centos7 – How to force creating an extended partition in before the last primary partition in kickstart?

I am searching for a way to write a kickstart partition section where I can control precisely which partitions are primary, and which partitions are secondary partitions in a extended partition.

I don’t want to use LVM, and I want to create this partition table :

enter image description here

On this screen, the sizes are not important, it was just a try to see if I really can do this kind of partition table.

I want the last partition (sdb4) to be physically at the end of the drive, and for that, I need to put sdb5 and sdb6 in an extended partition.

The problem is that the kickstart file parse the instructions sequentially, and the last partition is the last part instruction. But the problem is that kickstart will declare everything as primary and auto-create an extended partition automatically.

The problem is that I don’t seems to find a way to do what I want.

Is that even possible ?

hard drive – How to set vdb as primary disk

I am not a Linux expert but I trying the best I can, learning by doing :-]

I hire a VPS and it has two disks.
vda = 90 GB SSD
vdb = 200 GB HDD

vda has 4 partition and that is were the operating systems is.
vdb has 1 partition ans is empty

It seems to me that all software I install, such as PLESK control panel and applications via PLESK is installed on vda.
I assume vda is the primary disk

Is it possible to change this so that the HDD is primary disk and the smaller SSD is the secondary disk ??
This might require a full reinstallation of the system I guess ?

Thank you for your comments.

By the way I am using DEBIAN 10


email – Gsuite – Mailed-by and from. How to have a mailed-by as subdomain but the from is the primary domain – in gSuite (Gmail)

I want to have my email “from” to be from my primary email (gSuite), adn the mailed-by be from my subdomain (which I have added to my Gsuite account). Is this possible in GSuite? I know through different email providers, ex. MailerLite, they allow you to domain alignment to a subdomain, so that the from will be your primary domain but the mailed-by will be the subdomain. I have attached screenshot from Russel Brunson that he does this, and Neil Patel. He also does this.
Thanks for your help.
This is my first post.



distributed systems – If primary server’s UTC source fails, how will synchronization be possible in NTP(network time protocol)?

If a primary server’s UTC source fails, then it can become a stratum 2 secondary server. If a secondary server’s normal source of synchronization fails or becomes unreachable, then it may synchronize with another server.

I have made a picture based on what I understood. I don’t understand how will synchronization be possible when primary server(stratum 1 server) fails? Shed some light please here.

Figure 1 shows UTC fails. Figure 2 shows primary server becoming secondary server

❓ASK – Forex trading as primary source of earning | NewProxyLists

Forex trading is a professional business. Many people work here as full time trader. Even there are many companies out there, who trades in forex with their investors fund and gives incentives to them.

But as an Individual, does forex trading really can provide earnings for lifetime? I would say, surely yes. Only thing is that it’s us who will need to develop our skills at par with the market. Once we are settled, we can make hefty money.

MySQL 8 – Primary Keys – Sequential UUID or INT with AI?

My DB has ~100 tables and 3 for which I need UUIDs.

For these 3 tables, the main goal is simply to hide sensitive IDs to visitors.

MySQL 8 has introduced UUID support and I was planning for these 3 tables to use an UUID as a sequential PK doing:

UUID_TO_BIN(UUID(), true) -- Stored as BINARY(16)

My concerns are:

  • No Last Inserted ID available anymore, I would have to select the UUID first
  • No Auto-increment anymore (you can use a trigger instead, adding some complexity)
  • Slight performance loss

For these reasons, I’m thinking about a hybrid solution:

  • Using an INT with AI as the PK
  • Adding a separate BINARY(16) column for the UUID, keeping it sequential and indexed (but not unique as unnecessary and resource consuming on inserts)

Am I missing something here? Hopefully MySQL will follow PostgreSQL and introduce a UUID() data type in the future to avoid these issues.

dnd 3.5e – How can you use Primary Contact for early access to feats or prestige classes?

A marshal dip when you already have Skill Focus (Diplomacy)

The marshal from Miniatures Handbook gainst Skill focus (Diplomacy) as a bonus feat at 1st-level. Unlike every other class feature offering a bonus feat, the marshal’s class feature reads

Skill Focus (Diplomacy): Because a marshal has a way with people, he gains this feat as a bonus feat. If the marshal already has the feat, he can choose a different one.

(emphasis mine)

So you can take Skill Focus (Diplomacy) ahead of time, and then dip marshal when you need Primary Contact and get it as a bonus feat. Note that, since Primary Contact itself requires the Favored feat, you also need to take that feat—so two feats and a class level burned on this trick.

Frostblood orc or half-orc and Endurance as a bonus feat

Kind of coming at this from the other angle, the frostblood orcs and half-orcs from Dragon Magic gain Endurance as a bonus feat, again with some unique wording:

—Endurance: Frostblood orcs and half-orcs have Endurance as a bonus feat. If a frostblood orc or half-orc would later gain Endurance as a bonus feat, he can select any other feat for which he qualifies.

(emphasis mine)

So again, by being a frostblood orc or half-orc, you can gain Primary Contact as a bonus feat instead of Endurance as a bonus feat. The trick is timing things so the bonus Endurance feat lands on the level you want. Notably for prestige classes requiring 8 ranks, a mystic ranger from Dragon vol. 336 gets Endurance at precisely 4th level, perfect for early entry (simply dipping something else to take ranger 3rd at 4th level works for too, of course). The pugilist fighter from Dragon vol. 310 gets it at 1st level—alongside another bonus feat, which could allow for taking a feat a level early, too (though the pugilist feat list is fairly limited).

Compared to the marshal, this approach locks in your racial choice in order to save a feat: compared to a human marshal, you basically gain the benefits of the frostblood orc or half-orc’s other features—which leave a lot to be desired, thanks to the vulnerability to fire. However, the opportunity to use something other than marshal—which is a pretty terrible class—can easily make this a win for the right character.

Half-elf or human paragon

The variant racial paragon classes from Unearthed Arcana for half-elves and humans both include a bonus feat that can be any the paragon qualifies for. The half-elf gets it at 1st level, while the human gets it at 2nd.

Superficially, this approach costs much less than the frostblood (half-)orc, being just a race and one or two levels. The half-elf version also saves a feat compared to dipping marshal, since you don’t need to take Skill Focus (Diplomacy). But the half-elf race is notoriously poor, so much so that a human marshal or, in the right build, a frostblood (half-)orc, could easily be better. Meanwhile, the human race is one of the best in the game, but human paragon 1st is an extremely poor level, with almost zero going for it—dipping fighter to make up for the feat spent on Skill Focus (Diplomacy) is the better use of your resources for most characters.

data structures – What is the primary reference for the observation/discussion of how neural networks struggle with ambiguous training datasets?

It is known that neural networks, such as convolutional neural networks, struggle with pattern recognition if training sets contain ambiguities (i.e. several labels can correspond to one and the same pattern). However, I struggle to locate the paper that directly discusses this issue, or demonstrates it for the first time, etc. If anybody can point me to the paper(s), that would really help. Thanks!

shared hosting – Can add-on domains be transferred to a new host before transfering the primary domain?

I currently have a few websites hosted with a web host, but I want to switch to a better host. I found one that I want to use on a trial basis, transferring one website at a time to it to see how well it works for me before committing fully. Both hosts are shared servers, and I have one primary domain and three add-on domains. All four domains are registered through a third party and I point the name servers to the older host.

My question is, would I likely run into a problem if I were to point the nameserver for the add-on domain before the primary domain? Or vise-versa? For example, if I make an account with the new host and select my existing primary domain as what will be the main domain for the account, but keep the name servers pointed at the old host, would that cause issues? Because, for example, for these servers www.addondomain.com is usually also the same as addondomain.example.com but that latter address wouldn’t actually work without example.com being set up.

Or alternatively, if I transfer the name server for the primary domain to the new host before doing so for the add-on domains, might that cause the add-on domains to stop functioning?

To be clear, the domain registrar is staying the same. The web host is what will be changing.

Potentially this is something that the new web host can answer, but I’m looking for a little grounding so I can ask the right questions.