Why is it considered hard to maintain strong consistency in a distributed system?

Why can’t you just use strongly consistent reads for all your DB reads, with retries on 500 responses? According to CAP theorem increasing consistency should probably lower availability, but can’t the decreased availability (increased 500 responses) be handled fairly easily using retries? (assuming you are fine with a small percentage of queries taking a bit longer due to retries)

Using DynamoDB as an example, but this can be generalized to any noSQL cloud offering – It also seems like DDB with on demand scaling will simply increase your read capacity units (RCU) used if you turn on strong consistency, incurring a higher cost ($) but keeping the same latency on db queries, so it seems like the negative is only higher cost. It seems like you can just keep vertical scaling the DB’s processing power to meet your needs. Is it actually plausible that with a noSQL cloud database with a high traffic level, you cannot just throw enough money at it, and it could hit some scaling limit to make strongly consistent reads slower?

And then with regard to the entire question generalized to distributed systems, what does it actually mean for a distributed system to be ‘strongly consistent.’ I’ve heard this used to describe systems before but I don’t actually know it means beyond ‘all DB interactions being strongly consistent.’


My second question might be more basic level but necessary to understand the cost of providing consistency, but why do consistent read issues actually occur, ie. why does stale data occur (in single queries like read/write, not transactions with multiple reads/writes per transaction)?

From what I understand, with decreasing probability any time after a write occurs, it’s possible for one reader to read correct data, then a second reader to read stale data AFTER the first reader reads correct data (correct me if this isn’t actually true, but my understanding was it is). Why does this happen? Doesn’t a read just involve a read from some location on the disk?

Any cheap place to purchase hard drives

Trying to sell some storage VPS, wondering if there’s any place you guys know can get great deals on hard drives.

Thank you…. | Read the rest of https://www.webhostingtalk.com/showthread.php?t=1834894&goto=newpost

terminal – How do I access files, copy them on an external hard drive, and see them on it?

I’m trying to recover the password of an old iMac with a little trick I’ve learned. To do that, I need to access recovery mode, which doesn’t work for some reason. I know the shortcut is cmd + R. I’ve also tried Shift + cmd + R but that didn’t work either. So I went into single user mode to access the files, which worked. I have an external hard drive that is formatted as Mac OS Extended (Journaled) to put the files I’m looking for on the iMac into the drive. I copied them and I went to my computer to access it. But now my drive isn’t mounting!

When I did the copy I ran the command cp <file> /dev/disk1s2. When I try to mount it on the other computer to see the files, I tried mounting it through the command line with diskutil mountDisk /dev/disk2 and diskutil mount /dev/disk2s2. Both commands gave me an error saying that the disk failed to mount. I also tried mounting it through Disk Utility but nothing happens.

I need those files because it contains the password of the computer. How am I going to see the files? I’ve also tried going into another user account and copying the files but that user needs sudo to access them and that user isn’t part of the /etc/sudoers file. In Single User mode, I tried to add the other user to the file but everything is read-only even though I ran /sbin/mount -uw. I think if someone told me how to fix the read-only issue in Single User mode, I would be able to create a new admin account by running rm /var/db/.AppleSetupDone The user account I’m trying to recover is the admin account of the computer so that’s why I think that would work.

windows 7 – Hard drives seemingly no longer has bad sectors on it after erase and reformat

I have a Seagate SSHD laptop hard drive. When I use Seagates Software SeaTools. The drive used to fail the short drive self-tests and the long generic test. When it failed tests sea tools would warn me about having bad sectors. It would however to pass the short generic tests. I could access the data on the drive without any problems. I ran these tests from two different computers both from inside Windows 7 and Windows 10 and using sea tool for dos. When I try to run fix-all fast and fix-all long on the drive to try and fix the bad sectors it fails. All of a sudden after I deleted all the partitions on the drive and reformatted the drive after backing up the data on the drive. I just did a quick format too. Now the drive somehow can pass any tests I throw at it. It seemingly has no bad sectors. I tested the drive with different software other than seatools to get a second opinion and it passed those tests too. In addition, I used to not be able read the drive at all and it would cause windows disk management, file explorer and sometimes it would cause the bios to hang on my computer. What should I do now? Should I still keep using it because it seems to be fine now or should I RMA it and get a repaired/refurbished drive since it is still under warranty?

Is Really Ruby A Language Hard To Debug?

As the title suggests, is Ruby really h… | Read the rest of https://www.webhostingtalk.com/showthread.php?t=1834051&goto=newpost

hard drive – Do i need power cable to turn internal HDD on (I use laptop) because i need to take some data from my client Internal HDD

I have a client that needs his data transfered to his cloud. Now i need to buy sata+pata cable (He has old 2 X 80gb barracuda( 1 HDD with PATA, and the other one uses SATA that im not sure gonna work / have a lot bad sector) but do i need to buy power cable aswell? If so could you give me a link to where i can buy the power cable? (I need Usb power cable if they exist)

hard drive – Unable to format internal disk: disk is read-only

I’m having a strange issue with my MacBook Pro (13″ early 2015 running macOS Catalina 10.15.7): few days ago I was using my computer when it has suddenly shut down, when i powered it back on and provided my password it has shut down again and it keeps doing this every time I try to log into the system, plus, it seems that the safe mode is unavailable.
I’d like to format the internal SSD drive (currently formatted as APFS encrypted with FileVault enabled) and proceed with a clean install of the OS but as I can see the internal disk is in read-only mode and when I try to format it from recovery I get permission denied.
I also tried to boot from a bootable USB drive having Big Sur on it but I’m unable to format the drive nor install the OS. I tried running the built-in diagnostic routine but no disk issue was found.
I finally tried to remove the SSD drive and put it on a Thunderbolt case in order to format it from another Mac but I get a similar error even if the “Ignore ownership on this volume” option is ticked.
I managed to copy some of my data from the internal drive to an external one using the terminal in recovery mode, now I would like to wipe out all the contents from my internal drive and install a fresh copy of macOS (Catalina or Big Sur), how can I format a SSD drive in this condition?

I described this issue on Reddit too and I added a link to another Reddit post where I listed some other software related issue I had and I though they were related to the disk or the filesystem, maybe this may be helpful. https://www.reddit.com/r/MacOS/comments/kpafn0/unable_to_format_internal_disk_disk_is_readonly/

hard drive – Question regarding restoring from Time-machine backup

I’m having a problem with an slightly older iMac owned by a friend. It has behaving in a problematic fashion for some time ending with it refusing to boot properly. He has a complete time-machine-backup from november. So here is the problem: When we boot up in safe-mode with cmd-R the machine starts fine and we can select the time-machine-backup from the attached external drive as the source for the operation. BUT when we try to select the target drive, which is the internal harddrive, we cannot choose it and proceed. The machine sees the internal harddrive, but we can somehow not restore to it. There seems to be one big 1TB partition on the internal disk, but we can’t choose it as destination. See the attached picture.
enter image description here
I’ve tried running fsck -fy from single-user-mode (cmd-S) on the internal disk and this is possible, but it doesn’t solve the problem. I’m a mostly Linux-oriented sysadmin and doesn’t have so much experience with The Mac. But I have done this operation before with succes on other machines. Any suggestions? Should we try to reformat the system disk, install the OS againfrom the internet and see if we can then do the time-machine-trick? Or maybe change the system-disk? New year greetings from Denmark

Como desfaço um git reset HEAD — hard

git reset HEAD –hard
HEAD is now at fe4673e Migrate
reinaldo@sakaar:~/projetos/fusion2$ arken ticket:create 67315
commit suas pendencias, por favor

tentei abrir uma nova branch mas apareceu isto)

On branch ticket_67307
Your branch is up-to-date with ‘origin/ticket_67307’.
Untracked files:
(use “git add …” to include in what will be committed)

    app/controllers/Rma/
    app/views/rma/
    tests/models/Rma/Fornecedor/

nothing added to commit but untracked files present (use “git add” to track)

pergunta é será que perdi tudo?

copy files from Windows hard drive

My original drive was a 2.5″ SATA III HDD. A few years ago I replaced it with a SSD. I accidentally deleted a lot of files recently.

I still have the old HDD and it will have some of my old photos. I see there are cable adapters to attach my old drive to my PC via USB.

If I get one of these cables and attach it will I see a “normal” drive with files or do I need to boot from that drive?