identifiers between the student's identifying information (name, gender, etc.) and his username and password.

sql – scenario builder database design

I'm working on moving a backup system of loose files to a (spatial SQLite) database.

The user places objects from a library in a 3D simulation. Scenarios are "what-if", what if we put 5 wind turbines here, what if we put 10 solar panel racks instead, etc. They are mutually exclusive within a "group".

Elements are prototype objects that can be instantiated by the user. Wind turbines and solar panel racks are specializations of items with additional data.

It is the first time that I have performed SQL outside of trivial school projects. So please do not hold back to name best practices / tips.

Do my FKs make sense?

Is it in the right shape?

Thank you in advance.

Interactive version:
You can export SQL from there if this is easier to read.

Pseudo-SQL below.

//a mutually exclusive group of scenarios
Table group
   group_id int (pk, increment)
   name text (unique, not null)
   active_scenario_id int (ref: - scenario.scenario_id)
   is_active bool

//a user created arrangment of item instances
Table scenario 
  scenario_id int (pk, increment)
  group_id int (ref: > group.group_id)
  name text (not null)

//an item is a prototype. an object that can be repeated many times.
//the name identifies a prefab in the application
Table item
  item_id int (pk, increment)
  name text (unique, not null)

//an instance of an item prototype in the world
Table instance
  instance_id int (pk, increment)
  scenario_id int (ref: < scenario.scenario_id)
  item_id int (ref: - item.item_id)
  position PointZ (not null)
  rotationX float (default: 0)
  rotationY float (default: 0)
  rotationZ float (default: 0)
  scaleX float (default: 1)
  scaleY float (default: 1)
  scaleZ float (default: 1)
  active_date datetime (null)
  inactive_date datetime (null)

//a LineString created by the user.
Table user_path
  instance_id int (pk, increment, ref: - instance.instance_id)
  win3d_style_id int (ref: < win3d_style.win3d_style_id)
  path LineStringZ
  viewPath LineStringZ
  float duration (note: 'duration if this path is an animation')

//a polygon created by the user
Table user_polygon
  instance_id int (pk, increment, ref: - instance.instance_id)
  name text (not null)
  geometry PolygonZ
  win3d_sytle_id int (ref: < win3d_style.win3d_style_id)

//a wind turbine is a kind of item with extra data
Table wind_turbine
  item_id int (pk, increment, ref: - item.item_id)
  rotor_diameter float
  shaft_height float
  shaft_base_diameter float
  shaft_neck_diameter float
  min_wind_speed float (note: 'wind speed in m/s at which the blades starts spinning')
  max_wind_speed float (note: 'wind speed in m/s at which the blades reaches peak rotations per minute')
  off_speed float (note: 'wind speed in m/s at which the blades is stopped for safety')
  min_rotations_per_minute float (note: 'rpm at which the blades spin when the wind speed equals min_wind_speed')
  max_rotations_per_minute float (note: 'rpm at which the blades spins when the wind speed equals max_wind_speed')
  watt_min long (note: 'Watt produced at the min_wind_speed')
  watt_max long (note: 'Watt produced at the max_wind_speed')

Enum solar_panel_rack_kind

Enum solar_panel_facing
//a solar panel rack is a kind of item with extra data.
//a solar panel rack is an arrangement of solar panels
Table solar_panel_rack
  item_id int (pk, increment, ref: - item.item_id)
  solar_panel_id int (ref: - solar_panel.solar_panel_id)
  num_columns int
  num_rows int
  column_gap int (note: 'centimeters between each column')
  row_gap int (note: 'centimeters between each row')
  kind solar_panel_rack_kind
  facing solar_panel_facing

Enum solar_panel_kind{

Table solar_panel
  solar_panel_id int (pk, increment)
  width int (note: 'centimeters')
  height int (note: 'centimeters')
  watt_peak long
  kind solar_panel_kind

Table win3d_style
  win3d_style_id int (pk, increment)
  primary_color uint32 (default: 4294967295, note: 'default is opaque white')
  secondary_color uint32 (default: 4294967295, note: 'default is opaque white')

Table point_of_view
  point_of_view_id int (pk, increment)
  scenario_id int (ref: < scenario.scenario_id)
  name text
  position PointZ (not null)
  rotationX float (default: 0)
  rotationY float (default: 0)
  rotationZ float (default: 0)
  field_of_view float (default: 30)
  thumbnail blob (note:'jpg encoded image')

Image of the scheme

SQL Server – What preventive measures can be taken to deal with this scenario with minimal data loss

I am SQL Server DBA and encountered strange scenario, our cluster had 10 nodes 5 primary 5 secondary each node had SQL role …. Example Ams1pd11 to Ams1pd15 are primary .. and Ams3 side of PD11 to 15 was secondary … In this scenario, the behavior of the entire cluster after 2 node failure was unusual and all node availability groups were inaccessible, resulting in multiple customer failure.

Explanation of the real-time scenario, how it started when I was on the shift ….

Ams1pd12 has failed and hosted primary server A.
So role A on the best possible node was automatically canceled and he chose Ams1pd11.

Ams1pd11 already had a role, e.g. B. B
Now Ams1pd11 hosted both A and B when Ams1pd12 failed

Since both primary roles were on a node, there was a risk of balance. I failed over the B node to one of the secondary nodes on the Ams3 side. Now it was balanced and I was just about to investigate on Ams1pd12 why it failed and everything. ,

But suddenly the Ams1pd11 node also failed and the role did not fail and was stuck there …

Now 2 out of 10 nodes had failed and one role was stuck, so the customers in this role were affected.

We troubleshooted the same issue at Microsoft and found that other nodes were shown by the availability group for all blocked nodes and were not opening and blocking when expanding.

This affected all nodes in this cluster and therefore our backups were stopped.
There was data loss …

The stuck role and the customer on this node only had to accept a data loss of 15 minutes because the service for us and for them also failed at the same time

The nodes that appeared and AG groups were inaccessible. It was strange that the users who were already logged in could change the data. Only new connections were rejected. But the old connection was still active.

So if the problem started and the backup was stopped at 7am, most customers could access the database until 6pm, resulting in 11 hours of data loss.

Manual database recovery took a lot of effort …
These online nodes were easy to restore because we only had to attach the database databases while migrating data and log files for databases with deadlocked roles. We had to restore it manually.

Please suggest the best strategies to follow in this type of disaster to achieve rapid recovery.

Observer pattern – Java Observable API – when is it a pull scenario?

The push and pull scenarios differ in how the registered observers experience what has actually changed when they are called by the subject / observable.

In the pull scenario, the call to the observer contains no information other than that which the subject / observable changed. The observer must then pull the information from the subject / observable object to determine What Change has happened so that it can know how to react.

In the push scenario the call to the Observer does Contain information about what has changed (possibly with more details than some observers need). The information is passed on to the observers, regardless of whether they want it or not.

Both scenarios have their place, and you can also have an incident in which some information is pushed and observers can get more information when needed. Which scenario to use depends on the typical information needs of the observers in order to react to changes in the subjects / observables.

ids – Scenario categorization with deep packet inspection – intrusion detection

I am researching intrusion detection systems (IDS) and deep packet inspections (DPI). For example, suppose a system in which values ​​are passed to a validation system and the validation system validates the data passed (verifies anomalies, such as statistics, machine learning, etc.).

  • Is the validation process called DPI, even if only the payload is examined?
  • Is the screening process called network-based attack detection or something else?

Air travel – is the connecting flight scenario given in this video possible and if so, how did it occur?

In this video, which was posted in 2011 by the well-known Youtubern Rhett and Link, a story is presented, which I can not agree with the reality. Can you help me a) determine if this scenario could possibly occur and b) how?

The following story is presented as a fact (although it is not Necessary I mean, we should believe it. Although it's not impossible, I think that as a long-time supporter of the broadcaster, it would be unusual for them to really have invented this. You may be confused!)

  • Rhett and Link will receive a route from Delta via email for a flight from RDU to SLC. (They don't tell us if they bought the ticket or if someone bought it for them).
  • Upon arrival at RDU, you will be issued boarding passes at the check-in desk indicating that the trip should be from RDU to SLC without any connection. The total travel time is about 5-5.5 hours. (I am not sure if they have considered time zones).
  • They find their gate and on the sign above the gate is "Salt Lake City". They scan their boarding passes and get on the plane.
  • Link claims to have slept through most of the flight, but Rhett says he stayed awake all the time.
  • About 2 hours after the flight began, Rhett noticed that they were landing at MSP – not SLC.
  • Confused, Rhett and Link ask a flight attendant what's going on, and the flight attendant asks them to ask a gate agent when they land. (Nobody else on the flight is confused).
  • Upon landing, the gate agent prompts you to take your boarding pass to another gate where another aircraft is waiting. (The sign above the gate is "Salt Lake City").
  • They use theirs original Boarding passes to board this new flight at the new gate occupy the same places as in the previous aircraft. (They have different people around).
  • The plane took her (and her checked bags) to SLC without incident.

Here are my previous thoughts: Occasionally, flights with the same flight number but with a different aircraft are continued ("change of lane"). In this case, however, I have found that this is announced frequently and clearly and that this is noted in large letters on the boarding pass. In addition, the sign at the gate in RDU should have been "Minneapolis". As far as I remember, passengers generally get new seat assignments – having the same seats seems really weird … What if it's a different plane or someone else just bought their seats for the second segment before doing so first booked?

Any help would be appreciated!