My e-mail box is full but I can’t access my archived mail and only have less than 800 e-mails, what gives?

Several months ago in an effort to save space I archived about 4000 messages from over the last 5 years thinking it would help. Now with only 800 in my inbox I’m getting a message saying my inbox is full. My Google Drive is showing 1GB while my gmail is showing about 14GB. I’m assuming it’s my archived mail causing the problem but nothing shows up when I go to Archived Mail or All Mail. I’m worried now that all those e-mails are gone forever and I’m not sure what could be causing the backlog, please help!

Does a transaction always get verified/processed by ALL full nodes in the network before being added into the block by some mining node(s)?

If not, then why is it often claimed that the blockchain scaling issue is due to the fact that all tx has to be verified by all full nodes?

As long as a mining node found the correct nonce, the tx will be added to the block and the blockchain advances. It does not matter if the tx is not yet verified by some full nodes in the network.

(Yes, it will most probably get verified by all full nodes after it got into the block and the block propagates throughout the network. But this is not relevant as the blockchain already advanced (assuming there is no fork).)

Exagrid with a 9 TB drive. It showing as full, but I’m only using 3 TB

0 down vote favorite I have an Exagrid with a 9 TB drive. It showing as full, but I’m only using 3 TB.

[root@RFExaGrid1 shares]# lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT sda 8:0 0 9.1T 0 disk ├─sda1 8:1 0 46.6G 0 part / ├─sda2 8:2 0 11.2G 0 part [SWAP] └─sda3 8:3 0 9T 0 part /home1 sdb 8:16 0 7.5G 0 disk └─sdb1 8:17 0 7.5G 0 part [root@RFExaGrid1 shares]#

Have transferred all music to SD card but still seem to have full system/ user data

My galaxy S7 external SD card seemed to have stopped working. Which caused my phone to show full internal storage. I replaced the SD card with a new and larger one and transferred all my google play music to it. My phone is still showing full storage. 30.5 GB of storage is consumed by userdata/system. I removed the external sd card and checked if the google play music was showing any data still. No data is showing. So I can not for the life of me figure out what data and where is consuming all my storage.

Next step is clear the cache.

Last step is master/hard reset. Really don’t feel like that though.

Ideas?

Data structure for fixed sized records that only need to support full scans

Background

I’m building a memory-mapped inverted index for document search. The inverted index stores (docId, tokenId) pairs, and the whole memory-mapped file is divided into blocks.

A block has a following structure:

  1. Every block can hold 64 inverse index records of size 8 bytes each. The first 4 bytes represent the tokenId, and the next 4 bytes the documentId. (I know the tokenId thing is redundant, but I’ll change that later)
  2. Every block has a 8 byte header. The first 4 bytes represent how many records out of 64 are actually present in that block. The next 4 bytes are a pointer to the next block that houses documentIds for the same tokenId.
  3. Thus, every block is of size 4 + 4 + 8 * 64 bytes.

Purpose

Another global in-memory index points to the first block for a given tokenId. Then, I start traversing the pointers in each block till I’ve read all the documentIds for that particular tokenId.

I’ll never need to get some documentIds for a given tokenId. I’ll always need to fetch all documentIds given a tokenId.

Why this structure?

I tried bbolt (B+Tree implementation) and Badger (LSM implementation) in Golang. Both make the following assumptions that are not true in my case:

  1. Above libraries support keys of varying length. Mine is always of fixed length.
  2. Above libraries support random access to a specific key. I’m only interested in a very specific type of range scan.
  3. I only need to store keys, no values, so I can do away with a lot of implementation complexity around that.

I’ve gained some performance by not having to make those assumptions.

Question

My current implementation in Golang arranges this inverted index in a block-like structure. However, the one thing it does not do is sort them on write. I’m looking for suggestions on data structures that can do it, without resorting to something more heavyweight like the B+Tree. We can assume that I’d be using an SSD.

I’m able to read about 500,000 inverse records this way (on an SSD) in less than 5ms, but sorting those amount of records take considerably longer. So it’s essential that I have them sorted beforehand so that I can “merge” documentIds from boolean queries.

After the final scoring (which will require document metadata to be now stored in a proper k-v store like badger or rocksdb), I’m planning on using a priority queue (like Lucene) to get the top “n” documents.

Where to get the full offline-capable installer for macOS Mojave (.1 update)?

When I originally downloaded the Install macOS Mojave.app for the original release 14.0.22, I got a 6.03 gig file.

Today I replaced that file by downloading the installer fresh from the App Store for 14.1.0. The app is only 22 megs! So obviously this is just a shell that I imagine downloads the needed content over the internet at runtime.

But I need an offline installer, for clean-installs on various Macs and for creating virtual machines.

➥ How to obtain fully-loaded offline installer app for Mojave 14.1.x?

What about just using the original installer, then updating immediately the newly installed OS to .1? That .1 update was a whopper, over 3 gigs! I really want to avoid repeating that updater download. I really need a fully-loaded offline installer.

enter image description here